ICCV	Shape and motion without depth.	Carlo Tomasi,Takeo Kanade	1990	Shape and motion without depth.
ICCV	Hypothesizing and testing geometric attributes of image data.	Kenichi Kanatani	1990	Hypothesizing and testing geometric attributes of image data.
ICCV	Determining back-facing curved model surfaces by analysis at the boundary.	Robert B. Fisher	1990	Determining back-facing curved model surfaces by analysis at the boundary.
ICCV	Computing optical flow from an overconstrained system of linear algebraic equations.	Marco Campani,Alessandro Verri	1990	Computing optical flow from an overconstrained system of linear algebraic equations.
ICCV	Vanishing point calculation as a statistical inference on the unit sphere.	Robert T. Collins,Richard S. Weiss	1990	Vanishing point calculation as a statistical inference on the unit sphere.
ICCV	A finite element method applied to new active contour models and 3D reconstruction from cross sections.	Laurent D. Cohen,Isaac Cohen	1990	A finite element method applied to new active contour models and 3D reconstruction from cross sections.
ICCV	Interesting patterns for model-based machine vision.	Jorja G. Henikoff,Linda G. Shapiro	1990	Interesting patterns for model-based machine vision.
ICCV	Computing two motions from three frames.	James R. Bergen,Peter J. Burt,Rajesh Hingorani,Shrnuel Peleg	1990	Computing two motions from three frames.
ICCV	Object recognition using a feature search strategy generated from a 3D model.	Yoshinori Kuno,Yasukazu Okamoto,Satoshi Okada	1990	Object recognition using a feature search strategy generated from a 3D model.
ICCV	A fast algorithm for active contours.	Donna J. Williams,Mubarak Shah	1990	A fast algorithm for active contours.
ICCV	Matching range images of human faces.	John C. Lee,Evangelos E. Milios	1990	Matching range images of human faces.
ICCV	Accurate corner detection: an analytical study.	Rachid Deriche,Gérard Giraudon	1990	Accurate corner detection: an analytical study.
ICCV	Appearance-model-based representation and matching of 3-D objects.	Jacques G. Verly,Richard L. Delanoy	1990	Appearance-model-based representation and matching of 3-D objects.
ICCV	Toward the automatic generation of mathematical morphology procedures using predicate logic.	Hyonam Joo,Robert M. Haralick,Linda G. Shapiro	1990	Toward the automatic generation of mathematical morphology procedures using predicate logic.
ICCV	Multiple light source optical flow.	Robert J. Woodham	1990	Multiple light source optical flow.
ICCV	Description and reconstruction from image trajectories of rotational motion.	Harpreet S. Sawhney,John Oliensis,Allen R. Hanson	1990	Description and reconstruction from image trajectories of rotational motion.
ICCV	Towards a computational theory of model based vision and perception.	Haim Shvaytser	1990	Towards a computational theory of model based vision and perception.
ICCV	Dynamic integration of height maps into a 3-D world representation from range image sequences.	Minoru Asada,Masahiro Kimura,Yoshiaki Shirai	1990	Dynamic integration of height maps into a 3-D world representation from range image sequences.
ICCV	Pose determination from line-to-plane correspondences: existence condition and closed-form solutions.	Homer H. Chen	1990	Pose determination from line-to-plane correspondences: existence condition and closed-form solutions.
ICCV	Representation and the dimensions of shape deformation.	E. Saud	1990	Representation and the dimensions of shape deformation.
ICCV	A locally adaptive window for signal matching.	Masatoshi Okutomi,Takeo Kanade	1990	A locally adaptive window for signal matching.
ICCV	Region-based reconstruction of an indoor scene using an integration of active and passive sensing techniques.	Junichi Hoshino,Tetsuya Uemura,Isao Masuda	1990	Region-based reconstruction of an indoor scene using an integration of active and passive sensing techniques.
ICCV	Direct computation of qualitative 3D shape and motion invariants.	Daphna Weinshall	1990	Direct computation of qualitative 3D shape and motion invariants.
ICCV	Segmentation by minimal description.	Trevor Darrell,Stan Sclaroff,Alex Pentland	1990	Segmentation by minimal description.
ICCV	Analysis of facial images using physical and anatomical models.	Demetri Terzopoulos,Keith Waters	1990	Analysis of facial images using physical and anatomical models.
ICCV	On the sensitivity of geometric hashing.	W. Eric L. Grimson,Daniel P. Huttenlocher	1990	On the sensitivity of geometric hashing.
ICCV	Computing spatiotemporal surface flow.	Mark C. Allmen,Charles R. Dyer	1990	Computing spatiotemporal surface flow.
ICCV	Parallel structure recognition with uncertainty: coupled segmentation and matching.	Paul R. Cooper	1990	Parallel structure recognition with uncertainty: coupled segmentation and matching.
ICCV	A dynamic depth extraction method.	Terunori Mori,Masanobu Yamamoto	1990	A dynamic depth extraction method.
ICCV	Photometric motion.	Alex Pentland	1990	Photometric motion.
ICCV	Direct estimation of deformable motion parameters from range image sequence.	Masanobu Yamamoto,Pierre Boulanger,J.-Angelo Beraldin,Marc Rioux,Jacques Domey	1990	Direct estimation of deformable motion parameters from range image sequence.
ICCV	Temporally integrated surface reconstruction.	Joachim Heel	1990	Temporally integrated surface reconstruction.
ICCV	Viewpoint invariant recovery of visual surfaces from sparse data.	Robert L. Stevenson,Edward J. Delp	1990	Viewpoint invariant recovery of visual surfaces from sparse data.
ICCV	Uniqueness, the minimum norm constraint, and analog networks for optical flow along contours.	Atul K. Chhabra,Timothy A. Grogan	1990	Uniqueness, the minimum norm constraint, and analog networks for optical flow along contours.
ICCV	From uncertainty to visual exploration.	Peter Whaite,Frank P. Ferrie	1990	From uncertainty to visual exploration.
ICCV	Dynamic edge warping: experiments in disparity estimation under weak constraints.	Kim L. Boyer,Daniel M. Wuescher,Sudeep Sarkar	1990	Dynamic edge warping: experiments in disparity estimation under weak constraints.
ICCV	Image interpretation using multi-relational grammars.	Staffan Truvé	1990	Image interpretation using multi-relational grammars.
ICCV	The integration of region and edge-based segmentation.	Chen-Chau Chu,Jake K. Aggarwal	1990	The integration of region and edge-based segmentation.
ICCV	Surface shape reconstruction of an undulating transparent object.	Hiroshi Murase	1990	Surface shape reconstruction of an undulating transparent object.
ICCV	Feed-forward recovery of motion and structure from a sequence of 2D-lines matches.	Thierry Viéville,Olivier D. Faugeras	1990	Feed-forward recovery of motion and structure from a sequence of 2D-lines matches.
ICCV	Tracking and grouping 3D line segments.	Zhengyou Zhang,Olivier D. Faugeras	1990	Tracking and grouping 3D line segments.
ICCV	Detecting height from constrained motion.	Xueyin Lin,Zhigang Zhu	1990	Detecting height from constrained motion.
ICCV	Dynamic stereo with self-calibration.	Arun P. Tirumalai,Brian G. Schunck,Ramesh C. Jain	1990	Dynamic stereo with self-calibration.
ICCV	Integrated treatment of matching and measurement errors for robust model-based motion tracking.	David G. Lowe	1990	Integrated treatment of matching and measurement errors for robust model-based motion tracking.
ICCV	Third International Conference on Computer Vision, ICCV 1990. Osaka, Japan, 4-7 December, 1990, Proceedings		1990	Third International Conference on Computer Vision, ICCV 1990. Osaka, Japan, 4-7 December, 1990, Proceedings
ICCV	BONSAI: 3D object recognition using constrained search.	Patrick J. Flynn,Anil K. Jain	1990	BONSAI: 3D object recognition using constrained search.
ICCV	Simple method for computing 3D motion and depth.	David J. Heeger,Allan D. Jepson	1990	Simple method for computing 3D motion and depth.
ICCV	Qualitative route scene description using autonomous landmark detection.	Jiang Yu Zheng,Matthew Barth,Saburo Tsuji	1990	Qualitative route scene description using autonomous landmark detection.
ICCV	A photometric invariant and shape constraints at parabolic points.	Lawrence B. Wolff	1990	A photometric invariant and shape constraints at parabolic points.
ICCV	Geometrical learning from multiple stereo views through monocular based feature grouping.	Eric Thirion,Long Quan	1990	Geometrical learning from multiple stereo views through monocular based feature grouping.
ICCV	Scale detection and region extraction from a scale-space primal sketch.	Tony Lindeberg,Jan-Olof Eklundh	1990	Scale detection and region extraction from a scale-space primal sketch.
ICCV	Hypothesis testing: a framework for analysing and optimising Hough transform performance.	John Princen,John Illingworth,Josef Kittler	1990	Hypothesis testing: a framework for analysing and optimising Hough transform performance.
ICCV	Invariance-a new framework for vision.	David A. Forsyth,Joseph L. Mundy,Andrew Zisserman,Christopher M. Brown	1990	Invariance-a new framework for vision.
ICCV	Active surface reconstruction by integrating focus, vergence, stereo, and camera calibration.	A. Lynn Abbott,Narendra Ahuja	1990	Active surface reconstruction by integrating focus, vergence, stereo, and camera calibration.
ICCV	An estimation-theoretic framework for image-flow computation.	Ajit Singh	1990	An estimation-theoretic framework for image-flow computation.
ICCV	Active 3D object models.	Ruud M. Bolle,Andrea Califano,Rick Kjeldsen,Rakesh Mohan	1990	Active 3D object models.
ICCV	Collaboration between computer graphics and computer vision.	André Gagalowicz	1990	Collaboration between computer graphics and computer vision.
ICCV	The dynamic analysis of apparent contours.	Roberto Cipolla,Andrew Blake	1990	The dynamic analysis of apparent contours.
ICCV	About lacunarity, some links between fractal and integral geometry, and an application to texture segmentation.	Jacques Lévy Véhel	1990	About lacunarity, some links between fractal and integral geometry, and an application to texture segmentation.
ICCV	A Bayesian decision theoretic approach for adaptive goal-directed sensing.	Hsiang-Lung Wu,Alec Cameron	1990	A Bayesian decision theoretic approach for adaptive goal-directed sensing.
ICCV	MRF model-based segmentation of range images.	Anil K. Jain,Sateesha G. Nadabar	1990	MRF model-based segmentation of range images.
ICCV	Reconstructing 3D lines from a sequence of 2D projections: representation and estimation.	Yuh-Lin Chang,Jake K. Aggarwal	1990	Reconstructing 3D lines from a sequence of 2D projections: representation and estimation.
ICCV	Dynamic 3D models with local and global deformations: deformable superquadrics.	Demetri Terzopoulos,Dimitris N. Metaxas	1990	Dynamic 3D models with local and global deformations: deformable superquadrics.
ICCV	Multispectral constraints for optical flow computation.	Vishal Markandey,Bruce E. Flinchbaugh	1990	Multispectral constraints for optical flow computation.
ICCV	Understanding assembly illustrations in an assembly manual without any model of mechanical parts.	Shoujie He,Norihiro Abe,Tadahiro Kitahashi	1990	Understanding assembly illustrations in an assembly manual without any model of mechanical parts.
ICCV	Occlusion detection in early vision.	Peng-Seng Toh,Andrew K. Forrest	1990	Occlusion detection in early vision.
ICCV	A framework for adaptive scale space tracking solutions to problems in computational vision.	Gary Whitten	1990	A framework for adaptive scale space tracking solutions to problems in computational vision.
ICCV	An approach to color constancy using multiple images.	Masato Tsukada,Yuichi Ohta	1990	An approach to color constancy using multiple images.
ICCV	The effect of indexing on the complexity of object recognition.	W. Eric L. Grimson	1990	The effect of indexing on the complexity of object recognition.
ICCV	Compact image representation from multiscale edges.	Sifen Zhong,Stéphane Mallat	1990	Compact image representation from multiscale edges.
ICCV	Shape from texture: the homogeneity hypothesis.	Constantinos Marinos,Andrew Blake	1990	Shape from texture: the homogeneity hypothesis.
ICCV	3D structure from a monocular sequence of images.	Jean-Luc Jezouin,Nicholas Ayache	1990	3D structure from a monocular sequence of images.
ICCV	Sensitivity of the pose refinement problem to accurate estimation of camera parameters.	Rakesh Kumar,Allen R. Hanson	1990	Sensitivity of the pose refinement problem to accurate estimation of camera parameters.
ICCV	A theory of image matching.	Juyang Weng	1990	A theory of image matching.
ICCV	A model for the detection of motion over time.	Michael J. Black,P. Anandan	1990	A model for the detection of motion over time.
ICCV	Perceptual organization of occluding contours.	Lance R. Williams	1990	Perceptual organization of occluding contours.
ICCV	Reconstruction without discontinuities.	Stan Z. Li	1990	Reconstruction without discontinuities.
ICCV	Shape from contour: straight homogeneous generalized cones.	Fatih Ulupinar,Ramakant Nevatia	1990	Shape from contour: straight homogeneous generalized cones.
ICCV	Decomposition theory and transformations of visual directions.	Fredrik Bergholm	1990	Decomposition theory and transformations of visual directions.
ICCV	On the extensive reconstruction of Hough transform.	Hiroyasu Koshimizu,Munetoshi Numada	1990	On the extensive reconstruction of Hough transform.
ICCV	Segmentation of 3D range images using pyramidal data structures.	Bikash Sabata,Farshid Arman,Jake K. Aggarwal	1990	Segmentation of 3D range images using pyramidal data structures.
ICCV	Segmentation of optical flow and 3D data for the interpretation of mobile objects.	Aimé Meygret,Monique Thonnat	1990	Segmentation of optical flow and 3D data for the interpretation of mobile objects.
ICCV	Determining of camera rotation from vanishing points of lines on horizontal planes.	Shigang Li,Saburo Tsuji,Masakazu Imai	1990	Determining of camera rotation from vanishing points of lines on horizontal planes.
ICCV	Cooperative integration of multiple stereo algorithms.	Masaki Watanabe,Yuichi Ohta	1990	Cooperative integration of multiple stereo algorithms.
ICCV	Object recognition by a Hopfield neural network.	Nasser M. Nasrabadi,Wei Li,Chang Y. Choo	1990	Object recognition by a Hopfield neural network.
ICCV	Robust curve detection by temporal geodesics.	Dan Gutfinger,R. Nishimura,H. Doi,Jack Sklansky	1990	Robust curve detection by temporal geodesics.
ICCV	Simultaneous estimation of shape and reflectance maps from photometric stereo.	Hemant D. Tagare,Rui J. P. de Figueiredo	1990	Simultaneous estimation of shape and reflectance maps from photometric stereo.
ICCV	Computing the visual potential of an articulated assembly of parts.	Maha Sallam,John H. Stewman,Kevin W. Bowyer	1990	Computing the visual potential of an articulated assembly of parts.
ICCV	A computational model for face location.	Venu Govindaraju,Sargur N. Srihari,David B. Sher	1990	A computational model for face location.
ICCV	Uncertainty in interpretation of range imagery.	David M. Chelberg	1990	Uncertainty in interpretation of range imagery.
ICCV	Multiple widths yield reliable finite differences.	Margaret M. Fleck	1990	Multiple widths yield reliable finite differences.
ICCV	Extended structure and motion analysis from monocular image sequences.	Ning Cui,Juyang Weng,Paul R. Cohen	1990	Extended structure and motion analysis from monocular image sequences.
ICCV	Modeling the rim appearance.	W. Brent Seales,Charles R. Dyer	1990	Modeling the rim appearance.
ICCV	Multiresolution image acquisition and surface reconstruction.	Subhodev Das,Narendra Ahuja	1990	Multiresolution image acquisition and surface reconstruction.
ICCV	The 2.1-D sketch.	Mark Nitzberg,David Mumford	1990	The 2.1-D sketch.
ICCV	Qualitative 3-D shape reconstruction using distributed aspect graph matching.	Sven J. Dickinson,Alex Pentland,Azriel Rosenfeld	1990	Qualitative 3-D shape reconstruction using distributed aspect graph matching.
ICCV	Direct motion stereo: Recovery of observer motion and scene structure.	Brian Y. Hayashi,Shahriar Negahdaripour	1990	Direct motion stereo: Recovery of observer motion and scene structure.
ICCV	Surface reconstruction using deformable models with interior and boundary constraints.	Yuan-Fang Wang,Jih-Fang Wang	1990	Surface reconstruction using deformable models with interior and boundary constraints.
ICCV	Indexing via color histograms.	Michael J. Swain,Dana H. Ballard	1990	Indexing via color histograms.
ICCV	Terrain matching by analysis of aerial images.	Jeffrey J. Rodríguez,Jake K. Aggarwal	1990	Terrain matching by analysis of aerial images.
ICCV	Steerable filters for early vision, image analysis, and wavelet decomposition.	William T. Freeman,Edward H. Adelson	1990	Steerable filters for early vision, image analysis, and wavelet decomposition.
ICCV	Representing surface curvature discontinuities on curved surfaces.	Hiromi T. Tanaka,Daniel T. L. Lee	1990	Representing surface curvature discontinuities on curved surfaces.
ICCV	Detecting and localizing edges composed of steps, peaks and roofs.	Pietro Perona,Jitendra Malik	1990	Detecting and localizing edges composed of steps, peaks and roofs.
ICCV	Interpolating cubic spline contours by minimizing second derivative discontinuity.	Saeid Tehrani,Terry E. Weymouth,Brian G. Schunck	1990	Interpolating cubic spline contours by minimizing second derivative discontinuity.
ICCV	Detection of interest points using symmetry.	Daniel Reisfeld,Haim J. Wolfson,Yehezkel Yeshurun	1990	Detection of interest points using symmetry.
ICCV	Determining reflectance parameters using range and brightness images.	Katsushi Ikeuchi,Kosuke Sato	1990	Determining reflectance parameters using range and brightness images.
ICCV	The evolution and testing of a model-based object recognition system.	Joseph L. Mundy,Aaron Heller	1990	The evolution and testing of a model-based object recognition system.
ICCV	Epicardial motion and deformation estimation from coronary artery bifurcation points.	Chang Wen Chen,Thomas S. Huang	1990	Epicardial motion and deformation estimation from coronary artery bifurcation points.
ICCV	An approach to 3D scene reconstruction from noisy binocular image sequences using information fusion.	Lang Hong,Dragana Brzakovic	1990	An approach to 3D scene reconstruction from noisy binocular image sequences using information fusion.
ICCV	Calculating surface reflectance using a single-bounce model of mutual reflection.	Mark S. Drew,Brian V. Funt	1990	Calculating surface reflectance using a single-bounce model of mutual reflection.
ICCV	An efficient method for multiple-circle detection.	Xing Cao,Farzin Deravi	1990	An efficient method for multiple-circle detection.
ICCV	Estimation of shape, reflection coefficients and illuminant direction from image sequences.	Naoki Mukawa	1990	Estimation of shape, reflection coefficients and illuminant direction from image sequences.
ICCV	Curved inertia frames and the skeleton sketch: Finding salient frames of reference.	J. Brian Subirana-Vilanova	1990	Curved inertia frames and the skeleton sketch: Finding salient frames of reference.
ICCV	Segmentation as the search for the best description of the image in terms of primitives.	Ales Leonardis,Alok Gupta,Ruzena Bajcsy	1990	Segmentation as the search for the best description of the image in terms of primitives.
ICCV	Recovering 3D motion and structure from stereo and 2D token tracking cooperation.	Nassir Navab,Rachid Deriche,Olivier D. Faugeras	1990	Recovering 3D motion and structure from stereo and 2D token tracking cooperation.
ICCV	Detection of convex and concave discontinuous points in a plane curve.	Joon H. Han	1990	Detection of convex and concave discontinuous points in a plane curve.
ICCV	Feature matching for object localization in the presence of uncertainty.	Todd A. Cass	1990	Feature matching for object localization in the presence of uncertainty.
ICCV	Reconstructing line drawings from wings: the polygonal case.	George C. Stockman,Greg Chungmou Lee,Sei-Wang Chen	1990	Reconstructing line drawings from wings: the polygonal case.
ICCV	Segmenting curves into elliptic arcs and straight lines.	Paul L. Rosin,Geoff A. W. West	1990	Segmenting curves into elliptic arcs and straight lines.
ICCV	Shape from interreflections.	Shree K. Nayar,Katsushi Ikeuchi,Takeo Kanade	1990	Shape from interreflections.
ICCV	Similarity extraction and modeling.	Luc J. Van Gool,Johan Wagemans,J. Vandeneede,André Oosterlinck	1990	Similarity extraction and modeling.
ICCV	Local spatial frequency analysis of image texture.	John Krumm,Steven A. Shafer	1990	Local spatial frequency analysis of image texture.
ICCV	Direct recovery of motion and shape in the general case by fixation.	M. Ali Taalebinezhaad	1990	Direct recovery of motion and shape in the general case by fixation.
ICCV	Robustness of correspondence-based structure from motion.	R. Dutta,M. A. Snyder	1990	Robustness of correspondence-based structure from motion.
ICCV	Learning 3D object recognition strategies.	Bruce A. Draper,Edward M. Riseman	1990	Learning 3D object recognition strategies.
ICCV	Omni-directional stereo for making global map.	Hiroshi Ishiguro,Masashi Yamamoto,Saburo Tsuji	1990	Omni-directional stereo for making global map.
ICCV	A new transform for curve detection.	Ping Liang	1990	A new transform for curve detection.
IEEE Visualization	Dynamic Graphics for Network Visualization.	Richard A. Becker,Stephen G. Eick,E. O. Miller,Allan R. Wilks	1990	Network data involves statistics that are associated with the nodes or links in a network. We describe several dynamic graphics tools for visualizing such data.
IEEE Visualization	Shape Coding of Multidimensional Data on a Microcomputer Display.	J. Beddow	1990	The visual representation of data from complex systems, whether databases, measured scientific data, or simulation output, holds the promise of discovering patterns in the data that will increase its management efficiency while revealing relationships invisible to numeric methods. In this paper we present a simple and flexible method of shape coding for higher dimensional data sets that allows the database operator or discipline scientist quick access to promising patterns within and among records or samples. The example used is a thirteen parameter set of solar wind, magnetosphere, and ground observation data collected hourly for twenty-one days in 1976.
IEEE Visualization	Visualizing Computer Memory Architectures.	Bowen Alpern,Larry Carter,Ted Selker	1990	"The Memory Hierarchy Framework is a conceputal model together with a visual language for using the model. The model is more faithful to the structure of computers than the Von Neumann and Turing models. It addresses the issues of data movement and exposes and unifies storage mechanisms such as cache, translation lookaside buffers, main memory, and disks. The visual language presents the details of a computer's memory hierarchy in a concise drawing composed of rectangles and connecting segments. Using this framework, we have improved the performance of a matrix multiplication algorithm by more than an order of magnitude. We believe the framework gives insight into computer architecture and performance bottlenecks by making effective use of human visual abilities."
IEEE Visualization	Design of an End-User Data Visualization System.	D. L. Brittain,J. Aller,M. Wilson,S. L. C. Wang	1990	We describe the architecture of an end-user visualization system that supports interactive analysis of 3D scalar and vector data in a heterogeneous hardware environment. The system supports a variety of visualization methods having applicability in disciplines such as computational fluid dynamics, earth and space sciences, and finite-element analysis. We discuss general concerns with current visualization methods and present how our approach simplifies the visualization process.
IEEE Visualization	FAST: A Multi-Processed Environment for Visualization of Computational Fluid Dynamics.	Gordon V. Bancroft,Fergus Merrit,Todd Plessel,P. G. Kelaita,R. K. McCabe,Al Globus	1990	"Three-dimensional, unsteady, multi-zoned fluid dynamics simulations over full scale aircraft is typical of problems being computed at NASA Ames' Numerical Aerodynamic Simulation (NAS) facility on CRAY2 and CRAY-YMP supercomputers. With multiple processor workstations available in the 10-30 Mflop range, we feel that these new developments in scientific computing warrant a new approach to the design and implementation of analysis tools. These larger, more complex problems create a need for new visualization techniques not possible with the existing software or systems available as of this writing and these visualization techniques will change as the supercomputing environment, and hence the scientific methods employed, evolve even further.Visualization of computational aerodynamics requires flexible, extensible, and adaptable software tools for performing analysis tasks. Flexible means the ability to handle a diverse range of problems. Extensible means the ability to interact at all levels of the software hierarchy, either through existing built-in functionality or through the implementation of custom ""plug-in"" modules. Adaptable means the ability to adapt to new software and hardware configurations through the use of modular structured programming methods, a graphics library standard, and the use of common network communication protocols (like UNIX sockets) for the distribution of processing.This paper discusses FAST (Flow Analysis Software Toolkit), an implementation of a software system for fluid mechanics analysis that is based on this approach."
IEEE Visualization	Interdisciplinary Visualization: Lessons Learned at NCSA.	D. J. Cox	1990	Interdisciplinary Visualization: Lessons Learned at NCSA.
IEEE Visualization	A Graphical Interface for Robotic Remediation of Underground Storage Tanks.	B. K. Christensen,L. M. Desjarlais	1990	"Sensor-rich, intelligent robots that function with respect to models of their environment have significant potential to reduce the time and cost for the cleanup of hazardous waste while increasing operator safety. An animated graphical interface allows operators to easily and safely program such robots. Sandia National Laboratories is performing experimental investigations into the application of intelligent robot control technology to the problem of removing waste stored in tanks. This paper describes the experimental environment employed at Sandia with particular attention to the hardware and software control environment, and the graphical interface. Intelligent system control is achieved through the integration of extensive geometric and kinematic world models with realtime sensor based control. All operator interactions with the system are through fully animated, graphical representations which validate all operator commands before execution to provide for safe operation. Sensing is used to add information to the robot system's world model and to allow sensor based servo control during selected operations. The results of a first Critical Features Test are reported and the potential for applying advanced intelligent control concepts to the removal of waste in storage tanks is discussed."
IEEE Visualization	Visualization of Non-Linear Engineering FEM Analysis in Manufacturing.	G. W. Edgar	1990	Visualization of Non-Linear Engineering FEM Analysis in Manufacturing.
IEEE Visualization	Rendering and Managing Spherical Data with Sphere Quadtrees.	G. Fekete	1990	"Most databases for spherically distributed data are not structured in a manner consistent with their geometry. As a result, such databases possess undesirable artifacts, including the introduction of ""tears"" in the data when they are mapped onto a flat file system. Furthermore, it is difficult to make queries about the topological relationship among the data components without performing real arithmetic. The sphere quadtree (SQT), which is based on the recursive subdivision of spherical triangles obtained by projecting the faces of an icosahedron onto a sphere, eliminates some of these problems. The SQT allows the representation of data at multiple levels and arbitrary resolution. Efficient search strategies can be implemented for the selection of data to be rendered or analyzed by a specific technique. Furthermore, sphere quadtrees offer significant potential for improving the accuracy and efficiency of spherical surface rendering algorithms as well as for spatial data management and geographic information systems. Most importantly, geometric and topological consistency with the data is maintained."
IEEE Visualization	A Numerical Method for Rendering Spherical Reflections.	David P. Dobkin,E. S. Panduranga,M. Zhu	1990	A Numerical Method for Rendering Spherical Reflections.
IEEE Visualization	Personal Visualization System: Applications in Research and Engineering.	Q. E. Docecek,K. Moorjani,B. F. Kim,D. G. Tilley,Thomas S. Denney Jr.	1990	"This paper describes an innovative personal visualization system and its application to several research and engineering problems. The system bridges both hardware and software components to permit a user to graphically describe a visualization problem to the computer; thereby reducing program development time to a few hours. Low-cost visualization is achieved using PC-based software that can either be executed on the PC or drive graphic workstations for high resolution displays. In either case, supercomputer computation rates are made available to the visualization process. On PC's this is done with one or more PiP plug in cards, each of which is capable of 100 million floating point operations per second. On workstations this is done with the QUEN™ (trademark of Interstate Electronics Corporation, Anaheim, California) array processor."
IEEE Visualization	Visualization of Irregular Multivariate Data.	Thomas A. Foley,David A. Lane	1990	Scientific data is often sampled at irregular spatial locations because of physical constraints, yet most visualization software applies only to gridded or regular data. We will discuss effective techniques for representing scalar and vector valued functions that interpolate to irregularly located data. Special attention will be given to the situations where the sampling domain is a 2D plane, a 3D volume, or a closed 3D surface. The globally defined interpolants can be evaluated on a fine regular grid and they can then be visualized using conventional techniques. Triangular and tetrahedral based visualization techniques are also presented.
IEEE Visualization	Factors Inducing Periodic Breathing in Humans Blunted Hypoxic Sensitivity.	W. E. Fordyce,J. J. Ventrella	1990	Factors Inducing Periodic Breathing in Humans Blunted Hypoxic Sensitivity.
IEEE Visualization	Automatic Illustration of 3D Geometric Models: Surfaces.	D. Dooley,Michael F. Cohen	1990	Computer-generated models are becoming increasingly complex in structure, and therefore more difficult to display within single concise images. To better convey to the viewer overall shape information about the model, the advantages of line drawings and shaded surface renderings need to be used cooperatively. A system based on the techniques of traditional illustrators is described for automatically generating illustrations of complex three dimensional models. The system relies upon a richer set of display primitives, which are also outlined. Algorithmic details for emphasizing significant model components are discussed, and some preliminary results are presented.
IEEE Visualization	A Journey into the Fourth Dimension.	Y. Ke,E. S. Panduranga	1990	This paper shows that by a simple (one way) mapping from quaternions to complex numbers, the problem of generating a 4D Mandelbrot set by iteration of a quadratic function in quaternions can be reduced to iteration of the same function in the complex domain, and thus, the function values in 4D can be obtained by a simple table lookup. The computations are cut down by an order. Simple ways of displaying the fractal without shading and also of fast raytracing such a fractal using the table so generated are discussed. Further speed up in raytracing can be achieved by estimates of a distance of a point from the Mandelbrot set. Animation is a key factor in visualizing 4D objects. Three types of animation are attempted: (a) translation in 4D, (b) rotation in 4D and (c) fly-through in 3D.
IEEE Visualization	Visualization and Three-Dimensional Image Processing of Positron Emission.	Nahum D. Gershon	1990	Visualization and Three-Dimensional Image Processing of Positron Emission.
IEEE Visualization	Methods for Surface Interrogation.	Hans Hagen,Thomas Schreiber,Ernst Gschwind	1990	The geometric modeling of free-form curves and surfaces is of a central importance for sophisticated CAD systems. Apart from the pure construction of these curves and surfaces, the analysis of their quality is equally important in the design process. In the scope of this article, we will discuss various visualization techniques, which have the goal of identifying unwanted curvature regions (inflection points, dents, ...) interactively on screen.
IEEE Visualization	"Techniques for Visualizing Fermat's Last Theorem."	Andrew J. Hanson,P. A. Heng,B. C. Kaplan	1990	"Techniques for Visualizing Fermat's Last Theorem."
IEEE Visualization	Interactive Visualization of Quaternion Julia Sets.	John C. Hart,Louis H. Kauffman,Daniel J. Sandin	1990	The first half of a two-step quaternion Julia set visualization system is described. This step uses a quaternion square root function to adapt the classic inverse iteration algorithm to the quaternions. The augmented version produces a 3-D Julia set defined by a point cloud that can be interactively manipulated on a graphics workstation. Several cues are assigned to the point cloud to increase depth perception. Finally, a short theorem is proven that extends the domain of the inverse iteration method to a rotational family of quadratic quaternion Julia sets.
IEEE Visualization	Surface Representations of Two- and Three-Dimensional Fluid Flow Topology.	James Helman,Lambertus Hesselink	1990	We discuss our work using critical point analysis to generate representations of the vector field topology of numerical flow data sets. Critical points are located and characterized in a two-dimensional domain, which may be either a two-dimensional flow field or the tangential velocity field near a three-dimensional body. Tangent curves are then integrated out along the principal directions of certain classes of critical points. The points and curves are linked to form a skeleton representing the two-dimensional vector field topology.When generated from the tangential velocity field near a body in a three-dimensional flow, the skeleton includes the critical points and curves which provide a basis for analyzing the three-dimensional structure of the flow separation. The points along the separation curves in the skeleton are used to start tangent curve integrations to generate surfaces representing the topology of the associated flow separations.
IEEE Visualization	Wide-band Relativistic Doppler Effect Visualization.	Ping-Kang Hsiung,Robert H. Thibadeau,Christopher B. Cox,Robert H. P. Dunn,M. Wu,P. A. Olbrich	1990	One of the most visible aspects of special relativity is the relativistic Doppler effect --- the dependence of observed radiation wavelengths upon the velocity of the source and the viewing conditions.In this paper, we present a flexible and efficient method to simulate the Doppler shift. This new method has the following features:• Surface properties and light composition are represented by splines as functions of wavelength. The entire electromagnetic (EM) spectrum can therefore be represented efficiently.• Doppler shift and shading operations are performed through the manipulation of spline coefficients. The evaluation of the spline functions is carried out at the end of each shading loop to generate the display R, G, B values.This method simplifies the management of the shift and reduces the calculations necessary to maintain a spectral description of lights and surfaces.The study of astrophysical phenomenon, which are being color-shifted by individual recession velocity and by the expansion of the universe, requires the use of knowledge about the Doppler effect. Our simulations may contribute visual insight and understanding that enhances such knowledge.
IEEE Visualization	Visualization of the Information Age.	L. Herr	1990	Visualization of the Information Age.
IEEE Visualization	Superposing Images with Shadow Casting.	P. C. Hsu,John Staudhamer	1990	Algorithms are presented to render dynamic foreground images with shadows cast on static background scenes. The algorithms are based on the ray-tracing technique. We can produce complex and shaded animation with a lower cost. The target display device for these algorithms is a two-channel display in which one is for the background and the other is for the foreground.
IEEE Visualization	The VIS-5D System for Easy Interactive Visualization.	William L. Hibbard,David A. Santek	1990	The VIS-5D System for Easy Interactive Visualization.
IEEE Visualization	Parallel Coordinates: A Tool for Visualizing Multi-dimensional Geometry.	Alfred Inselberg,B. Dimsdale	1990	"A methodology for visualizing analytic and synthetic geometry in RN is presented. It is based on a system of parallel coordinates which induces a non-projective mapping between N-Dimensional and 2-Dimensional sets. Hypersurfaces are represented by their planar images which have some geometrical properties analogous to the properties of the hypersurface that they represent. A point ← → line duality when N = 2 generalizes to lines and hyperplanes enabling the representation of polyhedra in RN. The representation of a class of convex and non-convex hypersurfaces is discussed together with an algorithm for constructing and displaying any interior point. The display shows some local properties of the hypersurface and provides information on the point's proximity to the boundary. Applications to Air Traffic Control, Robotics, Computer Vision, Computational Geometry, Statistics, Instrumentation and other areas are discussed."
IEEE Visualization	The Application of Transport Theory to Visualization of 3D Scalar Data Fields.	W. Kruger	1990	"This paper describes a visualization model for 3D scalar data fields based on linear transport theory. The concept of ""virtual"" particles for the extraction of information from data fields is introduced. The role of different types of interaction of the data field with those particles such as absorption, scattering, source and color shift are discussed and demonstrated.Special attention is given to possible tools for the enhancement of interesting data features. Random texturing can provide visual insights as to the magnitude and distribution of deviations of related data fields, e.g., originating from analytic models and measurements, or in the noise content of a given data field. Hidden symmetries of a data set can often be identified visually by allowing it to interact with a preselected beam of ""physical"" particles with the attendant appearance of characteristic structural effects such as channeling."
IEEE Visualization	Volume Visualization in Cell Biology.	Arie E. Kaufman,Roni Yagel,Reuven Bakalash,I. Spector	1990	Living cells are 3D objects with a complex inner structural organization. Understanding this organization and its relation to biological function constitutes a major challenge for cell biologists. This paper discusses the special properties of volumetric cell data (e.g., noise, discontinuity, raggedness) and the particular difficulties encountered when trying to visualize them in 3D. We describe some of the solutions we adopted, specifically in surface discrimination and shading. These solutions have been implemented as an integral part of the BioCube system, an environment for volume visualization of cellular structures. The paper presents volume visualization results for the actin cytoskeleton of single cells.
IEEE Visualization	Interpreting a 3D Object from a Rough 2D Line Drawing.	Del Lamb,Amit Bandopadhay	1990	Visualizing the third dimension while designing 3D objects is an awkward process in mechanical CAD systems, given the current state of the art. We describe a computer system that automatically constructs the shape of a 3D object from a single 2D sketch. The method makes it convenient to create and manipulate 3D objects and is thus seen as an intelligent user interface for CAD and 3D graphics applications. The proposed technique is built on well known results in image analysis. These results are applied in conjunction with some perceptual rules to determine 3D structure from a rough line drawing. The principles are illustrated by a computer implementation that works in a nontrivial object domain.
IEEE Visualization	Moving Iconic Objects in Scientific Visualization.	G. David Kerlick	1990	"The visualization and interpretation of multidimensional data in space can be substantially enhanced by the introduction of independently moving visual objects. These bird-oid objects or ""boids"" [1], derive from: (1) ""icons"" which are geometric objects whose shape and appearance are related to the field variables, (2) 3-dimensional cursors by which a user interactively picks a point in space, (3) particle traces, which are numerically integrated trajectories in space, (4) moving frames of vectors along space curves, and (5) ""actors"" which are programming objects which can create and destroy instances of themselves, act according to internal logic, and communicate with each other and with a user. A software prototype in the C++ language has been developed which demonstrates some of the capabilities of these objects for the visualization of scalar, vector, and tensor fields defined over finite elements or finite volumes. Visualization using boids requires fewer rendered graphical primitives, allows a higher degree of interactivity, and permits automated ""knowledge navigation"" amid data which is organized spatially."
IEEE Visualization	Visualization of Free-Form Volumes.	D. Lasser	1990	An algorithm is presented which creates planar as well as arbitrarily curved sections of free form volumes. The algorithm itself can be used as a subroutine for algorithms which are able to perform more general intersections of free form volumes, e.g. boolean operations on two free form volumes.
IEEE Visualization	Techniques for Visualizing 3-Dimensional Manifolds.	Michael J. Laszlo	1990	Computer graphics has long been concerned with representing and displaying surfaces in 3-dimensional space R3. We address the questions of representation and display in a higher dimensional setting, specifically, that of 3-manifolds immersed in R4. We describe techniques for visualizing the cross-section surfaces of a 3-manifold formed by a cutting hyperplane. The manifold is first triangulated, so that the cross-section may be computed on a per tetrahedron basis. The triangulated manifold is stored in a data structure which efficiently supports calculation of curvature. We expect the techniques, which we have implemented on the Personal IRIS, will support a highly-interactive system when implemented on a fine-grain parallel computer such as the Connection Machine.
IEEE Visualization	Exploring N-Dimensional Databases.	J. LeBlanc,Matthew O. Ward,N. Wittels	1990	The ability of researchers in the scientific and engineering community to generate or acquire data far outstrips their ability to analyze it. This problem is even more pronounced when the data is of high dimensionality. Visualization has been identified as a critical technique for exploring data sets, but the visualization tools developed to date have mostly concentrated on the display of low (one to four) dimensional data. Ideally a tool for examining N-dimensional data should allow the presentation of the data in a way that can be intuitively interpreted and allow the display of arbitrary views and subsets of the data. The work presented in this paper describes the creation of such a tool using a technique which we term dimensional stacking.
IEEE Visualization	Interactive Investigations of Fluid Mechanics Data Sets.	S. M. Legensky	1990	Scientific Visualization techniques provide convenient means to graphically represent various types of experimental and computational data. Fluid mechanics data tends to be challenging to visualize. FIELDVIEW is a visual analysis tool designed to facilitate the interactive investigation of these data sets by providing an easy-to-use interface to the flow field data. Operating on NASA Plot 3D format data, FIELDVIEW computes scalar and vector flow quantities and displays them using a variety of representations including animation. An interactive viewing interface allows free motion around the data under study to allow the researcher to locate and study the interesting flow features of three-dimensional fluid dynamic data.
IEEE Visualization	Classifying Visual Knowledge Representations: A Foundation for Visualization Research.	Gerald L. Lohse,Henry H. Rueter,Kevin Biolsi,Neff Walker	1990	This research is an exploratory effort to classify visual representations into homogeneous clusters. Our goal is to determine the type of knowledge conveyed by various visual representations. We collected hierarchical sorting data from twelve subjects. Five principal groups of visual representations emerged from a cluster analysis of sorting data: graphs and tables, maps, diagrams, networks, and icons. Two dimensions appear to distinguish these clusters: the amount of spatial information and cognitive processing effort. Our future research will continue to identify properties that characterize knowledge expressed with visual representations.
IEEE Visualization	Animation Techniques for Chain-Coded Objects.	Anthony J. Maeder	1990	Chain coding is a common boundary representation for objects used in image analysis. When animating digital image sequences for visualization, it is often convenient to retain this representation. Methods for scaling, rotation and elastic deformation of objects based solely on chain code elements are discussed here. Quantized methods transform groups of chain code elements into other groups while incremental methods construct the transformed chain code element by element.
IEEE Visualization	A Procedural Interface for Volume Rendering.	J. L. Montine	1990	"This paper presents a simple, procedural interface for doing volume rendering. The interface is built upon three types of objects: volumes, which contain the data to be visualized, environments, which set up viewing and lighting, and image objects, which convert results to a user definable format. A volume is rendered against a particular environment with the results sent to an image object for conversion. By defining volume qualities such as color, opacity, and gradient in terms of user definable transfer functions, the rendering process is made independent of the data set's underlying representation."
IEEE Visualization	Techniques for the Interactive Visualization of Volumetric Data.	Gregory M. Nielson,Bernd Hamann	1990	"Some ideas and techniques for visualizing volumetric data are introduced. The methods presented are quite different from either ""volume rendering"" techniques or ""surface contour"" methods. All of the methods are conceptually quite simple and rather easy to implement. In addition, they are intended to be used interactively."
IEEE Visualization	Ray Traced Scalar Fields with Shaded Polygonal Output.	R. J. Meyers,M. B. Stephenson	1990	An algorithm is presented for rendering scalar field data which reduces rendering times by as much as two orders of magnitude over traditional full resolution images. Less than full resolution sampling of the scalar field is performed using a fast ray tracing method. The sampling grid points are output as a set of screen based Gouraud shaded polygons which are rendered in hardware by a graphics workstation. A gradient based variable resolution algorithm is presented which further improves rendering speed. Several examples are presented.
IEEE Visualization	Visualizing a Scalar Field on an N-dimensional Lattice.	Ted Mihalisin,E. Grawlinksi,John Timlin,J. Schwegler	1990	A new hierarchical method of plotting is presented which allows one to interactively view millions of data points with up to 10 independent variables.
IEEE Visualization	Extracting Geometric Models Through Constraint Minimization.	James V. Miller,David E. Breen,Michael J. Wozny	1990	Geometric models that represent the features in a greyscale image allow for alternative visualizations of the features. These alternatives can lead to a greater understanding of the extents and interrelationships of the features in the image. We present a technique that extracts the geometric nature of a feature by constraining a known geometric model to conform to the characteristics of the feature. The constraints embody local properties of simple polygons and the nature of the relationship between noise and the features in the image. By minimizing these constraints we are able to deform our initial model to fit the feature.
IEEE Visualization	Visualization of Scalar Data Defined on a Structured Grid: Applications to Petroleum Research.	J. L. Pajon,V. B. Tran	1990	We will describe in this article some simple visualization techniques that may be used to explore dynamic 3D scalar fields in an interactive way. Scalar data are assumed to have been already computed, and graphic manipulations are done afterwards on a graphics workstation. Structured grids (finite difference grids) are used, leading to an easy and fast exploration of the interior of a volume. Smooth animation and simultaneous visualization of two or three scalar fields are also described. We have tested our methods on various types of data issuing from different fields of petroleum engineering, i.e. oil reservoir simulation, geophysics, geology and combustion engine simulations.A demonstration videotape is available.
IEEE Visualization	Spline-Based Color Sequences for Univariate, Bivariate and Trivariate Mapping.	Binh Pham	1990	Colors are often used for displaying data in order to reveal their characteristics and to facilitate the task of data analysis. Much research and experimental work has been carried out to identify principles for making effective use of color and to find good computational models for constructing color schemes for representing data values. Most existing color schemes are based on simple geometric models which facilitates the computation, but do not allow sufficient flexibility for manipulating colors in a sequence. This paper introduces alternative models which uses B-spline curves and surfaces for generating color sequences for univariate, bivariate and trivariate mapping. The models provide more flexibility for interactive modification of the range of color locally in order to obtain customized color schemes for a particular map.
IEEE Visualization	An Interpersonal Multimedia Visualization System.	P. Phillips	1990	An Interpersonal Multimedia Visualization System.
IEEE Visualization	Real-World Applications of Visualization Solutions.	D. A. Prawel	1990	Visual Data Analysis (VDA) methods have found application in aerospace engineering research. VDA is being used to develop new non-destructive evaluation testing techniques for graphite epoxy composites by providing new insights into stress waves propagating through them. Visual Data Analysis was used to analyze stress wave propagation, determine the origin of an unexplained wave distortion, and create a theoretical model to eliminate the distortion utilizing mathematical modeling.
IEEE Visualization	A Methodology for Scientific Visualization: Choosing Representations based on Natural Scene Paradigm.	Philip K. Robertson	1990	A Methodology for Scientific Visualization: Choosing Representations based on Natural Scene Paradigm.
IEEE Visualization	Accurate Display of Tensor Product Isosurfaces.	Alyn P. Rockwood	1990	A general method for rendering isosurfaces of multivariate rational and polynomial tensor products is described. The method is robust up to degree 15, handling singularities without introducing spurious rendering artifacts. It is based on finding real roots of a polynomial in Bernstein form. This makes it particularly suitable for parallel and pipelined processing.It is envisioned that the tensor products will be used as approximants or interpolants for empirical data or scalar fields. An interpolation scheme is given as an example.
IEEE Visualization	ierarchical Triangulation Using Terrain Features.	Lori L. Scarlatos,Theo Pavlidis	1990	ierarchical Triangulation Using Terrain Features.
IEEE Visualization	Volume Microscopy of Biological Specimens Based on Nonconfocal Imaging Techniques.	Stephen L. Senft,V. J. Argio,W. L. van Zandt	1990	Volume Microscopy of Biological Specimens Based on Nonconfocal Imaging Techniques.
IEEE Visualization	Applying Space Subdivision Techniques to Volume Rendering.	Kalpathi R. Subramanian,Donald S. Fussell	1990	We present a new ray-tracing algorithm for volume rendering which is designed to work efficiently when the data of interest is distributed sparsely through the volume. A simple preprocessing step identifies the voxels representing features of interest. Frequently this set of voxels, arbitrarily distributed in three dimensional space, is a small fraction of the original voxel grid. A mediancut space partitioning scheme, combined with bounding volumes to prune void spaces in the resulting search structure, is used to store the voxels of interest in a k-d tree. The tree is then efficiently ray-traced to render the voxel data. The k-d tree is view independent and can be used for animation sequences involving changes in positions of the viewer or positions of lights. We have applied this search structure to render voxel data from MRI, CAT Scan and electron density distributions.
IEEE Visualization	Scattered Data Interpolation Tools in a Microcomputer Visualization Environment.	K. Voegde	1990	Scattered Data Interpolation Tools in a Microcomputer Visualization Environment.
IEEE Visualization	A Problem-oriented Classification of Visualization Techniques.	Stephen Wehrend,Clayton Lewis	1990	Progress in scientific visualization could be accelerated if workers could more readily find visualization techniques relevant to a given problem. This paper describes an approach to this problem based on a classification of visualization techniques that is independent of particular application domains. A user breaks up a problem into subproblems, describes these subproblems in terms of the objects to be represented and the operations to be supported by a representation, locates applicable visualization techniques in a catalog, and combines these representation into a composite representation for the original problem. The catalog and its underlying classification provide a way for workers in different application disciplines to share methods.
IEEE Visualization	A System for Three-Dimensional Acoustic Visualization in a Virtual Environment Workstation.	E. M. Wenzel,S. S. Fisher,P. K. Stone,S. H. Foster	1990	"This paper describes the real time acoustic display capabilities developed for the VIrtual Environment Workstation (VIEW) project at NASA-Ames Research Center. The acoustic display is capable of generating localized acoustic cues in real time over headphones. An auditory symbology, a related collection of representational auditory ""objects"" or ""icons,"" can be designed using ACE, the Auditory Cue Editor, which links both discrete and continuously-varying acoustic parameters with information or events in the display. During a given display scenario, the symbology can be dynamically co-ordinated in real time with three-dimensional visual objects, speech, and gestural displays. The types of displays feasible with the system range from simple warnings and alarms to the acoustic representation of multidimensional data or events."
IEEE Visualization	Displaying Voxel-Based Objects According to Their Qualitative Shape Synthesis.	Yaser Yacoob	1990	The use of qualitative shape synthesis for the display of 3-D binary objects, is presented. Instead of the commonly used shading approaches, we propose that intrinsic shape features of the object(s) be the means for the display. The proposed approach is applicable to multi-object scenes and to outdoor scenery as well. It makes use of a new method, the diffusion process, that simulates diffusion of particles within the interior of a 3-D discrete object. Starting with initial concentrations of particles at the boundary-voxels, the diffusion procedure simulates the propagation of these particles inwards. Boundary voxels of the object are colored according to the concentration of particles obtained by suspending the diffusion process. This method assists shape characterization by providing a qualitative measure of boundary curvature and was used in achieving display of a variety of voxel-based objects. Examples of the use of this approach on synthetic, terrain, and range data, are provided.
IEEE Visualization	A Three-Dimensional/Stereoscopic Display and Model Control System for Great Lakes Forecasts.	C. J. Yen,K. W. Bedford,J. L. Kempf,R. E. Marshell	1990	In a forecasting system for the Great Lakes, the data generated by the three-dimensional numerical model is visualized by a 3D/stereoscopic display module. The module consists of a control panel and a display window with the capability of interactively rendering the results. The event scheduling for scenario testing to steer the 3D numerical model is achieved by a similar panel. These panels set up the simulation and control the data flow between the graphics workstation and supercomputer. Many rendering methods, stereo imagery and animation are incorporated to display the results. Interaction between the user, the workstation and the supercomputer allows steering of the simulation and tracing of the simulation output. Distributed software for post-processing and volume rendering are also used to enhance the representation. Further enhancement will provide operational predictions of Great Lakes physics and transport features with these visualization methods and control algorithms.
IEEE Visualization	"Proceedings IEEE Visualization'90."	Arie E. Kaufman	1990	"Proceedings IEEE Visualization'90."
IEEE Visualization	The Hyperbox.	Bowen Alpern,Larry Carter	1991	A hyperbox is a 2-dimensional depiction of an N-dimensional box (rectangular parallelepiped). This paper defines the visual syntax of hyperboxes, states some properties, and sketches two applications. Hyperboxes can be evocative visual names for tensors or multidimensional arrays in visual programming languages. They can also be used to simultaneously display all pairwise relationships in an N-dimensional dataset. We show that this can be helpful in choosing a sequence of dimension-reducing transformations that preserve interesting properties of the dataset.
IEEE Visualization	The Virtual Windtunnel: An Environment for the Exploration of Three-Dimensional Unsteady Flows.	Steve Bryson,C. Levit	1991	"We describe a recently completed implementation of a virtual environment for exploring numerically-generated three-dimensional unsteady flowfields. A boom-mounted six degree of freedom head-position-sensitive stereo CRT system is used for viewing. A hand position sensitive glove controller is used for injecting various tracers (e.g., ""smoke"") into the virtual flowfield. A mutiprocessor graphics workstation is used for computation and rendering. We describe our techniques for visualizing unsteady flows and discuss the computer requirements for a variety of visualization techniques. These techniques generalize to visualization of other three-dimensional vector fields."
IEEE Visualization	Interactive Data Visualization Using Focusing and Linking.	Andreas Buja,John Alan McDonald,J. Michalak,Werner Stuetzle	1991	This paper discusses two basic principles for interactive visualization of high dimensional data: focusing and linking. The paper and the accompanying video give examples of how graphical data analysis methods based on focusing and linking are used in applications including linguistics, geographic information systems, time series analysis, of multi-channel images arising in radiology and remote sensing.
IEEE Visualization	Visualizing Environmental Data for Program Decision Support.	J. Burnetti,R. Manley,W. Mitchell,D. Varnadore	1991	The McClellan Air Force Base Installation Restoration Program (IRP) is responsible for identifying and remediating environmental contamination from past operations and disposal practices. Since 1979, the IRP has generated over 200 volumes of technical reports regarding the degree and extent of contamination at the base. The base is in the process of automating the storage, retrieval, and analysis of the technical data generated by the cleanup program. This paper discusses the requirements for the IRP technical information system, presents the development approach taken, illustrates visualization results from the system prototype, and outlines future plans for development of the system.
IEEE Visualization	A Scientific Visualization Synthesizer.	Roger Crawfis,M. J. Allison	1991	We describe methods for displaying scientific data using textures and raster operations rather than geometric techniques. The flexibility and simplicity of raster operations allow a greater choice of visualization techniques with only a small set of basic operations. In addition, texture mapping techniques will be shown that allow the representation of several variables simultaneously, without a high degree of clutter. The combination of traditional geometric techniques, image composition techniques and image rendering techniques can be integrated into a single framework for the display of scientific data. This paper presents a system for generating and operating on textures and images for the purposes of scientific visualization. The advantages of using such a system are demonstrated through the use of examples. In particular, the development of bump maps for vector filters and contour lines is demonstrated.
IEEE Visualization	NetV: An Experimental Network-Based Volume Visualization System.	T. Todd Elvins,David R. Nadeau	1991	Volume data imaging is a computationally expensive process. Imaging small volume data sets is slow when executed on typical workstation-class machines and imaging large volume data sets on such machines is nearly impossible. Executing the same tasks on mini-supercomputer class systems can significantly reduce imaging times and job-size restrictions, but at the cost of increased user-interface and accessibility problems. In a well-integrated system, computational tasks should be handled by high-powered compute engines, while user-interface tasks are left to graphics workstations.At the San Diego Supercomputer Center (SDSC), an experimental volume visualization system is being tested that distributes volume imaging tasks to appropriate network resources. Remote high-powered compute engines process rendering tasks, while local workstations run the user-interface.
IEEE Visualization	Gray Scale Diagrams as Business Charts.	W. R. Feeney	1991	"The flood of numeric data in science and technology has prompted new graphic representations called visualization. Business also has a serious need for new graphic representations. One candidate comes from the unlikely gray scale. Gray scale diagrams can present large amounts of quantitative information in a compact format. Hundreds of data points can easily be represented in one diagram, using small gray scale squares (or tiles), without visually overloading a viewer. An experiment was done to compare subjects' responses to answer questions from three types of charts, traditional column and line charts and gray scale tile charts. The results showed that questions were answered more correctly and more quickly using gray scale tile charts than using traditional charts. However, subjects reported they experienced more strain using gray scale tile charts."
IEEE Visualization	Multidimensional Real-Time Visualization on Personal Computers.	Q. E. Dolocek	1991	This case study describes a low-cost, high-performance visualization tool based on the IBM PC. Characteristics of scientific and engineering visualization and requirements for real time analysis are discussed. Application programming without coding by use of flowgraphs is also presented.
IEEE Visualization	Multi-Valued Volumetric Visualization.	Thomas A. Foley,David A. Lane	1991	Effective methods for visualizing several sets of volumetric data simultaneously are presented which involve the composition of multiple volumetric rendering techniques. Although there are many methods for visualizing volumetric data, most of these methods can only render one set of volumetric data at a time. By rendering several volumetric data sets simultaneously, relationships among the data sets can be visualized, which may be difficult or impossible to visualize when the data sets are rendered individually.
IEEE Visualization	Span Filering: An Optimization Scheme for Volume Visualization of Large Finite Element Methods.	R. S. Gallangher	1991	Techniques for displaying 3D isovalues of scalar fields such as stress within a solid finite element model generally involve examining each element for values of interest. An inexpensive, straightforward method is discussed to reduce the number of elements searched for such isovalues. It takes advantage of one traversal of the element data to yield a compact classification of the model by result values and ranges, with no sorting required. This data structure can then relate any scalar isovalue to a set of element groups which are closely inclusive of the isovalue.This method is intended for applications requiring repeated access to the analysis data, such as animation and interactive rendering of isosurfaces and scalar fields. While applicable to general volume visualization problems, it is particularly well suited to optimizing real-valued continuum field result such those found in finite element result data.
IEEE Visualization	Enhanced Visualization of Multi-dimensional Structures: Applications in Positron Emission Tomography and Cimate Data.	Nahum D. Gershon	1991	In this work, we have developed an algorithm based on mathematical morphology, image processing, and volume rendering to enhance the visual perception of definite and abstract structures embedded in multidimensional data undergoing visualization. This erosion procedure enhances the depth and shape perception of structures present in the data beyond the perception facilitated by shading and contrasty colors alone [Gershon, 1990]. The utility of this algorithm is demonstrated in medical imaging (Positron Emission Tomography) and climate (sea surface temperature) data. The resulting information is displayed in stereo.
IEEE Visualization	A Tool for Visualizing the Topology of Three-Dimensional Vector Fields.	Al Globus,C. Levit,T. Lasinski	1991	"We describe a software system, TOPO, that numerically analyzes and graphically displays topological aspects of a three dimensional vector field, v, to produce a single relatively simple picture that characterizes v. The topology of v that we consider consists of its critical points (where v = 0), their invariant manifolds, and the integral curves connecting these invariant manifolds. Many of the interesting features of v are associated with its critical points. The field in the neighborhood of each critical point is approximated by the Taylor expansion. The coefficients of the first non-zero term of the Taylor expansion around a critical point are the 3x3 matrix ▿v. Critical points are classified by examining ▿v's eigenvalues. The eigenvectors of ▿v span the invariant manifolds of the linearized field around a critical point. Curves integrated from initial points on the eigenvectors a small distance from a critical point connect with other critical points (or the boundary) to complete the topology. In addition, one class of critical surfaces important in computational fluid dynamics is analyzed.TOPO is implemented as a module in the FAST [1] visualization environment. FAST is general purpose visualization software with modules for isosurface generation, particle tracing, etc. TOPO operates on curvilinear, structured grids, including large multi-zone grids. We have used TOPO to visualize a number of computational fluid dynamics (CFD) data sets. The results agree well with other topology software and hand generated topologies. TOPO has proved useful in finding surface topology, flow attachment and separation points, vortex cores, scalar field local extrema, and generally interesting regions of v. We believe there may be other interesting applications yet to be discovered. This paper, along with the references, contains most of the information needed for a scientific programmer to code a topology module in another environment."
IEEE Visualization	How Shall We Connect Our Software Tools?	Eric Grosse	1991	Traditionally we connect our software tools using human-readable files. This is a conscious decision to buy flexibility and understandability at some cost in performance relative to binary file formats. This note explores the possibility of using shared memory functions to retain most of the existing style while leapfrogging the speed of reading binary files, at least in some environments and for some applications.
IEEE Visualization	Data Model for Scientific Visualization with Provisions for Regular and Irregular Grids.	Robert B. Haber,Bruce Lucas,Nancy S. Collins	1991	Data Model for Scientific Visualization with Provisions for Regular and Irregular Grids.
IEEE Visualization	Visualization in Computational Fluid Dynamics: A Case Study.	Robert Haimes,D. Darmofal	1991	VISUAL3 is a highly-interactive environment for the visualization of 3D volumetric scientific data. The volume may be broken up in a structured or unstructured manner. Also, the problem may be static or unsteady in time. Because the data is volumetric (not just surfaces), and all the information can be changing, traditional CAD techniques are not appropriate. Therefore VISUAL3 was developed using immediate mode rendering methods.A unique aspect of VISUAL3 is the dimensional windowing approach coupled with cursor mapping which allows efficient pointing in 3D space.VISUAL3 is composed of a large number of visualization tools that can be generally classified into identification, scanning and probing techniques.
IEEE Visualization	Visualizing the Fourth Dimension Using Geometry and Light.	Andrew J. Hanson,P. A. Peng	1991	Visualizing the Fourth Dimension Using Geometry and Light.
IEEE Visualization	Deixis and the Future of Visualization Excellence.	William C. Hill,James D. Hollan	1991	The nature of visualizations and the social uses to which they are put rely heavily on pointing, or to use the linguistic term, deixis. Of particular importance for visualization excellence is an understanding of effective deictic facilities, especially new techniques enabled by computation. In this paper, we (1) explain what deixis is, (2) say why it is fundamental to visualization, (3) analyze some of what is required for effective deixis in the context of emergent visualization technology, and (4) provide an accompanying video demonstration of deictic techniques for visualization.
IEEE Visualization	The Visual Comparison of Three Sequences.	K. P. Hinkley,Matthew O. Ward	1991	Scientists (particularly biologists) currently lack effective tools for comparing multiple sequences of numbers or symbols. This paper describes a method of visual comparison which provides the scientist with a new and unique tool to study the qualitative relationships between three such sequences. The program displays a three-dimensional shape containing the sequence similarities and differences, which manifest themselves as simple geometric shapes and colors that a human observer can easily detect and classify. The method presents all possible correlations to the user, giving it a considerable advantage over existing sequence comparison tools which only search for a programmed subset of all possible correlations. Thus, using this technique, researchers may detect sequence similarities which available analytic methods might completely overlook. The program can also filter out undesirable or insignificant correlations, letting the user focus full attention on the more interesting sequence relationships. The technique enjoys facile adaptation to a wide range of applications, including DNA and protein sequence analysis, speech analysis, signal processing, text comparison, and image analysis.
IEEE Visualization	Visualizing Chemical Kinetics in Fractal Domains.	N. E. Hurlburt,L. W. Anacker,R. Kopelman	1991	Chemical reactions occurring within complex domains, such as fractals, can display behavior which differs radically from the expectations of classical chemical kinetics. Rather than relaxing to a uniform distribution at the steady state, these non-classical systems display large-scale order on many scales. Such self-organization is difficult to measure using the usual statistical techniques, but is visually apparent. Hence visualization is essential to the understanding of these chemical systems. We discuss some of the problems of visualizing chemical kinetics in fractal domains and describe evolution of the visualization as the chemist and visualization scientist collaborated.
IEEE Visualization	Tree maps: A Space-Filling Approach to the Visualization of Hierarchical Information Structures.	B. Johnson,Ben Shneiderman	1991	This paper describes a novel method for the visualization of hierarchically structured information. The Tree-Map visualization technique makes 100% use of the available display space, mapping the full hierarchy onto a rectangular region in a space-filling manner. This efficient use of space allows very large hierarchies to be displayed in their entirety and facilitates the presentation of semantic information.
IEEE Visualization	Integration of Visualization and Scientific Calculation in a Software System.	Ulrich Lang,R. Lang,R. Rule	1991	This paper presents the problems and advantages of the integration of scientific computations and visualization into one common program system. An important point is the direct feedback of information from the visualization into an ongoing simulation. Some strong and weak points of the varying approaches in different software packages are shown. The visualization component of the program system developed at our computer center and the advantages of its integration into the overal system is explained. Finaly the weak points in our system and the open work items to deal with them are described.
IEEE Visualization	Topographical Mapping of Brain Electrical Activity.	S. K. Law,P. L. Nunez,A. F. Westdorp,A. V. Nelson,K. L. Pilgreen	1991	An interdisciplinary approach is applied toward the development of methods to improve spatial resolution of brain electrical activity. Methods to interpolate the potential distribution and to estimate the surface Laplacian from multi-channel data are presented and applied to human evoked potential data. Although developed for electroencephalographic (EEG) data, these spline algorithms can be applied to a variety of fields where visualization of spatial information is desired.
IEEE Visualization	Shadowed Hedgehogs: A Technique for Visualizing 2D Slices of 3D Vector Fields.	R. Victor Klassen,Steven J. Harrington	1991	The technique of placing directed line segments at grid points, known as hedgehogging, has been used for some time as a tool for visualizing two dimensional vector fields. We provide a means of rapidly rendering a slice of a three dimensional field, suitable for a bilevel display. Shape and shadowing are used to disambiguate orientation. Liberal use of lookup tables makes the technique very fast.
IEEE Visualization	Cooperative, Computer-Aided Design of Scientific Visualizations.	Sandeep Kochhar,Mark Friedell,Mark Vincent LaPolla	1991	While the field of scientific visualization has grown rapidly over the last decade, the task of designing graphical displays that effectively depict the data to be visualized is still a time-consuming, difficult, and essentially manual process. This paper proposes an approach to partially automating the process through cooperative computer-aided design (CCAD)--a paradigm that combines the strengths of manual and automated design by interspersing guiding design operations by the human user with the exploration of design alternatives by the computer. We demonstrate this approach in the context of the IVE design system--a CCAD environment for the design of scientific visualizations. Given a set of design requirements, the system is able to generate several alternative visualizations using a set of design rules that combine primitive visualization components in different ways. These alternatives are presented graphically to the user, who can browse through them, select the most promising visualization and refine it manually.
IEEE Visualization	Computer Assisted Sphere Packing in Higher Dimensions.	Nelson L. Max	1991	"A computer was used to help study the packing of equal spheres in dimension four and higher. A candidate for the densest packing in 4-space is described. The configuration of 24 spheres touching a central sphere in this packing is shown to be rigid, unlike the analogue in 3-space, in which the spheres can slide past each other. A system for interactively manipulating and visualizing such configurations is described.The Voronoi cell for a sphere is the set of points closer to its center than to any other sphere center in the packing. The packing density is the ratio of a sphere's volume to the average of the volumes of the Voronoi cells. A method of constructing Voronoi cells and computing their volumes is presented, which works in any dimension. Examples of Voronoi cell volumes are given."
IEEE Visualization	Advanced Visualization on Desktop Workstations.	S. M. Legensky	1991	Computer graphics and visualization play a central role in software tools for Computer-Aided Engineering (CAE) and scientific research. As capabilities of low-cost workstations are enhanced both through price/performance breakthroughs and distributed computing, more sophisticated interactive techniques become locally available to a wider range of users.This paper explores some of the primary challenges that face designers of hardware and software for visualization who are attempting to create tools that will be used and widely accepted. Possible solutions to some of these challenges have been incorporated into FIELDVIEW, a commercial tool for increasing engineering productivity in Computational Fluid Dynamics (CFD).
IEEE Visualization	Color Icons: Merging Color and Texture Perception for Integrated Visualization of Multiple Parameters.	Haim Levkowitz	1991	"Multiparameter data collection is becoming increasingly common in many scientific disciplines. The need exists for integrated displays to enable the scientist to see the combined information, some of which may not be adequately perceived or may not be perceived at all when viewed individually. In order for such displays to succeed, they must exploit the visual perceptual capabilities as much as possible. We present a new technique that harnesses color and texture perception to create integrated displays of two-dimensional ""imagelike"" multiparameter distributions. The power of the new technique is demonstrated by an example of a synthesized dataset comparing it with several other proposed techniques. We discuss the nature of studies required to measure objectively and accurately the effectiveness of such displays."
IEEE Visualization	Golf Green Visualization.	William E. Lorensen,B. Yamron	1991	"Television coverage of golf fails to bring the viewer an appreciation of the complex topography of a golf green and how that topography affects the putting of golf balls, We present a computer graphics simulation that enhances the viewer perception of these features using shaded polygonal models of the actual golf greens used in tournaments. Mathematical modeling of the golf ball's trajectory on its way towards the hole further enhances viewer understanding. A putting from each location on the green to a given pin position. The object-oriented system is written in C and runs on a variety of 3D graphics workstations, As an experiment, the system was used at a professional golf tournament and correctly simulated all putts during the final round."
IEEE Visualization	Image Handling in a Multi-Vendor Environment.	David R. Nadeau,T. Todd Elvins,Michael J. Bailey	1991	"In an ideal world every hardware and software vendor stores the same type of data in the same file format. Software packages transparently exchange their data and results. Different vendor's hardware platforms read and write each other's files without a hitch. And the skies are not cloudy all day.This paper discusses software developed by the San Diego Supercomputer Center (SDSC) to deal with the real world of differing image file formats, mismatched byte orders and word sizes, and confusing hardcopy device interfaces."
IEEE Visualization	Achieving Direct Volume Visualization with Interactive Semantic Region Selection.	Terry S. Yoo,Ulrich Neumann,Henry Fuchs,Stephen M. Pizer,Tim J. Cullip,John Rhoades,Ross T. Whitaker	1991	Interactive direct visualization of 3D data requires fast update rates and the ability to extract regions of interest from the surrounding data. Parallel volume rendering yields rates that make interactive control of image viewing possible for the first time. We have achieved rates as high as 15 frames per second by trading some function for speed. while volume rendering with a full complement of ramp classification capabilities is performed at 1.4 frames per second. These speeds have made the combination of region selection with volume rendering practical for the first time. Semantic driven selection, rather than geometric clipping, has proven to be a natural means of interacting with 3d data. Internal organs in medical data or other regions of interest can be built from preprocessed region primitives. We have applied the resulting combined system to real 3D medical data with encouraging results. The ideas presented are not just limited to our platform, but can be generalized to include most parallel architectures. We present lessons learned from writing fast volume renderers and from applying image processing techniques to viewing volumetric data.
IEEE Visualization	The Asymptotic Decider: Resolving the Ambiguity in Marching Cubes.	Gregory M. Nielson,Bernd Hamann	1991	A method for computing isovalue or contour surfaces of a trivariate function is discussed. The input data are values of the trivariate function, Fijk, at the cuberille grid points (Xi, yj, zk) and the output is a collection of triangles representing the surface consisting of all points where F(x, y, z) is a constant value. The method described here is a modification that is intended to correct a problem with a previous method.
IEEE Visualization	Visualization and Analysis of Multivariate Data: A Technique for all Fields.	Ted Mihalisin,John Timlin,J. Schwegler	1991	Visualization and Analysis of Multivariate Data: A Technique for all Fields.
IEEE Visualization	Two Widely Different Architectural Approaches to Computer Image Generation.	H. W. Park,K. S. Eo,D. L. Kim,B. K. Choi,Y. Kim,T. Alexander	1991	Among several systems from the UWGSP (University of Washington Graphics System Processor) series, two recent architectures were designed for imaging and graphics, One of them, UWGSP3, has already been completed and is being used for selected applications, while the other one, UWGSP4, is being implemented now.These two systems use parallel and pipelined architectures for high performance graphics operations. UWGSP3 uses only commercially available of-the-shelf chips, and consists of a TMS34082 graphics system processor and four TMS34082 floating point coprocessor that can be configured into pipelined or SIMD modes depending on the algorithm. UWGSP4, however, uses dedicated ASIC (Application Specific IC) chips for higher performance, and consists of two main computational parts: a parallel vector processor with 16 vector processing units, used mainly for image processing, and a graphics subsystem which utilizes a parallel pipelined architecture for image synthesis.IN this paper, the computer graphics aspects of both UWGSP3 and UWGSP4 will be described.
IEEE Visualization	Distributed Visualization using Workstations, SuperJacques Lévy-Vehelcomputers and High Speed Networks.	David W. Robertson,V. L. Jacobson,William E. Johnston,S. C. Loken,E. H. Theil,B. L. Tieney	1991	Distributed Visualization using Workstations, SuperJacques Lévy-Vehelcomputers and High Speed Networks.
IEEE Visualization	Acoustic Imaging: The Reconstruction of Underwater Objects.	Lawrence J. Rosenblum,Behzad Kamgar-Parsi,Edward O. Belcher,O. Engelsen	1991	While many sensors now produce a plethora of data, some still produce an insufficient number of samples. This is true for the marine scientific community, because the theory of wave propagation in water dictates the use of acoustic frequencies for underwater imaging. In this paper we discuss reconstruction of 3D scenes using data from an acoustic imaging sonar. Two methods for visualizing objects in an acoustic snapshot of the ocean are discussed: mathematical morphology and a synthesis of 3D digital imaging with volume rendering.
IEEE Visualization	Experiments with Interdisciplinary Projects and Scientific Visualization Applications at the Undergraduate Level.	Nan C. Schaller	1991	This paper describes the interaction between computer graphics students from the computer science department at Rochester Institute of Technology and faculty from various disciplines, in their attempts to utilize state-of-the-art computer graphics techniques for the visualization of physical systems. The structure of a computer graphics course designed to act as the vehicle for this interaction is also described.
IEEE Visualization	Fast Rotation of Volume Data on Data Parallel Architectures.	P. Schroder,J. B. Salem	1991	Data parallel computer architectures hold great promises for high performance computing. Volume visualization (raytracing) is an application that can greatly benefit from these architectures. We describe an algorithm for rendering of orthographic views of volume data on such architectures. In particular the problem of rotating the volume in regard to the communication overhead associated with finely distributed memory is analyzed. We extend an earlier technique (shear decomposition) to 3D and show how this can be mapped onto a data parallel architecture using only grid communication during the resampling associated with the rotation. The rendering uses efficient parallel computation constructs that allow us to use sophisticated shading models and still maintain high speed throughout. This algorithm has been implemented on the Connection MachineR parallel supercomputer and is used in an interactive volume rendering application, with multiple frames per second performance.
IEEE Visualization	The Stream Polygon: A Technique for 3D Vector Field Visualization.	William J. Schroeder,Christopher R. Volpe,William E. Lorensen	1991	We present a new method for the visualization of 3D vector fields called the Stream polygon: a regular, n-sided polygon, oriented normal to the local vector, The polygon can represent local deformation due to rigid body rotation and both normal and shear strain. In addition, the effects of translation and scalar functions may be represented by sweeping the stream polygon along the streamline, and by appropriately varying the radius and shading the surface of the resulting streamtube. In this paper we develop a mathematical foundation for the stream polygon, and provide examples with application to velocity field visualization.
IEEE Visualization	A Fast Ray Casting Algorithm Using Adaptive Isotriangular Subdivision.	R. Shu,A. Lui	1991	Ray casting is a popular method of rendering high quality images from volume data. Its main shortcoming has been the high computational expense. This paper proposes a method of subdividing the image plane with isosceles triangles instead of quadrants as is usually done, resulting in fewer rays being fired without sacrificing image quality. A brief theoretical analysis of the algorithm in comparison with other methods is given.
IEEE Visualization	Visualizing Causal Effects in 4D Space-Time Vector Fields.	Deborah Silver,M. Gao,Norman Zabusky	1991	"In this paper, we present a method to juxtapose 4D space-time vector fields in which one contains a ""source"" variable and the other the ""response"" field. The technique helps highlight the topological relationship between the two in an effort to understand the causal connection. These concepts are applied to ongoing research in evolving fluid dynamical problems."
IEEE Visualization	Designing a Distributed Scientific Visualization Tool.	L. van der Sluis	1991	Technology Applications, Inc. (TAI) provides supercomputing services for the Naval Underwater Systems Center (NUSC) in New London, Connecticut and Newport, Rhode Island. Many NUSC engineers and scientists use the CRAY X-MP/28 to accomplish their work in such diverse areas as acoustic modeling, finite element analysis, and computational fluid dynamics. This paper focuses on the benefits of using a distributed scientific visualization tool in the field of acoustic modeling.
IEEE Visualization	Interactive Data Exploration with a Supercomputer.	Stuart Smith,Georges G. Grinstein,R. Daniel Bergeron	1991	Scientific data of high-dimensionality is particularly difficult to understand if there are complex (and perhaps unknown) interactions among parameters. In such situations, the scientist can turn to a process Tukey has called exploratory data visualization. In exploratory data visualisation, we typically do not know exactly what we are looking for; instead, we explore the data with a variety of visualization techniques that can help us understand the nature of the data by demonstrating patterns in it. Given the complexity of the data and the large number of ways to map the data to a display, exploratory visualization can be truly effective only if it can also be interactive. In this paper we describe an experiment in exploratory data visualization using a massively parallel processor. With this approach we were able to find new features in some of our oldest datasets and to create more vivid presentations of familiar features in these datasets. Our experience has also led us to a better understanding of the nature of exploratory visualization and has resulted in some formal representations of the interaction process in this environment.
IEEE Visualization	Visualization Tools for Semi-Conductor Modeling Software.	Duncan Stevenson	1991	"This case study looks at the issues involved in using a visualisation software package to extend the scope of an existing suite of semi-conductor modelling software.The visualisation software and its hardware platform represent the ""state of the art"" in powerful interactive workstation visualisation systems. The semi-conductor modelling software is the result of joint development between research groups and industry, and is at the leading edge in terms of the underlying models used and the solution techniques employed.The result of this case study is the identification of a range of important issues to be considered when applying ""off the shelf"" visualisation software to a real-world scientific problem."
IEEE Visualization	Volume Rendering of Flow-Visualization Point Data.	P. G. Swann,Sudhanshu Kumar Semwal	1991	"A survey of two and three-dimensional flow-visualization techniques is provided. Our approach is base on applying volume rendering to flow-visualization data. Linear interpolation and B-Spline approximation is used. Several views are given for linear interpolation and B-spline approximation. Our aim is to offer researchers to ""see"" three-dimensional (3-D) views of multiple slices. Thus, relieving them from ""mental"" visualization of two-dimensional (2-D) slices. We conclude by providing suggestions for efficient volume rendering."
IEEE Visualization	Visualizing 4-D Medical Ultrasound Data.	Nils Thune,Bjørn Olstad	1991	There is an increasing interest in visualizing 4-D ultrasound data for use in medical diagnosis. This paper discusses different standard rendering methods applied to 4-D medical ultrasound data. In particular we have tested maximum value projection, sum of values projection, transparent gray level gradient shading, and surface shading. Due to the fact that ultrasound data suffer from a low signal to noise ratio issues such as image processing and image analysis are used with the purpose to enhance and classify the volumetric data set.
IEEE Visualization	Run-Time Visualization of Program Data.	Allan Tuchman,David Jablonowski,George Cybenko	1991	"An important improvement to visualization systems will provide a graphics ""window into an application"" displaying program data at run-time through an easy-to-use graphical interface. With little or no instrumentation of the application the user will be able to dynamically select data for graphical display as the program executes on a remote computer system. The data to be displayed and the type of display to be used are chosen interactively while the application is executing. Any data display can be enabled or disabled at any time; it is not necessary to specify the data or graphics technique before compilation as with conventional graphics tools. We propose an architecture for such a remote visualization system, and describe Vista, our implementation. Designed primarily for scientific visualization, Vista also offers an environment for more effective debugging and program development."
IEEE Visualization	Visualization of Equations in an Interactive Environment.	David Watson,Jakub Wejchert,David W. Williams,Bri M. Collins	1991	The interactive visualization of equations in 3D is potentially an important tool for both research and education. We describe a method of visualizing equations in their explicit form using 3D fields. Equations are written algebraically, interpreted by an equation parser and then expressed as scalar fields. Fields are represented as isosurfaces making use of an algorithm similar to the method of marching cubes. The implementation allows the real time interaction of equation parameters, isosurface and colouring. A variety of applications from mathematics and physics are given together with examples of the construction of data probes using equations.
IEEE Visualization	The Electronic Structure of Oxygen in Silicon as Revealed by Volume Visualization of Ab Initio Calculations.	R. H. Wolfe,M. Needels,J. D. Jaonnopoulos	1991	"We apply volumetric rendering to the interpretation of atomic-scale data generated from quantum molecular dynamics computations. In particular, we examine silicon computations, and we discover that volumetric visualization of the computed 3D electronic charge density is a valuable tool for identifying defect states in silicon lattices in which oxygen atoms occur as impurities. Rendering of several judiciously selected ranges of charge density in translucent colors provides an effective means of identifying broken or altered molecular bonds and induced charge excesses in the lattice. The resulting 3D images reveal important features missed previously in 2D charge density contour maps. We find that stereoscopic ""blink comparison"" of image pairs is an extremely valuable way to study the structural differences among various configurations and that time animation provides significant in sight into the molecular dynamics. We conclude that our success with silicon implies that the technique has the potential to provide a significant benefit to other problems in computational condensed matter physics."
IEEE Visualization	Realistic Volumetric Imaging.	Roni Yagel,Arie E. Kaufman,Q. Zhang	1991	We present a set of volume visualization tools that are based on the use of recursive ray tracing as the primary vehicle for realistic volume imaging. The visualization tools include shadows, mirrors, specularity, and constructive solid geometry. The underlying representation for the ray tracer is a 3D raster of voxels that holds the discrete form of the scene. Unlike traditional volume rendering techniques, our discrete recursive ray tracer models many illumination phenomena by traversing discrete rays in voxel space. Our approach provides true ray tracing of sampled or computed datasets, as well as ray tracing of hybrid scenes where sampled or computed datasets, as well as ray tracing of hybrid scenes where sampled or computed data are intermixed with geometric models. We demonstrate that our techniques enhance the understanding of complex biomedical datasets.
IEEE Visualization	In Vivo Blood Flow Visualization with Magnetic Resonance Imaging.	Guang-Zhong Yang,Peter Burger,Philip J. Kilner,Raad Mohiaddin	1991	Blood movement investigated by magnetic resonance (MR) velocity mapping is generally presented in the form of velocity components in one or more chosen velocity encoding directions. By viewing these components separately, it is difficult for MR practitioners to conceptualize and comprehend the underlying flow structures, especially when the image data has strong background noise. The flow visualization technique presented in this paper adapts the idea of particle tracing used in classical fluid dynamics for visualizing flow. By rendering and animating the processed flow signal, clear flow structures can be revealed.
IEEE Visualization	"Proceedings IEEE Visualization'91."	Gregory M. Nielson,Lawrence J. Rosenblum	1991	"Proceedings IEEE Visualization'91."
IEEE Visualization	Towards a Comprehensive Volume Visualization System.	Ricardo S. Avila,Lisa M. Sobierajski,Arie E. Kaufman	1992	The VolVis system has been developed to satisfy the diverse requirements of the volume visualization community by comfortably housing numerous visualization algorithms and methods within a consistent and well organized framework. The VolVis system is supported by a generalized abstract model which provides for both geometric and volumetric constructs. VolVis contains several rendering algorithms that span the speed versus accuracy continuum. A fast volume rendering algorithm has been developed, which is capable of exploiting existing graphics hardware without placing any viewing restrictions or compromising accuracy. In addition. VolVis includes a volumetric navigation facility, key-frame animation generator, quantitative analysis tools, and a generalized protocol for communicating with 3D input devices.
IEEE Visualization	Virtual Spacetime: An Environment for the Visualization of Curved Spacetimes via Geodesic Flows.	Steve Bryson	1992	We describe an implementation of a virtual environment for visualizing the geometry of curved spacetime by the display of interactive geodesics. This technique displays the paths of particles under the influence of gravity as described by the general theory of relativity, and is useful in the investigation of solutions to the field equations of that theory. A boom-mounted six degree of freedom head position sensitive stereo CRT system is used for display. A hand position sensitive glove controller is used to control the initial positions and directions of geodesics in spacetime. A multiprocessor graphics workstation is used for computation and rendering. We describe and illustrate with examples several techniques for visualizing the geometry of spacetime using geodesics. Though this work is described exclusively in the context of physical four-dimensional spacetimes, it extends to arbitrary geometries in arbitrary dimensions. While this work is intended for researchers, it is also useful for the teaching of general relativity.
IEEE Visualization	Automated Design of Virtual Worlds for Visualizing Multivariate Relations.	Clifford Beshers,Steven Feiner	1992	"Interactive visualization systems provide a powerful means to explore complex data, especially when coupled with 3D interaction and display devices to produce virtual worlds. While designing a quality static 2D visualization is already a difficult task for most users, designing an interactive 3D one is even more challenging. To address this problem, we are developing Auto Visual, a research system that designs interactive virtual worlds for visualizing and exploring multivariate relations of arbitrary arity. Auto Visual uses ""worlds within worlds,"" an interactive visualization technique that exploits nested, heterogeneous coordinate systems to map multiple variables onto each of our spatial dimensions. Auto Visual's designs are guided by user-specified visualization tasks, and by a catalog of design principles encoded using a rule-based language."
IEEE Visualization	Visualization of High-Resolution, 3-D, Nonlinear, Finite Element Analyses.	Mark A. Christon,T. Spelce	1992	Visualization of High-Resolution, 3-D, Nonlinear, Finite Element Analyses.
IEEE Visualization	Volume Rendering on a Distributed Memory Parallel Computer.	T. Todd Elvins	1992	Volume rendering ideally produces high-quality images at interactive rates. High CPU and memory requirements make this goal unreachable with current off-the-shelf technology. Exploiting highly parallel computers is one way that future systems may approach acceptable speeds. This paper discusses the adaptation of a known volume rendering algorithm to a new commercially available distributed memory MIMD parallel architecture.
IEEE Visualization	Visualizing n-Dimensional Implications of 2-D Design Decisions.	Stephen M. Ervin	1992	Visualizing n-Dimensional Implications of 2-D Design Decisions.
IEEE Visualization	Logical Time in Visualizations Produced by Parallel Programs.	Janice E. Cuny,Alfred Hough,Joydip Kunda	1992	"Visualization tools that display data as it is manipulated by a parallel, MIMD computation must contend with the effects of asynchronous execution. We have developed techniques that manipulate logical time in order to produce coherent animations of parallel program behavior despite the presence of asynchrony. Our techniques ""interpret"" program behavior in light of user-defined abstractions and generate animations based on a logical rather than a physical view of time. If this interpretation succeeds, the resulting animation is easily understood; if it fails, the programmer can be assured that the failure was not an artifact of the visualization. Here we demonstrate that these techniques can be generally applied to enhance visualizations of a variety of types of data as it is produced by parallel, MIMD computations."
IEEE Visualization	Visualization of Second-Order Tensor Fields and Matrix Data.	Thierry Delmarcelle,Lambertus Hesselink	1992	We present a study of the visualization of 3-D second order tensor fields and matrix data. The general problem of visualizing unsymmetric real or complex Hermitian second order tensor fields can be reduced to the simultaneous visualization of a real and symmetric second order tensor field and a real vector field. As opposed to the discrete iconic techniques commonly used in multivariate data visualization, the emphasis is on exploiting the mathematical properties of tensor fields in order to facilitate their visualization and to produce a continuous representation of the data. We focus on interactively sensing and exploring real and symmetric second order tensor data by generalizing the vector notion of streamline to the tensor concept of hyperstreamline. We stress the importance of a structural analysis of the data field analogous to the techniques of vector field topology extraction in order to obtain a unique and objective representation of second order tensor fields.
IEEE Visualization	Visualizing Code Profiling Line Oriented Statistics.	Stephen G. Eick,Joseph L. Steffen	1992	"Code tuning is a well-known technique for improving the run-time performance of software. There are several widely used profilers available that show the heavily used functions. Other profilers provide fine grain profiling detail such as basic block counts. The difficulty in using fine grain profilers is coping with the large volumes of data that they generate.We describe a visualization technique that enables us to display and analyze line count profile data. Our technique is to make a reduced picture of code with the line execution counts identified with the color. We show ""hot spots"" in red, ""warm spots"" in orange, and so on. We are also able to identify nonexecuted code and nonexecutable code such as declarations and static tables."
IEEE Visualization	Visualization for the Document Space.	X. Lin	1992	"Visualization for the document space is an important issue for future information retrieval systems. This article describes an information retrieval framework that promotes graphical displays which will make documents in the computer visualizable to the searcher. As examples of such graphical displays, two simulation results of using Kohonen's feature map to generate map displays for information retrieval are presented and discussed. The map displays are a mapping from a high-dimensional document space to a two-dimensional space. They show document relationships by various visual cues such as dots, links, clusters and areas as well as their measurement and spatial arrangement. Using the map displays as an interface for document retrieval systems, we will provide the user with richer visual information to support browsing and searching."
IEEE Visualization	Visualizing the Universe.	M. J. Geller,E. E. Flaco,D. G. Fabricant,B. Estus	1992	"These decades are the first in which we can begin to map the universe. Recent surveys reveal patterns in the distribution of galaxies --- patterns coherent on scales of 150 million light years or more. These patterns contrast with the smoothness of the radiation background measured by the COBE satellite. Together these observations challenge our understanding of the origin of galaxies --- structure --- in the universe.""Visualizing the universe"" is crucial for exploring the 3-dimensional map, for analyzing it, for designing instruments to make deeper maps with new large telescopes, and for sharing the excitement of discovery with the public."
IEEE Visualization	Visualization of Fuzzy Data Using Generalized Animation.	Nahum D. Gershon	1992	"In this paper, we present methods for visualization of fuzzy data based on the sensitivity of the human visual system to motion and dynamic changes and the ease with which electronic display devices can change their display. The methods presented include taking an otherwise static image and displaying in an animation loop either its segmented components or a series of blurred versions of the whole image. This approach was applied to seasurface temperature data and was found to be effective in showing fuzzy details embedded in the data and in drawing the viewer's attention. This approach and methods could play a significant role in the display of browse products for massive data and information systems."
IEEE Visualization	Flow Visualization as a Basic Tool to Investigate the Dynamics and Topology of Jets.	Fernando Grinstein,Upul Obeysekare,Gopal Patnaik	1992	We address relevant issues and difficulties involved in the practical implementation of flow visualization techniques based on database generated in numerical simulations of unsteady square jets. Instantaneous visualizations provide basic information on the topological features of the flow, while animation of these visualizations gives an insight into the detailed dynamics of formation, development and interaction of the coherent structures controlling the entrainment and mixing processes.
IEEE Visualization	Generalized Focal Surfaces: A New Method for Surface Interrogation.	Hans Hagen,Stefanie Hahmann	1992	The generation of smooth surfaces from a mesh of three-dimensional data points is an important problem in geometric modeling. Apart from the pure construction of these curves and surfaces, the analysis of their quality is equally important in the design and manufacturing process. Generalized focal surfaces are presented here as a new surface interrogation tool.
IEEE Visualization	Massively Parallel Isosurface Extraction.	Charles D. Hansen,P. A. Hinker	1992	We describe experiences during the investigation of parallel methods for faster isosurface generation on SIMD machines. A sequential version of a well known isosurfacing algorithm is algorithmically enhanced for a particular type of SIMD architecture. The SIMD implementation takes full advantage of the data parallel nature of the algorithm and experiments have proven the implementation to be highly scalable. A parallel tool, which can generate 170K polygons / second, gives scientists the means to explore large 3D scalar or vector fields interactively.
IEEE Visualization	Four-Dimensional Views of 3-D Scalar Fields.	Andrew J. Hanson,P. A. Hinker	1992	Four-Dimensional Views of 3-D Scalar Fields.
IEEE Visualization	Display of Scientific Data Structures for Algorithm Visualization.	William L. Hibbard,Charles R. Dyer,Brian E. Paul	1992	"We present a technique for defining graphical depictions for all the data types defined in an algorithm, The ability to display arbitrary combinations of an algorithm's data objects in a common frame of reference, coupled with interactive control of algorithm execution, provides a powerful way to understand algorithm behavior. Type definitions are constrained so that all primitive values occurring in data objects are assigned scalar types. A graphical display, including user interaction with the display, is modeled by a special data type. Mappings from the scalar types into the display model type provide a simple user interface for controlling how all data types are depicted, without the need for type-specific graphics logic."
IEEE Visualization	Anatomical Atlases Based on Volume Visualization.	Karl Heinz Höhne,Andreas Pommert,Martin Riemer,Thomas Schiemann,Rainer Schubert,Ulf Tiede,W. Lierse	1992	"In current practice computerized anatomical atlasses are based on a collection of images that can be accessed via a hypermedia program shell. In order to overcome the drawback of a limited number of available views, we propose an approach that uses an anatomical model as data base. The model has a two layer structure. The lower level is a volume model with a set of semantic attributes belonging to each voxel. Its spatial representation is derived from data sets of Magnetic Resonance Imaging and Computer Tomography. The semantic attributes are assigned by an anatomist using a volume editor. The upper level is a set of relations between these attributes which are specified by the expert as well. Interactive visualization tools such as multiple surface display, transparent rendering and cutting are provided. As a substantial feature of the implementation the semantic and the visualization oriented descriptions are stored in a knowledge base. It is shown that the combination of this object oriented data structure with advanced volume visualization tools provides the ""look and feel"" of a real dissection. The concept which even allows simulations like surgery rehearsal, is claimed to be superior to all presently known atlas techniques."
IEEE Visualization	Constructing Stream Surfaces in Steady 3-D Vector Fields.	Jeff P. Hultquist	1992	Constructing Stream Surfaces in Steady 3-D Vector Fields.
IEEE Visualization	SuperGlue: A Programming Environment for Scientific Visualization.	Jeff P. Hultquist,E. L. Raible	1992	Visualization environments have two audiences: scientists and programmers. We suggest that many existing platforms overemphasize ease-of-use and do not adequately address issues of extensibility. We have built a visualization testbed, called SuperGlue, which is particularly suited for the rapid development of new visualization methods. An interpreter supports rapid development of new code and an extensive class hierarchy encourages code reuse.By explicitly designing for ease of programming, we have produced a visualization system which is powerful, easy to use, and rapidly improving. This report describes the motivation of the work, the architecture of the system, and our plans for further development.
IEEE Visualization	Visualizing a 3-D Hydrodynamic Model.	C. S. Jones,J. A. Baca	1992	Visualizing a 3-D Hydrodynamic Model.
IEEE Visualization	Interactive Terrain Rendering van Volume Visualization on the Princeton Engine.	J. Kaba,J. Matey,Gordon Stoll,Herb Taylor,Pat Hanrahan	1992	Interactive Terrain Rendering van Volume Visualization on the Princeton Engine.
IEEE Visualization	Visualization in Anthropology: Reconstruction of Human Fossils.	Alan D. Kalvin,D. Dean,J. J. Hublin,M. Braun	1992	Visualization in Anthropology: Reconstruction of Human Fossils.
IEEE Visualization	Visualization of Simulated Airflow in a Clean Room.	K. Koyamada	1992	Techniques for visualizing a simulated air flow in a clean room are developed by using an efficient cell traverse of tetrahedral cells generated from irregular volumes. The proposed techniques are probing and stream line display, which are related to the measurement techniques used in actual clean rooms. The efficient traverse makes it possible to move freely around a given irregular volume and to spawn off stream lines. A successful application of these techniques to a problem in a clean room is also described.
IEEE Visualization	Virtual Smoke: An Interactive 3-D Flow Visualization Technique.	K. L. Ma,P. J. Smith	1992	Virtual Smoke: An Interactive 3-D Flow Visualization Technique.
IEEE Visualization	Visual Query Specification in a Multimedia Database System.	Daniel A. Keim,Vincent Y. Lum	1992	In this paper we describe a visual interface for a Multimedia Database Management System (MDBMS). In spite of the technological advances in display devices, DBMS query languages are still linear in syntax as it was two decades ago. Although natural language interfaces have been found to be useful, natural language is ambiguous and difficult to process. For queries on standard (relational) data these difficulties may be easily avoided with the use of a visual, graphical interface to guide the user in specifying the query. For image and other media data which are ambiguous in nature, we use natural language processing combined with direct graphical access to the domain knowledge to interpret and evaluate the natural language query. The system fully supports graphical and image input/output in different formats. The combination of visual effect and natural language specification, the support of media data, and the allowance of incremental query specification are very effective to simplify the process of query specification not only for image or multimedia databases but also for all databases.
IEEE Visualization	A 3-D Streamline Tracking Algorithm Using Dual Stream Functions.	David N. Kenwright,G. D. Mallison	1992	A new methodology has been developed for constructing streamlines and particle paths in numerically generated fluid velocity fields. A graphical technique is used to convert the discretely defined flow within a cell into one represented by two three-dimensional stream functions. Streamlines are calculated by tracking constant values of each stream function, a process which corresponds to finding the intersection of two stream surfaces. The tracking process is mass conservative and does not use a time stepping method for integration, thus eliminating a computationally intensive part of traditional tracking algorithms. The method can be applied generally to any three-dimensional compressible or incompressible steady flow. Results presented here compare the performance of the new method to the most commonly used scheme and show that calculation times can be reduced by an order of magnitude.
IEEE Visualization	An Efficient Range Search Algorithm for Visualizing Extrema of Volume Data.	X. Wu,Y. Fang	1992	This article presents a new, fast range search algorithm for visualizing extrema of d-dimensional volume data in real time as the user interactively moves the query range. The new algorithm is based on an efficient data structure, called index heap, which needs only O(N/logN) space and O(d2dN) preprocessing time to be set up, where N is the size of d-dimensional data volume. The algorithm can answer an extremum query in O(4d) expected time and its worst case time complexity is O(2d log N) per query. For dimensions two and three, we can consider that the range search for extrema is effected in average O(1) time per query independent of the size of query range. Unlike the present range query algorithms in the computational geometry literature, the proposed algorithm is very simple and can be easily implemented by practitioners.
IEEE Visualization	A Scientific Visualization Renderer.	Bruce Lucas	1992	While scientific visualization systems share many requirements with other graphical applications, they also pose special requirements that make solutions based on standard rendering hardware or software not entirely satisfactory. This paper illustrates those requirements by describing the renderer used in a production scientific visualization system, Data Explorer [1].We begin by setting forth the requirements for a visualization renderer derived from our experiences. We then describe implementation techniques used to meet the requirements of parallelism, volume rendering of irregular data, clipping, and integration of rendering modalities. The renderer described is a software renderer, but it is hoped that the requirements and implementation described here might influence the design of future generations of rendering hardware.
IEEE Visualization	An Architecture for a Scientific Visualization System.	Bruce Lucas,G. D. Abrams,Nancy S. Collins,D. A. Epstien,Donna L. Gresh,Kevin P. McAuliffe	1992	This paper describes the architecture of Data Explorer, a scientific visualization system. Data Explorer supports visualization of a wide variety of data by means of a flexible set of visualization modules. This paper discusses five elements of the system architecture: 1) A single powerful data model common to all modules that allows a wide range of data types to be imported and passed between modules. 2) Integral support for parallelism, affecting the data model and the execution model. 3) A powerful set of visualization modules that are highly interoperable, due in part to the common data model, and exemplified by the renderer. 4) An execution model designed to facilitate parallelization of modules and incorporating optimizations such as caching. 5) A two-process clientserver system structure consisting of a user interface that communicates with an executive via a dataflow language.
IEEE Visualization	Visualization of Cardiac Bioelectricty - A Case Study.	Robert S. MacLeod,N. E. Harrison,Christopher R. Johnson,Michael A. Matheson	1992	Visualization of Cardiac Bioelectricty - A Case Study.
IEEE Visualization	Approximation and Rendering of Volume Data Using Wavelet Transforms.	Shigeru Muraki	1992	A method is presented to obtain a unique shape description of an object by using wavelet transforms. Wavelet transform is a new signal analysis technique which decomposes a signal using a family of functions having a local property in both time and frequency domains. A multiresolution expression of 3D volume data was first obtained by applying 3D orthogonal wavelet transforms, with the shape then being approximated with a relatively small number of 3D orthogonal functions using only the significant functions. In addition, the resolution of the approximation can be varied point by point using the local property of the wavelets. The method is applied to real volume data, i.e., facial range data and MR images of a human head, and typical results are shown.
IEEE Visualization	Network Video Device Control.	David R. Nadeau,Michael J. Bailey	1992	"Today's highly-networked visualization computing environments include systems from a wide variety of hardware vendors, each running its own operating system and sporting its own input, output, display, and storage peripherals. Faced with such a bewildering variety of hardware and software, today's visualization user is in dire need of software systems that integrate these resources over a network and allow him or her to take maximum advantage of them with a minimum of hassle and networking technical knowledge.Video equipment has become one such network resource. Visualization video equipment is used to record visualization animations, process the video signal, play back animations at varying speeds forward and backward, and edit animations into polished final productions. Computer control of video equipment allows many of these operations to be performed automatically or through slick graphical user interfaces. However, the single serial communications line connecting the video device to a host means that one can only access that device via that host. This paper discusses the Video Tools software developed at the San Diego Supercomputer Center (SDSC) to overcome these restrictions and turn a site's video equipment into a network-accessible resource."
IEEE Visualization	Visualizing Wind Velocities by Advecting Cloud Textures.	Nelson L. Max,Roger Crawfis,D. Williams	1992	In order to visualize both clouds and wind in climate simulations, we have rendered clouds using a 3-D texture which is advected by the wind flow.
IEEE Visualization	Visualizing Seafloor Structures with Satellite Gravity Measurements.	J. McLeod,C. Small	1992	"In recent years, the development of satellite altimetry has opened previously unexplored regions of the world's oceans to exploration by marine geoscientists. By correlating sea surface gravity with seafloor depth, researchers gain insight into the kinematics of seafloor spreading and the physical mechanisms by which the seafloor is formed and deformed. Given that the ocean comprises two thirds of the earth's surface, under which lies a vastly unexplored domain, visualization holds the key in isolating regions of exploration and providing an encompassing perspective that can not be achieved in nature. This case study describes the process of seafloor spreading and discusses some of the concerns with visualizing gravitational acceleration."
IEEE Visualization	"The Microscopist's Workstation."	Philip J. Mercurio,T. Todd Elvins,Stephen J. Young	1992	"This case study looks at the issues involved in operating a sophisticated scientific instrument as a computer peripheral accessible over a high-speed network.A custom interactive visualization application was constructed to support investigation using a unique computer-controlled high voltage electron microscope. The researcher's workstation forms the visible third of a triumvirate, with the instrument and the compute resource comprising the other two parts. The software was designed to support not only image acquisition, but also many of the tasks that microscope researchers perform in analyzing images.The result of this case study is the identification of some of the issues regarding interacting with scientific instrumentation over high-speed networks and the construction of custom applications to support many of the tasks within a laboratory's research methodology."
IEEE Visualization	Visualization of Neutron Scattering Data Using AVS.	R. Popovic	1992	This case study is a result of a six weeks feasibility exercise, the aim of which was to explore the extent to which the existing visualisation software can be used for visualising ISIS Neutron Scattering Data.ISIS, situated at Rutherford Appleton Laboratory, is a world class experimental facility devoted to the use of pulsed neutrons and muons to investigate the microscopic structure and dynamics of all classes of condensed matter.The result of the feasibility study demonstrated the benefits of using visualisation in exploring material science data and it also proved that it is possible to satisfy most of the requirements ISIS researchers place on a software environment by using AVS and without writing any new code.The paper lists the problems encountered and where possible suggests solutions.
IEEE Visualization	Color, Change and Control for Quantitative Data Display.	Penny Rheingans	1992	Color is used widely and reliably to display the value of a single scalar variable. It is more rarely, and far less reliably, used to display multivariate data. Calico, a dynamic tool for the creation and manipulation of color mappings, adds the element of dynamic control over the color mapping to that of color itself for the more effective display and exploration of multivariate spatial data. Using Calico, a one-or two variable color mapping can be created using parametric equations and a variety of color models. This mapping can be manipulated by moving input devices referenced in the parametric expressions, by applying affine transforms, or by performing free-form deformations. As the user changes the mapping, an image showing the data displayed using the current mapping is updated in real time, along with the images describing the mapping. This paper presents an empirical study investigating the effects of user control and smooth change in the display of quantitative data on user accuracy, confidence, and preference. While the user control did not quite produce a significant effect on accuracy, it did produce significant increases in user preference and confidence.
IEEE Visualization	Visualization Requirements in the Atmospheric and Environmental Sciences (Five Case Study Reports).	Theresa-Marie Rhyne,M. Bolstad,Penny Rheingans,L. Petterson,W. Shackelford,Mike E. Botts,E. Pepke,K. W. Johnson,William L. Hibbard,Charles R. Dyer,Brian E. Paul,Lloyd Treinish	1992	Visualization Requirements in the Atmospheric and Environmental Sciences (Five Case Study Reports).
IEEE Visualization	Interactive Visualization of Large Scalar Voxel Fields.	Georgios Sakas,J. Hartig	1992	"We present a technique allowing interactive visualization of large, scalar, discrete volume fields as semi-transparent clouds ""on the fly"", i.e., without preprocessing. Interactivity is not restricted to geometric transformations, but includes all possible methods of processing the data (thresholds, filtering, clipping, illumination, etc.). Our system flexibly trades-off quality for performance at any desirable level. In particular, by using a scanline based method and a DDA-based traversing scheme instead of ray-tracing we achieve real-time processing during previewing. By means of the ""pyramidal volume"" traversing technique we achieve highquality, constant-time filtering independent of the data resolution. Several filters help to detect ""fuzzy"", obscured hot spots, even within noisy data. Our visualization pipeline allows the application of filters at four different stages, maximizing thereby flexibility. Four different illumination models have been implemented. Our method can be used in numerous areas, including 3-D medical imaging, geologic, technical, and environmental applications."
IEEE Visualization	Optimizing Triangulation by Curvature Equalization.	Lori L. Scarlatos,Theo Pavlidis	1992	Triangulated irregular networks (TINs) are an attractive form of surface approximation because triangle vertices and edges may be adaptively selected to produce a good fit with a minimal number of triangles. Finding methods for selecting these vertices and edges, however, is still an active area of research. Numerous refinement algorithms have been proposed, but these may produce more triangles than necessary. In this paper we present an algorithm that attempts to improve a triangulation by shifting the vertices so that curvature within the triangles is nearly equal. In addition, unnecessary triangles are removed. We finish with results produced by running this algorithm on simple geometric surfaces and real terrain data.
IEEE Visualization	VISAGE: An Object-Oriented Scientific Visualization System.	William J. Schroeder,William E. Lorensen,G. D. Montanaro,Christopher R. Volpe	1992	VISAGE is a scientific visualization system implemented in an object-oriented, message passing environment. The system includes over 500 classes ranging from visualization and graphics to Xlib and Motif user interface. Objects are created using compiled C and interact through an interpreted scripting language. The result is a flexible yet efficient system that has found wide application in our user community. This paper describes the object architecture and the major issues we faced when designing the visualization classes. Sample applications are also described.
IEEE Visualization	Surface Curvature Analysis Using Color.	L. R. Seidenberg,Robert B. Jerard,J. Megewick	1992	In the automotive industry it is highly important that the exterior body panels be esthetically pleasing. One aspect of creating esthetically pleasing surfaces is to require that they be fair. In this paper we present a system which has proven useful for diagnosis of surface fairness problems. We describe how to choose a set of colors with perceptually uniform spacing, and also show the usefulness of a logarithmic scale for relating curvature to colors.
IEEE Visualization	A Characterization of the Scientific Data Analysis Process.	R. R. Springmeyer,Meera Blattner,Nelson L. Max	1992	Extensible scientific visualization tools are often offered as data analysis tools. While images may be the goal of visualization, insight is the goal of analysis. Visualization tools often fail to reflect this fact both in functionality and in their user interfaces, which typically focus on graphics and programming concepts rather than on concepts more meaningful to end-user scientists. This paper presents a characterization which shows how data visualization fits into the broader process of scientific data analysis. We conducted an empirical study, observing scientists from several disciplines while they analyzed their own data. Examination of the observations exposed process elements outside conventional image viewing. For example, analysts queried for quantitative information, made a variety of comparisons, applied math, managed data, and kept records. The characterization of scientific data analysis reveals activity beyond that traditionally supported by computer. It offers an understanding which has the potential to be applied to many future designs, and suggests specific recommendations for improving the support of this important aspect of scientific computing.
IEEE Visualization	Representing Medical Images with Partitioning Trees.	Kalpathi R. Subramanian,Bruce F. Naylor	1992	Discrete space representation of images arise as a consequence of the transducers between the physical and informational domains. While discrete representations (arrays of pixels) are simple, they are also verbose and structureless. We present a method of converting between a discrete space representation to a particular continuous space representation, viz. the <u>binary space partitioning tree</u>. The conversion is accomplished using standard discrete space operators developed for edge detection, followed by a Hough transform to generate candidate hyperplanes that are used to construct the partitioning tree. The result is a segmented and compressed image represented in continuous space suitable for elementary computer vision operations and improved image transmission/storage. The method is more noise tolerant than methods whose target is a topological representation, and more adaptive than axis-aligned spatial partitioning schemes. Affine transformations needed for interactive manipulation are fast and edges do not blur with enlargement of the image. Efficient algorithms are known for spatial operations, such as masking/clipping and compositing. We give several examples of 256x256 medical images for which we have estimated the compression to range between 1 and 0.5 bits/pixel.
IEEE Visualization	Volume Warping.	T. J. True,J. F. Hughes	1992	We present volume warping, a technique for deforming sampled volumetric data using B-splines that is related to image warping and to the free-form deformations of Sederberg/Parry and Coquillart. We accelerate the process to near-real-time speed, and explain the compromises that are made to effect such speeds. This technique expands the repertoire of volumetric modeling techniques, and can be applied to any form of volumetric data.
IEEE Visualization	Improving the Visualization of Hierarchies with Treemaps: Design Issues and Experimentation.	David Turo,B. Johnson	1992	Controlled experiments with novice treemap users and real data highlight the strengths of treemaps and provide direction for improvement. Issues discussed include experimental results, layout algorithms, nesting offsets, labeling, animation and small multiple displays. Treemaps prove to be a potent tool for hierarchy display. The principles discussed are applicable to many information visualization situations.
IEEE Visualization	Rendering Surface Particles.	Jarke J. van Wijk	1992	Rendering Surface Particles.
IEEE Visualization	Interactive Splatting of Nonrectilinear Volumes.	Peter L. Williams	1992	This paper describes various techniques for achieving interactive direct volume rendering of nonrectilinear data sets using fast projection (splatting) methods. The use of graphics hardware, rendering approximations, parallelization and reduced resolution meshes are discussed. Results from the use of these techniques are presented in the form of color photos and comparative timings.
IEEE Visualization	The State of the Art of Visual Languages for Visualization.	Carla S. Williams,John Rasure,C. Hansen	1992	Data flow visual language systems are now being used to provide sophisticated environments for the visualization of scientific data. These systems are evolving rapidly and are beginning to encompass related technologies such as distributed computing and user interface development systems. This paper presents a hierarchical classification of the components and issues involved, giving the reader an understanding of the design decisions and trade-offs that the developers of these systems are making. The component categories can be used as a framework for discussing where interoperability of competing visual programming environments might occur and what the future holds for these systems.
IEEE Visualization	Direct Volumetric Visualization.	R. D. Williams,F. L. Wefer,T. E. Clifton	1992	The need for true 3D volumetric displays derives from the fact that the human visual system has evolved to perceive and comprehend the world in three dimensions. The Texas Instruments OmniView&trade; display technology provides the capability to display computer graphics images in true 3D. Viewers can see the display from all angles and change views simply by walking around the display volume. The OmniView&trade;; device uses lasers of three different colors to project images on a moving surface sweeping through the 3D cylindrical display volume. The concept of operations is discussed, along with some details of the moving surface and parameters of the resulting display. The transport theory for this type of display involves only the source term in the linear Boltzmann equation. Constraints of the technology imposed by this limitation are discussed, along with discussions of image quality, current applications, and plans for improved devices.
IEEE Visualization	A Voxel-Based, Forward-Projection Algorithm for Rendering Surface and Volumetric Data.	John R. Wright,Julia C. Hsieh	1992	In this paper we present a voxel-based, forward projection algorithm with a pipeline architecture for real-time applications. The multi-sensor capabilities (electro-optical, or visual, and infrared) currently implemented in software have also been applied to non-real-time imaging applications on workstations and minicomputers. Most suited for terrain based applications, the system features haze, imbedded targets, moving objects, smooth shading, and specular reflections.
IEEE Visualization	Visualizing Classical Problems in CFD.	Norman Zabusky,Deborah Silver	1992	Visualizing Classical Problems in CFD.
IEEE Visualization	"Proceedings IEEE Visualization '92"	Arie E. Kaufman,Gregory M. Nielson	1992	"Proceedings IEEE Visualization '92"
ICCV	Direct estimation of multiple disparities for transparent multiple surfaces in binocular stereo.	Masahiko Shizawa	1993	Direct estimation of multiple disparities for transparent multiple surfaces in binocular stereo.
ICCV	2-D digital curve analysis: A regularity measure.	Bruno Vasselle,Gérard Giraudon	1993	2-D digital curve analysis: A regularity measure.
ICCV	Reflectance ratio: A photometric invariant for object recognition.	Shree K. Nayar,Ruud M. Bolle	1993	Reflectance ratio: A photometric invariant for object recognition.
ICCV	Renormalization for unbiased estimation.	Kenichi Kanatani	1993	Renormalization for unbiased estimation.
ICCV	A modal framework for correspondence and description.	Stan Sclaroff,Alex Pentland	1993	A modal framework for correspondence and description.
ICCV	Recovering reflectance and illumination in a world of painted polyhedra.	Pawan Sinha,Edward H. Adelson	1993	Recovering reflectance and illumination in a world of painted polyhedra.
ICCV	Learning recognition and segmentation of 3-D objects from 2-D images.	John (Juyang) Weng,Narendra Ahuja,Thomas S. Huang	1993	Learning recognition and segmentation of 3-D objects from 2-D images.
ICCV	Relative depth from vergence micromovements.	Antônio Francisco	1993	Relative depth from vergence micromovements.
ICCV	Tracking non-rigid objects in complex scenes.	Daniel P. Huttenlocher,Jae J. Noh,William Rucklidge	1993	Tracking non-rigid objects in complex scenes.
ICCV	Texture discrimination by local generalized symmetry.	Yoram Bonneh,Daniel Reisfeld,Yehezkel Yeshurun	1993	Texture discrimination by local generalized symmetry.
ICCV	Diffuse shading, visibility fields, and the geometry of ambient light.	Michael S. Langer,Steven W. Zucker	1993	Diffuse shading, visibility fields, and the geometry of ambient light.
ICCV	Extracting projective structure from single perspective views of 3D point sets.	Charlie Rothwell,David A. Forsyth,Andrew Zisserman,Joseph L. Mundy	1993	Extracting projective structure from single perspective views of 3D point sets.
ICCV	Recognizing algebraic surfaces from their outlines.	David A. Forsyth	1993	Recognizing algebraic surfaces from their outlines.
ICCV	Cooperation of visually guided behaviors.	Jana Kosecka,Ruzena Bajcsy	1993	Cooperation of visually guided behaviors.
ICCV	Integration of quantitative and qualitative techniques for deformable model fitting from orthographic, perspective, and stereo projections.	Dimitris N. Metaxas,Sven J. Dickinson	1993	Integration of quantitative and qualitative techniques for deformable model fitting from orthographic, perspective, and stereo projections.
ICCV	Automatic feature point extraction and tracking in image sequences for unknown camera motion.	Qinfen Zheng,Rama Chellappa	1993	Automatic feature point extraction and tracking in image sequences for unknown camera motion.
ICCV	Dynamic fixation [active vision].	Kourosh Pahlavan,Tomas Uhlin,Jan-Olof Eklundh	1993	Dynamic fixation [active vision].
ICCV	Multiple knowledge sources and evidential reasoning for shape recognition.	B. Besserer,S. Estable,B. Ulmer	1993	Multiple knowledge sources and evidential reasoning for shape recognition.
ICCV	Head-centered orientation strategies in animate vision.	Enrico Grosso,Dana H. Ballard	1993	Head-centered orientation strategies in animate vision.
ICCV	Optimal estimation of object pose from a single perspective view.	Thai-Quynh Phong,Radu Horaud,Adnan Yassine,Dinh-Tuan Pham	1993	Optimal estimation of object pose from a single perspective view.
ICCV	A binocular stereo algorithm for reconstructing sloping, creased, and broken surfaces in the presence of half-occlusion.	Peter N. Belhumeur	1993	A binocular stereo algorithm for reconstructing sloping, creased, and broken surfaces in the presence of half-occlusion.
ICCV	Motion segmentation and local structure.	David Sinclair	1993	Motion segmentation and local structure.
ICCV	Visual echo analysis.	Esfandiar Bandari,James J. Little	1993	Visual echo analysis.
ICCV	Large deformable splines, crest lines and matching.	André Guéziec	1993	Large deformable splines, crest lines and matching.
ICCV	Motion detection robust to perturbations: A statistical regularization and temporal integration framework.	Jean-Michel Létang,Veronique Rebuffel,Patrick Bouthemy	1993	Motion detection robust to perturbations: A statistical regularization and temporal integration framework.
ICCV	Mathematical morphology: The Hamilton-Jacobi connection.	Alan B. Arehart,Luc Vincent,Benjamin B. Kimia	1993	Mathematical morphology: The Hamilton-Jacobi connection.
ICCV	Segmentation and 2D motion estimation by region fragments.	Minoru Etoh,Yoshiaki Shirai	1993	Segmentation and 2D motion estimation by region fragments.
ICCV	The reciprocal-wedge transform for space-invariant sensing.	Frank Tong,Ze-Nian Li	1993	The reciprocal-wedge transform for space-invariant sensing.
ICCV	Multiscale Markov random field models for parallel image classification.	Zoltan Kato,Marc Berthod,Josiane Zerubia	1993	Multiscale Markov random field models for parallel image classification.
ICCV	A quantitative methodology for analyzing the performance of detection algorithms.	Tapas Kanungo,Mysore Y. Jaisimha,John Palmer,Robert M. Haralick	1993	A quantitative methodology for analyzing the performance of detection algorithms.
ICCV	Experiments with monocular visual tracking and environment modeling.	Olli Silvén,Tapio Repo	1993	Experiments with monocular visual tracking and environment modeling.
ICCV	Robust structure from motion using motion parallax.	Roberto Cipolla,Yasukazu Okamoto,Yoshinori Kuno	1993	Robust structure from motion using motion parallax.
ICCV	An improved algorithm for algebraic curve and surface fitting.	Gabriel Taubin	1993	An improved algorithm for algebraic curve and surface fitting.
ICCV	A note on existence and uniqueness in shape from shading.	Ryszard Kozera	1993	A note on existence and uniqueness in shape from shading.
ICCV	Estimation of the light source distribution and its use in integrated shape recovery from stereo and shading.	Darrell R. Hougen,Narendra Ahuja	1993	Estimation of the light source distribution and its use in integrated shape recovery from stereo and shading.
ICCV	Combining stereo and motion analysis for direct estimation of scene structure.	Keith J. Hanna,Neil E. Okamoto	1993	Combining stereo and motion analysis for direct estimation of scene structure.
ICCV	A finite element model for 3D shape reconstruction and nonrigid motion tracking.	Tim McInerney,Demetri Terzopoulos	1993	A finite element model for 3D shape reconstruction and nonrigid motion tracking.
ICCV	Stereo matching, reconstruction and refinement of 3D curves using deformable contours.	Benedicte Bascle,Rachid Deriche	1993	Stereo matching, reconstruction and refinement of 3D curves using deformable contours.
ICCV	Robust 3D-3D pose estimation.	Xinhua Zhuang,Yan Huang	1993	Robust 3D-3D pose estimation.
ICCV	Grasping visual symmetry.	Andrew Blake,Michael J. Taylor,Adrian Cox	1993	Grasping visual symmetry.
ICCV	The critical sets of lines for camera displacement estimation: A mixed Euclidean-projective and constructive approach.	Nassir Navab,Olivier D. Faugeras,Thierry Viéville	1993	The critical sets of lines for camera displacement estimation: A mixed Euclidean-projective and constructive approach.
ICCV	A framework for the robust estimation of optical flow.	Michael J. Black,P. Anandan	1993	A framework for the robust estimation of optical flow.
ICCV	Enhanced image capture through fusion.	Peter J. Burt,Raymond J. Kolczynski	1993	Enhanced image capture through fusion.
ICCV	Occam algorithms for computing visual motion.	Haim Schweitzer	1993	Occam algorithms for computing visual motion.
ICCV	3D object recognition by indexing structural invariants from multiple views.	Rakesh Mohan,Daphna Weinshall,Ramesh R. Sarukkai	1993	3D object recognition by indexing structural invariants from multiple views.
ICCV	A computational model of neural contour processing: Figure-ground segregation and illusory contours.	Friedrich Heitger,Rüdiger von der Heydt	1993	A computational model of neural contour processing: Figure-ground segregation and illusory contours.
ICCV	A linear complexity procedure for labelling line drawings of polyhedral scenes using vanishing points.	Pietro Parodi,Vincent Torre	1993	A linear complexity procedure for labelling line drawings of polyhedral scenes using vanishing points.
ICCV	Robust vergence with concurrent detection of occlusion and specular highlights.	Wee-Soon Ching,Peng-Seng Toh,Kap-Luk Chan,Meng-Hwa Er	1993	Robust vergence with concurrent detection of occlusion and specular highlights.
ICCV	Euclidean constraints for uncalibrated reconstruction.	Boubakeur Boufama,Roger Mohr,Francoise Veillon	1993	Euclidean constraints for uncalibrated reconstruction.
ICCV	Minimum description length based 2D shape description.	Mengxiang Li	1993	Minimum description length based 2D shape description.
ICCV	Probabilistic relaxation for matching problems in computer vision.	Josef Kittler,William J. Christmas,Maria Petrou	1993	Probabilistic relaxation for matching problems in computer vision.
ICCV	"Active exploration: Knowing when we're wrong."	Peter Whaite,Frank P. Ferrie	1993	"Active exploration: Knowing when we're wrong."
ICCV	A perceptually plausible model for global symmetry detection.	Francois Labonte,Yerucham Shapira,Paul Cohen	1993	A perceptually plausible model for global symmetry detection.
ICCV	Surface discontinuities in range images.	Tomás Pajdla,Václav Hlavác	1993	Surface discontinuities in range images.
ICCV	A system for automatic vectorization and interpretation of map-drawings.	Sergey Ablameyko,Vladimir Bereishik,Oleg Frantskevich,Maria Homenko,E. Melnik,Oleg Okun,Nadeshda Paramonova	1993	A system for automatic vectorization and interpretation of map-drawings.
ICCV	Learning object recognition models from images.	Arthur R. Pope,David G. Lowe	1993	Learning object recognition models from images.
ICCV	Computation of ego-motion and structure from visual and inertial sensors using the vertical cue.	Thierry Viéville,Emmanuelle Clergue,P. E. Dos Santos Facao	1993	Computation of ego-motion and structure from visual and inertial sensors using the vertical cue.
ICCV	Projective depth: A geometric invariant for 3D reconstruction from two perspective/orthographic views and for visual recognition.	Amnon Shashua	1993	Projective depth: A geometric invariant for 3D reconstruction from two perspective/orthographic views and for visual recognition.
ICCV	Exploiting the generic view assumption to estimate scene parameters.	William T. Freeman	1993	Exploiting the generic view assumption to estimate scene parameters.
ICCV	Projectively invariant decomposition and recognition of planar shapes.	Stefan Carlsson	1993	Projectively invariant decomposition and recognition of planar shapes.
ICCV	Egomotion analysis based on the Frenet-Serret motion model.	Zoran Duric,Azriel Rosenfeld,Larry S. Davis	1993	Egomotion analysis based on the Frenet-Serret motion model.
ICCV	Localization using combinations of model views.	Ronen Basri,Ehud Rivlin	1993	Localization using combinations of model views.
ICCV	Reactions to peripheral image motion using a head/eye platform.	David W. Murray,Philip F. McLauchlan,Ian D. Reid,Paul M. Sharkey	1993	Reactions to peripheral image motion using a head/eye platform.
ICCV	Relative 3D positioning and 3D convex hull computation from a weakly calibrated stereo pair.	Luc Robert,Olivier D. Faugeras	1993	Relative 3D positioning and 3D convex hull computation from a weakly calibrated stereo pair.
ICCV	Looking for trouble: Using causal semantics to direct focus of attention.	Lawrence Birnbaum,Matthew Brand,Paul R. Cooper	1993	Looking for trouble: Using causal semantics to direct focus of attention.
ICCV	Ridge-detection for the perceptual organization without edges.	J. Brian Subirana-Vilanova,Kah Kay Sung	1993	Ridge-detection for the perceptual organization without edges.
ICCV	Affine-invariant contour tracking with automatic control of spatiotemporal scale.	Andrew Blake,Rupert W. Curwen,Andrew Zisserman	1993	Affine-invariant contour tracking with automatic control of spatiotemporal scale.
ICCV	Vision-based construction of CAD models from range images.	Xin Chen,Francis Schmitt	1993	Vision-based construction of CAD models from range images.
ICCV	Robust line-based pose enumeration from a single image.	Takeshi Shakunaga	1993	Robust line-based pose enumeration from a single image.
ICCV	Interpretation of natural scenes using multi-parameter default models and qualitative constraints.	Michael Hild,Yoshiaki Shirai	1993	Interpretation of natural scenes using multi-parameter default models and qualitative constraints.
ICCV	Shape from texture from a multi-scale perspective.	Tony Lindeberg,Jonas Gårding	1993	Shape from texture from a multi-scale perspective.
ICCV	Recognizing mice, vegetables and hand printed characters based on implicit polynomials, invariants and Bayesian methods.	Jayashree Subrahmonia,Daniel Keren,David B. Cooper	1993	Recognizing mice, vegetables and hand printed characters based on implicit polynomials, invariants and Bayesian methods.
ICCV	Incremental image sequence enhancement with implicit motion compensation.	A. Singh	1993	Incremental image sequence enhancement with implicit motion compensation.
ICCV	Robust computation of optical flow in a multi-scale differential framework.	Joseph Weber,Jitendra Malik	1993	Robust computation of optical flow in a multi-scale differential framework.
ICCV	Distance accumulation and planar curvature.	Joon Hee Han,Timothy Poston	1993	Distance accumulation and planar curvature.
ICCV	Understanding noise: The critical role of motion error in scene reconstruction.	J. Inigo Thomas,Allen R. Hanson,John Oliensis	1993	Understanding noise: The critical role of motion error in scene reconstruction.
ICCV	Building and using flexible models incorporating grey-level information.	Timothy F. Cootes,Christopher J. Taylor,Andreas Lanitis,David H. Cooper,Jim Graham	1993	Building and using flexible models incorporating grey-level information.
ICCV	Using hyperquadrics for shape recovery from range data.	Song Han,Dmitry B. Goldgof,Kevin W. Bowyer	1993	Using hyperquadrics for shape recovery from range data.
ICCV	Recognition of object classes from range data.	Ian D. Reid,J. Michael Brady	1993	Recognition of object classes from range data.
ICCV	Silhouette-based object recognition through curvature scale space.	Farzin Mokhtarian,Hiroshi Murase	1993	Silhouette-based object recognition through curvature scale space.
ICCV	An all-transputer visual autobahn-autopilot/copilot.	Ernst D. Dickmanns,Reinhold Behringer,C. Brudigam,Dirk Dickmanns,Frank Thomanek,Volker van Holt	1993	An all-transputer visual autobahn-autopilot/copilot.
ICCV	A global algorithm for shape from shading.	John Oliensis,Paul Dupuis	1993	A global algorithm for shape from shading.
ICCV	Fast segmentation, tracking, and analysis of deformable objects.	Chahab Nastar,Nicholas Ayache	1993	Fast segmentation, tracking, and analysis of deformable objects.
ICCV	Quantitative analysis of the viewpoint consistency constraint in model-based vision.	L. Du,Geoffrey D. Sullivan,Keith D. Baker	1993	Quantitative analysis of the viewpoint consistency constraint in model-based vision.
ICCV	Active and intelligent sensing of road obstacles: Application to the European Eureka-PROMETHEUS project.	Ming Xie,Laurent Trassoudaine,Joseph Alizon,Monique Thonnat,Jean Gallice	1993	Active and intelligent sensing of road obstacles: Application to the European Eureka-PROMETHEUS project.
ICCV	Linear and incremental acquisition of invariant shape models from image sequences.	Daphna Weinshall,Carlo Tomasi	1993	Linear and incremental acquisition of invariant shape models from image sequences.
ICCV	Discrete models for energy-minimizing segmentation.	Andrew Ackah-Miezan,André Gagalowicz	1993	Discrete models for energy-minimizing segmentation.
ICCV	A complete two-plane camera calibration method and experimental comparisons.	Guo-Qing Wei,Song De Ma	1993	A complete two-plane camera calibration method and experimental comparisons.
ICCV	A design for a visual motion transducer.	Andrew Blake,Gabriel Hamid,Lionel Tarassenko	1993	A design for a visual motion transducer.
ICCV	Diagonal transforms suffice for color constancy.	Graham D. Finlayson,Mark S. Drew,Brian V. Funt	1993	Diagonal transforms suffice for color constancy.
ICCV	Tracking foveated corner clusters using affine structure.	Ian D. Reid,David W. Murray	1993	Tracking foveated corner clusters using affine structure.
ICCV	Optical flow from 1D correlation: Application to a simple time-to-crash detector.	Nicola Ancona,Tomaso Poggio	1993	Optical flow from 1D correlation: Application to a simple time-to-crash detector.
ICCV	Fourth International Conference on Computer Vision, ICCV 1993, Berlin, Germany, 11-14 May, 1993, Proceedings		1993	Fourth International Conference on Computer Vision, ICCV 1993, Berlin, Germany, 11-14 May, 1993, Proceedings
ICCV	A robust active contour model with insensitive parameters.	Gang Xu,Eigo Segawa,Saburo Tsuji	1993	A robust active contour model with insensitive parameters.
ICCV	"An extension of Marr's signature based edge classification and other methods determining diffuseness and height of edges, and bar edge width."	Wei Zhang,Fredrik Bergholm	1993	"An extension of Marr's signature based edge classification and other methods determining diffuseness and height of edges, and bar edge width."
ICCV	Eliciting qualitative structure from image curve deformations.	Andrew Zisserman,Andrew Blake,Charlie Rothwell,Luc J. Van Gool,Marc Van Diest	1993	Eliciting qualitative structure from image curve deformations.
ICCV	Contextual feature similarities for model-based object recognition.	Detlev Noll,Michael Schwarzinger,Werner von Seelen	1993	Contextual feature similarities for model-based object recognition.
ICCV	A generalized brightness change model for computing optical flow.	Shahriar Negahdaripour,Chih-Ho Yu	1993	A generalized brightness change model for computing optical flow.
ICCV	Dynamic calibration of an active stereo head.	James L. Crowley,Philippe Bobet,Cordelia Schmid	1993	Dynamic calibration of an active stereo head.
ICCV	Accurate line parameters from an optimising Hough Transform for vanishing point detection.	Phil L. Palmer,Maria Petrou,Josef Kittler	1993	Accurate line parameters from an optimising Hough Transform for vanishing point detection.
ICCV	A spherical representation for the recognition of curved objects.	Herve Delingette,Martial Hebert,Katsushi Ikeuchi	1993	A spherical representation for the recognition of curved objects.
ICCV	Fast and robust 3D recognition by alignment.	Tao Daniel Alter,W. Eric L. Grimson	1993	Fast and robust 3D recognition by alignment.
IEEE Visualization	A Climate Simulation Case Study.	P. C. Chen	1993	A Climate Simulation Case Study.
IEEE Visualization	The Quantum Coulomb Three Body Problem - Visualization of Simulation Results and Numerical Methods.	D. I. Abramov,V. V. Gusev,Stanislav V. Klimenko,L. I. Ponomarev,W. Krueger,W. Renz	1993	Some years ago it was established that the muon catalysed fusion phenomenon could be used for the production of energy. This fact has been causing a rebirth of interest in the universal methods of solving the Quantum Coulomb Three-Body Problem. The Adiabatic Hyperspherical (AHS) approach considered in this joint project has definite advantages in comparison with other methods.The case study proposed will focus on the study of the structure and behavior of the wave function of bound states of quantum three-body system as well as of the basis functions of the AHS approach. Adapted scientific visualization tools such as surface rendering, volume raytracing and texturing will be used. Visualization allows to discover interesting features in the behavior of the basis functions and to analyze the convergence of the AHS-expansion for the wave functions.
IEEE Visualization	Visualization of Acoustic Lens Data.	A. J. Bladek	1993	"3-dimensional data visualization from any input source involves the study and understanding of several steps. These steps include data acquisition. Signal processing, image processing and image generation. Using a forward-looking high frequency sonar system (which focuses sound much like the eye focuses light), standard and non-standard data processing algorithms, and industry ""standard"" visualization algorithms, this project produced accurate 3-dimensional representations of several underwater objects."
IEEE Visualization	3D Simulation of Delivery.	Jean-Daniel Boissonnat,Bernhard Geiger	1993	3D Simulation of Delivery.
IEEE Visualization	Orientation Maps: Techniques for Visualizing Rotations.	Bowen Alpern,Larry Carter,M. Grayson,C. Pelkie	1993	Orientation Maps: Techniques for Visualizing Rotations.
IEEE Visualization	GRASPARC: A Problem Solving Environment Integrating Computation and Visualization.	Ken Brodlie,A. Poon,Helen Wright,L. Brankin,G. Banecki,A. Gay	1993	Visualization has proved an effective tool in the understanding of large data sets in computational science and engineering. There is growing interest today in the development of problem solving environments which integrate both visualization and the computational process which generates the data. The GRASPARC project has looked at some of the issues involved in creating such an environment. An architecture is proposed in which tools for computation and visualization can be embedded in a framework which assists in the management of the problem solving process. This framework has an integral data management facility which allows an audit trail of the experiments to be recorded. This design therefore allows not only steering but also backtracking and more complicated problem solving strategies.A number of demonstrator case studies have been implemented.
IEEE Visualization	Optimal Filter Design for Volume Reconstruction and Visualization.	Ingrid Carlbom	1993	Digital filtering is a crucial operation in volume reconstruction and visualization. Lowpass filters are needed for subsampling and minification. Interpolation filters are needed for registration and magnification, and to compensate for geometric distortions introduced by scanners. Interpolation filters are also needed in volume rendering for ray-casting and slicing. In this paper, we describe a method for digital filter design of interpolation filters based on weighted Chebyshev minimization. The accuracy of the resulting filters are compared with some commonly used filters defined by piecewise cubic polynomials.A significant finding of this paper is that although piecewise cubic interpolation have some computational advantages and may yield visually satisfactory results for some data, other data results in artifacts such as blurring. Furthermore, piecewise cubic filters are inferior for operations such as registration. Better results are obtained by the filters derived in this papers at only small increases in computation.
IEEE Visualization	Visualization and Modeling of Geophysical Data.	G. Celniker,I. Chakravarty,J. Moorman	1993	We present the visualization and modeling techniques used in a case study to build feature-based computational models from geophysical data. Visualization was used to inspect the quality of the interpretation of the geophysical data. We describe the geophysical data graphical representation used to support rapid rendering and to enhance the perception of differences between the interpretation of the data and the data itself. In addition, we present the modeling techniques used to convert the geophysical data into a feature-based computational model suitable for use by a numerical simulation package.
IEEE Visualization	Data Shaders.	Brian Corrie,Paul Mackerras	1993	The process of visualizing a scientific data set requires an extensive knowledge of the domain in which the data set is created. Because an in-depth knowledge of all scientific domains is not available to the creator of visualization software, a flexible and extensible visualization system is essential in providing a productive tool to the scientist.This paper presents a shading language, based on the RenderMan shading language, that extends the shading model used to render volume data sets. Data shaders, written in this shading language, give the users of a volume rendering system a means of specifying how a volume data set is to be rendered. This flexibility is useful both as a visualization tool in the scientific community and as a research tool in the visualization community.
IEEE Visualization	Texture Splats for 3D Scalar and Field Visualization.	Roger Crawfis,Nelson L. Max	1993	Texture Splats for 3D Scalar and Field Visualization.
IEEE Visualization	Interactive Shading for Surface and Volume Visualization on Graphics Workstations.	Peter A. Fletcher,Philip K. Robertson	1993	Shading is an effective exploratory visualization tool widely used in scientific visualization. Interactive, or close to interactive, shading of images offers significant benefit, but is generally too computationally expensive for graphics workstations. A novel method for providing interactive diffuse and specular shading capability on low-cost graphics workstations is described. Application to digital elevation models, iso-surfaces in volumetric images, and colour-coded aspect maps are illustrated and an analysis of artifacts, and of ways of minimising artifacts, is given.
IEEE Visualization	The Vision Camera: An Interactive Tool for Volume Data Exploration and Navigation.	Hans-Heino Ehricke,G. Daiber,Wolfgang Straßer	1993	In this paper we focus on one of the key problems of Scientific Visualization, the object recognition dilemma. The necessity to pre-interpret application data in order to classify object surface voxels prior to rendering has prevented many visualization methods from becoming practical. We propose the concept of vision by visualization which integrates Computer Vision methods into the visualization process. Based on this, we present the vision camera, a new tool allowing for interactive object recognition during volume data walkthroughs. This camera model is characterized by a flexible front-plane which, under the control of user-specified parameters and image features elastically matches to object surfaces, while shifted through a data volume. Thus, objects are interactively carved out and can be visualized by standard volume visualization methods. Implementation and application of the model are described. Our results suggest that by the integration of human and machine vision new perspectives for data exploration are opened up.
IEEE Visualization	Non Conventional Methods for the Visualization of Events from High Energy Physics.	H. Drevermann,D. Kuhm,B. Nilsson	1993	"Visualization of events in high energy physics is an important tool to check hard- and software and to generate pictures for presentation purposes. The radial pattern of all events suggests the use of predefined projections, especially p1/Z and Y/X. The representation can be improved by a ""fish-eye"" transformation and by angular projections, which produce straight track patterns and allow extensive magnifications. Three dimensional data of radial structure are best displayed in the 3D V-Plot, which has optimal track separation and presents all relevant information in a clear way."
IEEE Visualization	Navigating Large Networks with Hierarchies.	Stephen G. Eick,Graham J. Wills	1993	This paper is aimed at the exploratory visualization of networks where there is a strength or weight associated with each link, and makes use of any hierarchy present on the nodes to aid the investigation of large networks. It describes a method of placing nodes on the plane that gives meaning to their relative positions. The paper discusses how linking and interaction principles aid the user in the exploration. Two examples are given; one of electronic mail communication over eight months within a department, another concerned with changes to a large section of a computer program.
IEEE Visualization	Virtual Input Devices for 3D Systems.	Taosong He,Arie E. Kaufman	1993	"The device unified interface is a generalized and easily expandable protocol for the communication between applications and input devices. The key idea is to unify various device data into the parameters of a so-called ""virtual input device."" The device information-base, which includes device dependent information, is also incorporated into the virtual input device. Using the device unified interface, system builders are able to design their applications independent of the input devices as well as utilize the capabilities of several devices in the same application."
IEEE Visualization	Rapid Exploration of Curvilinear Grids Using Direct Volume Rendering.	Allen Van Gelder,Jane Wilhelms	1993	Fast techniques for direct volume rendering over curvilinear grids (common to computational fluid dynamics and finite element analysis) are developed. Three new projection methods that use polygon-rendering hardware for speed are presented and compared with each other and with previous methods for tetrahedral grids and rectilinear grids. A simplified algorithm for visibility ordering, based on a combination of breadth-first and depth-first searches, is described. A new multi-pass blending method is described that reduces visual artifacts that are introduced by linear interpolation in hardware where exponential interpolation is needed. Visualization tools that permit rapid data banding and cycling through transfer functions, as well as region restriction, are described.
IEEE Visualization	Interactive Visualization Methods for Four Dimensions.	Andrew J. Hanson,Robert A. Cross	1993	Making accurate computer graphics representations of surfaces and volumes (2-manifolds and 3-manifolds) embedded in four-dimensional space typically involves complex and time-consuming computations. In order to make simulated worlds that help develop human intuition about the fourth dimension, we need techniques that permit real-time, interactive manipulation of the most sophisticated depictions available. We propose the following new methods that bring us significantly closer to this goal: an approach to high-speed 4D illuminated surface rendering incorporating 4D shading and occlusion coding; a procedure for rapidly generating 2D screen images of tessellated 3-manifolds illuminated by 4D light. These methods are orders of magnitude faster than previous approaches, enabling the real-time manipulation of high-resolution 4D images on commercial graphics hardware.
IEEE Visualization	Visualization of Turbulent Flow with Particles.	Andrea J. S. Hin,Frits H. Post	1993	In this paper a new method for visualization of three-dimensional turbulent flow using particle motion animation is presented. The method is based on Reynolds decomposition of a turbulent flow field into a convective and a turbulent motion. At each step of particle path generation a stochastic perturbation is added, resulting in random-walk motions of particles. A physical relation is established between the perturbations and the eddy-diffusivity, which is calculated in a turbulent flow simulation. The flow data used is a mean velocity field, and an eddy-diffusivity field. The erratic particle motions are more than just a visual effect, but represent a real physical phenomenon. An implementation of the method is described, and an example of a turbulent channel flow is given, which clearly shows the random particle motions in their context of general fluid motion patterns.
IEEE Visualization	Geometric Optimization.	P. A. Hinker,Charles D. Hansen	1993	An algorithm is presented which describes an application independent method for reducing the number of polygonal primitives required to faithfully represent an object. Reducing polygon count without a corresponding reduction in object detail is important for: achieving interactive frame rates in scientific visualization, reducing mass storage requirements, and facilitating the transmission of large, multi-timestep geometric data sets. This paper shows how coplanar and nearly coplanar polygons can be merged into larger complex polygons and re-triangulated into fewer simple polygons than originally required. The notable contributions of this paper are: 1) a method for quickly grouping polygons into nearly coplanar sets, 2) a fast approach for merging coplanar polygon sets and, 3) a simple, robust triangulation method for polygons created by 1 and 2. The central idea of the algorithm is the notion of treating polygonal data as a collection of segments and removing redundant segments to quickly from polygon hulls which represent the merged coplanar sets.
IEEE Visualization	Visual Feedback in Querying Large Databases.	Daniel A. Keim,Hans-Peter Kriegel,Thomas Seidl	1993	In this paper, we describe a database query system that provides visual relevance feedback in querying large databases. The goal of our system is to support the query specification process by using each pixel of the display to represent one data item of the database. By arranging and coloring the pixels according to their relevance for the query, the user gets a visual impression of the resulting data set. Using sliders for each condition of the query, the user may change the query dynamically and receives immediate feedback by the visual representation of the resulting data set. By using multiple windows for different parts of a complex query, the user gets visual feedback for each part of the query and, therefore, will easier understand the overall result. The system may be used to query any database that contains tens of thousands to millions of data items, but it is especially helpful to explore large data sets with an unknown distribution of values and to find the interesting hot spots in huge amounts of data. The direct feedback allows to visually display the influence of incremental query refinements and, therefore, allows a better, easier and faster query specification.
IEEE Visualization	Cloud Tracing in Conection-Diffusion Systems.	K. L. Ma,P. J. Smith	1993	This paper describes a highly interactive method for computer visualization of simultaneous three-dimensional vector and scalar flow fields in convection-diffusion systems. This method allows a computational fluid dynamics user to visualize the basic physical process of dispersion and mixing rather than just the vector and scalar values computed by the simulation. It is based on transforming the vector field from a traditionally Eulerian reference frame into a Lagrangian reference frame. Fluid elements are traced through the vector field for the mean path as well as the statistical dispersion of the fluid elements about the mean position by using added scalar information about the root mean square value of the vector field and its Lagrangian time scale. In this way, clouds of fluid elements are traced not just mean paths. We have used this method to visualize the simulation of an industrial incinerator to help identify mechanisms for poor mixing.
IEEE Visualization	Visualization of Time-Dependent Flow Fields.	David A. Lane	1993	Presently, there are very few visualization systems available for time-dependent flow fields. Although existing visualization systems for instantaneous flow fields may be used to view time-dependent flow fields at discrete points in time, the time variable is usually not considered in the visualization technique. We present a simple and effective approach for visualizing timedependent flow fields using streaklines. A system was developed to demonstrate this approach. The system can process many time frames of flow fields without requiring that all the data be in memory simultaneously, and it also handles flow fields with moving grids. We have used the system to visualize streaklines from several large 3D time-dependent flow fields with moving grids. The system was able to provide useful insights to the physical phenomena in the flow fields.
IEEE Visualization	Towards Interactive Steering, Visualization and Animation of Unsteady Finite Element Simulations.	G. David Kerlick,E. Kirby	1993	"Progress towards interactive steering of the time-accurate, unsteady finite-element simulation program DYNA3D is reported. Rudimentary steering has been demonstrated in a distributed computational environment encompassing a supercomputer, multiple graphics workstations, and a single frame animation recorder. The coroutine facility of AVS (Application Visualization System from AVS Inc.) and software produced in-house has been coordinated to prove the concept. This work also applies to other large batch-oriented FORTRAN simulations (""dusty decks"") presently in production use."
IEEE Visualization	An Environment for Telecollaborative Data Exploration.	Gudrun Klinker	1993	This paper presents an environment for telecollaborative data exploration. It provides the following capabilities essential to data exploration: (1) Users can probe the data, defining regions of interest with arbitrary shapes. (2) The selected data can be transformed and displayed in many different ways. (3) Linked cursors can be established between several windows showing data sets with arbitrary relationships. (4) Data can be displayed on any screen across a computer network, allowing for telecollaboration arrangements with linked cursors around the world. (5) Our system is user-extensible, allowing programmers to change any component of it while keeping the remaining functionality. We demonstrate how the system can be used in several applications, such as biomedical imaging, robotics, and wood classification.
IEEE Visualization	A Probe for Local Flow Field Visualization.	Wim C. de Leeuw,Jarke J. van Wijk	1993	A probe for the interactive visualization of flow fields is presented. The probe can be used to visualize many characteristics of the flow in detail for a small region in the data set. The velocity and the local change of velocity (the velocity gradient tensor) are visualized by a set of geometric primitives. To this end, the velocity gradient tensor is transformed to a local coordinate frame, and decomposed into components parallel with and perpendicular to the flow. These components are visualized as geometric objects with an intuitively meaningful interpretation. An implementation is presented which shows that this probe is a useful tool for flow visualization.
IEEE Visualization	Bridging the Gap Between Visualization and Data Management: A Simple Visualization Management System.	Peter Kochevar,Zahid Ahmed,J. Shade,Colin Sharp	1993	A prototype visualization management system is described which merges the capabilities of a database management system with any number of existing visualization packages such as AVS or IDL. The prototype uses the Postgres database management system to store and access Earth science data through a simple graphical browser. Data located in the database is visualized by automatically invoking a desired visualization package and downloading an appropriate script or program. The central idea underlying the system is that information on how to visualize a data set is stored in the database with the data set itself.
IEEE Visualization	Enhancing Reality in the Operating Room.	William E. Lorensen,Harvey E. Cline,C. Nafis,D. Altobelli,L. Gleason	1993	Enhancing Reality in the Operating Room.
IEEE Visualization	Geometric Clipping Using Boolean Textures.	W. E. Lorenson	1993	"Texture mapping is normally used to convey geometric detail without adding geometric complexity. This paper introduces boolean textures, a texture mapping technique that uses implicit functions to generate texture maps and texture coordinates. These boolean textures perform clipping during a renderer's scan conversion step. Any implicit function is a candidate boolean texture clipper. The paper describes how to use quadrics as clippers. Applications from engineering and medicine illustrate the effectiveness of texture as a clipping tool."
IEEE Visualization	DIVIDE: Distributed Visual Display of the Execution of Asynchronous, Distributed Algorithms on Loosely-Coupled Parallel Processors.	Tom M. Morrow,Sumit Ghosh	1993	"The issue of monitoring the execution of asynchronous, distributed algorithms on loosely-coupled parallel processor systems, is important for the purposes of (i) detecting inconsistencies and flaws in the algorithm, (ii) obtaining important performance parameters for the algorithm, and (iii) developing a conceptual understanding of the algorithm's behavior, for given input stimulus, through visualization. For a particular class of asynchronous distributed algorithms [1,5] that may be characterized by independent and concurrent entities that execute asynchronously on multiple processors and interact with one another through explicit messages, the following reasoning applies. Information about the flow of messages and the activity of the processors may contribute significantly towards the conceptual understanding of the algorithm's behavior and the functional correctness of the implementation. The computation and subsequent display of important parameters, based upon the execution of the algorithm, is an important objective of DIVIDE. For instance, the mean and standard deviation values for the propagation delay of ATM cells between any two given Broadband-ISDN (BISDN) nodes in a simulation of BISDN network under stochastic input stimulus[36], as a function of time, are important clues to the degree of congestion in the Broadband-ISDN network. Although the execution of the algorithm typically generates high resolution data, often, a coarse-level visual representation of the data may be useful in facilitating the conceptual understanding of the behavior of the algorithm. DIVIDE permits a user to specify a resolution less than that of the data from the execution of the algorithm, which is then utilized to coalesce the data appropriately. Given that this process requires significant computational power, for efficiency, DIVIDE distributes the overall task of visual display into a number of user specified workstations that are configured as a loosely-coupled parallel processor. DIVIDE has been implemented on a heterogeneous network of SUN sparc 1+, sparc 2, and 3/60 workstations and performance measurements indicate significant improvement over that of a uniprocessor-based visual display."
IEEE Visualization	Flow Volumes for Interactive Vector Field Visualization.	Nelson L. Max,Barry G. Becker,Roger Crawfis	1993	Flow volumes are the volumetric equivalent of stream lines. They provide more information about the vector field being visualized than do stream lines or ribbons. Presented is an efficient method for producing flow volumes, composed of transparently rendered tetrahedra, for use in an interactive system. The problems of rendering, subdivision, sorting, compositing artifacts, and user interaction are dealt with. Efficiency comes from rendering only the volume of the smoke, and using hardware texturing and compositing.
IEEE Visualization	Visualizing Results of Transient Flow Simulations.	H. F. Mayer,B. Tabatabai	1993	Visualizing Results of Transient Flow Simulations.
IEEE Visualization	Feature Extraction for Oceanographic Data Using a 3D Edge Operator.	R. J. Moorehead,Z. Zhu	1993	Feature Extraction for Oceanographic Data Using a 3D Edge Operator.
IEEE Visualization	Fast Volume Rendering of Compressed Data.	Paul Ning,Lambertus Hesselink	1993	Volume rendering has been proposed as a useful tool for extracting information from large datasets, where non-visual analysis alone may not be feasible. The scale of these applications implies that data management is an important issue that needs to be addressed. Most volume rendering algorithms, however, process data in raw, uncompressed form. In previous work, we introduced a compressed volume format that may be volume rendered directly with minimal impact on rendering time. In this paper, we extend these ideas to a new volume format that not only reduces storage space and transmission time, but is designed for fast volume rendering as well. The volume dataset is represented as indices into a small codebook of representative blocks. With this data structure, volume shading calculations need only be performed on the codebook and image generation is accelerated by reusing precomputed block projections.
IEEE Visualization	Unsteady Phenomena, Hypersonic Flows and Co-Operative Flow Visualization.	Hans-Georg Pagendarm	1993	Unsteady Phenomena, Hypersonic Flows and Co-Operative Flow Visualization.
IEEE Visualization	The Virtual Restoration of the Visir Tomb.	Patrizia Palamidese,M. Betro,G. Muccioli	1993	The Virtual Restoration of the Visir Tomb.
IEEE Visualization	Spray Rendering: Visualization Using Smart Particles.	Alex Pang,Kyle Smith	1993	We propose a new framework for doing scientific visualization. The basis for this framework is a combination of particle systems and behavioral animation. Here, particles are not only affected by the field that they are in, but can also exhibit different programmed behaviors. An intuitive delivery system, based on virtual cans of spray paint, is also described to introduce the smart particles into the data set. Hence the name spray rendering. Using this metaphor, different types of spray paint are used to highlight different features in the data set. Spray rendering offers several advantages over existing methods: (1) it generalizes the current techniques of surface, volume and flow visualization under one coherent framework; (2) it works with regular and irregular grids as well as sparse and dense data sets; (3) it allows selective progressive refinement; (4) it is modular, extensible and provides scientists with the flexibility for exploring relationships in their data sets in natural and artistic ways.
IEEE Visualization	Dichromatic Color Representations for Complex Display Systems.	M. S. Peercy,Lambertus Hesselink	1993	New display technologies have begun to provide more innovative and potentially powerful methods to present information to a viewer. However, many of these techniques struggle to deliver accurate full color. In this paper, we address this difficulty by employing the dichromatic theory of color reflection, which implies that many objects can be rendered accurately using only two primaries. Complex display systems with two primaries can be produced with significantly less work than is required for the traditional three primaries. We discuss methods for selecting objects that can be rendered accurately on two-color displays, and we present our experiments with a two-color display using monochromatic primaries.
IEEE Visualization	MRIVIEW: An Interactive Computational Tool for Investigation of Brain Structure and Function.	D. Ranken,J. George	1993	MRIVIEW is a software system that uses image processing and visualization to provide neuroscience researchers with an integrated environment for combining functional and anatomical information. Key features of the software include semi-automated segmentation of volumetric head data and an interactive coordinate reconciliation method which utilizes surface visualization.The current system is a precursor to a computational brain atlas. We describe features this atlas will incorporate, including methods under development for visualizing brain functional data obtained from several different research modalities.
IEEE Visualization	Towards a Texture Naming System: Identifying Relevant Dimensions in Texture.	A. Ravishankar Rao,Gerald L. Lohse	1993	Towards a Texture Naming System: Identifying Relevant Dimensions in Texture.
IEEE Visualization	An Architecute for Rule-Based Visualization.	Bernice E. Rogowitz,Lloyd Treinish	1993	In Rogowitz and Treinish [20] we introduced an architecture for incorporating perceptual rules into the visualization process. In this architecture, higher-level descriptors of the data, metadata, flow to perceptual rules, which constrain visualization operations. In this paper, we develop a deeper analysis of the rules, the prerequisite metadata, and the system for enabling their operation.
IEEE Visualization	Fine-Grain Visualization Algorithms in Dataflow Environments.	D. Song,Eric J. Golin	1993	Most of the current dataflow visualization systems are based on coarse-grain dataflow computing models. In this paper we propose a fine-grain dataflow model that takes advantage of data locality properties of many visualization algorithms. A fine-grain module works on small chunks of data one at a time by keeping a dynamically adjusted moving window on the input data stream. It is more memory efficient and has the potential of handling very large data sets without taking up all the memory resources. Two popular visualization algorithms, an iso-surface extraction algorithm [1] and a volume rendering algorithm [2], are implemented using the fine-grain model. The performance measurements showed faster speed, reduced memory usage, and improved CPU utilization over a typical coarse-grain system.
IEEE Visualization	InfoCrystal: A Visual Tool for Information Retrieval.	Anselm Spoerri	1993	This paper introduces a novel representation, called the Info Crystal&trade;, that can be used as a visualization tool as well as a visual query language to help users search for information. The InfoCrystal visualizes all the possible relationships among N concepts. Users can assign relevance weights to the concepts and use thresholding to select relationships of interest. The InfoCrystal allows users to specify Boolean as well as vectorspace queries graphically. Arbitrarily complex queries can be created by using the InfoCrystals as building blocks and organizing them in a hierarchical structure. The InfoCrystal enables users to explore and filter information in a flexible, dynamic and interactive way.
IEEE Visualization	Applying Observations of Work Activity in Designing Prototype Data Analysis Tools.	R. R. Springmeyer	1993	"Designers, implementers, and marketers of data analysis tools typically have different perspectives than end users. Consequently, data analysts often find themselves using tools focused on graphics and programming concepts rather than concepts which reflect their own domain and the context of their work. Some user studies focus on usability tests late in development; others observe work activity, but fail to show how to apply that knowledge in design. This paper describes a methodology for applying observations of data analysis work activity in prototype tool design. The approach can be used both in designing improved data analysis tools, and customizing visualization environments to specific applications. We present an example of user-centered design for a prototype tool to cull large data sets. We revisit the typical graphical approach of animating a large data set from the point of view of an analyst who is culling data. Field evaluations using the prototype tool not only revealed valuable usability information, but initiated in-depth discussions about users' work, tools, technology, and requirements."
IEEE Visualization	Tioga: A Database-Oriented Visualization Tool.	Michael Stonebraker,Jolly Chen,Nobuko Nathan,Caroline Paxson,Alan Su,Jiang Wu	1993	"In this paper we present a new architecture for visualization systems that is based on Data Base Management System (DBMS) technology. By building on the mechanisms present in a next-generation DBMS, rather than merely on the capabilities of a standard file manager, we show that a simpler and more powerful visualization system can be constructed. We retain the popular ""boxes and arrows"" programming notation for constructing visualization programs, but add a ""flight simulator"" model of movement to navigate the output of such programs. In addition, we provide a means to specify a hierarchy of abstracts of data of different types and resolutions, so that a ""zoom"" capability can be supported. The underlying DBMS support for this system, Tioga, is briefly described, as well as the current state of the implementation."
IEEE Visualization	Developing Modular Applications Builders to Exploit MIMD Parallel Resources.	C. Thornborrow,A. J. S. Wilson,Chris Faigle	1993	Modular application builders (MABs), such as AVS and Iris Explorer[6, 7] are increasingly being used in the visualisation community. Such systems can already place compute intensive modules on supercomputers in order to utilise their power. This paper details two major projects at EPCC which attempted to fully integrate the MAB concept with a distributed memory MIMD (DM-MIMD) environment.The work presented was driven by two goals, efficient use of the resource and ease of use by programmer and end user.We present a model of MABs and describe the major problems faced, giving solutions to them through two case studies.
IEEE Visualization	Visualization of Stratospheric Ozone Depletion and the Polar Vortex.	Lloyd Treinish	1993	Visualization of Stratospheric Ozone Depletion and the Polar Vortex.
IEEE Visualization	Visualization of Oil Reservoirs Ovew a Large Range of Scales as a Catalyst for Multi-Disciplinary Integration.	S. Tyson,B. Williams	1993	We discuss a system which provides a single, unified model of oil and gas reservoirs that is used across a range of disciplines from geologists to reservoir engineers. It has to store, manipulate and display reservoir phenomena which are observed over several orders of magnitude from 1mm to 10km.We propose that the current capabilities of visualization, over this range of scales, can remove perception barriers that have existed between disciplines and provide clear insights into the problems of modelling reservoirs from geological and engineering perspectives.
IEEE Visualization	"Fast Analytical Computation of Richard's Smooth Molecular Surface."	Amitabh Varshney,Frederick P. Brooks Jr.	1993	"Fast Analytical Computation of Richard's Smooth Molecular Surface."
IEEE Visualization	Fanel: A Relational Analysis and Visualization Package for High Energy Physics.	H. Videau,P. Mora de Freitas	1993	The package described in this paper has been designed for analysing the data collected in the LEP experiment ALEPH. Its main graphical feature is a deep interplay between the description of the objects manipulated and their relationships, and their graphical representation. The easy access to information through navigation between objects and its display makes possible a thorough study of the events produced by the detector. This has proved to be very powerful in numerous occasions for analysing data and testing programs. The package provides as well statistical analysis tools and a graphic editor. It is based on the PHIGS graphics standard. It will develop towards a more elaborate usage of the data structure in particular in the geometrical representations and towards object oriented languages to overcome some heaviness linked to the use of Fortran.
IEEE Visualization	Performance Visualization of Parallel Programs.	Abdul Waheed,Diane T. Rover	1993	The user of a parallel computer system would like to know the performance of a program in terms of how optimally it uses the system resources. This task is increasingly performed by program performance visualization. The limitations of conventional performance data analysis techniques necessitate better visual analysis methods that are scalable with the problem and system sizes and extensible. They should represent some physical and logical structure of the parallel system and program. The analysis techniques presented here have been motivated by the use of signal and (two- and three-dimensional) image processing techniques being applied in some areas of scientific visualization. Results of applying selected techniques are shown. These techniques and tools have advantages and disadvantages when applied in this area.
IEEE Visualization	Volume Sampled Voxelization of Geometric Primitives.	Sidney W. Wang,Arie E. Kaufman	1993	We present a 3D antialiasing algorithm for voxelbased geometric models. The technique band-limits the continuous object before sampling it at the desired 3D raster resolution. By precomputing tables of filter values for different types and sizes of geometric objects, the algorithm is very efficient and has a complexity that is linear with the number of voxels generated. The algorithm not only creates voxel models which are free from object space aliasing, but it also incorporates the image space antialiasing information as part of the view independent voxel model. The resulting alias-free voxel models have been used to model synthetic scenes, for discrete ray tracing applications. The discrete raytraced image is superior in quality to the image generated with a conventional surface-based ray tracer, since silhouettes of objects, shadows, and reflections appear smooth (jaggy-less). In addition, the alias-free models are also suitable for intermixing with sampled datasets, since they can be treated uniformly as one common data representation.
IEEE Visualization	Implicit Stream Surfaces.	Jarke J. van Wijk	1993	Streamlines and stream surfaces are well known techniques for the visualization of fluid flow. For steady velocity fields, a streamline is the trace of a particle, and a stream surface is the trace of a curve. Here a new method is presented for the construction of stream surfaces. The central concept is the representation of a stream surface as an implicit surface f (x)=C. After the initial calculation off a family of stream surfaces can be generated efficiently by varying C. The shapes of the originating curves as defined by the value off at the boundary. Two techniques are presented for the calculation of f: one based on solving the convection equation, the other on backward tracing of the trajectories of grid points. The flow around objects is discussed separately.With this method irregular topologies of the originating curves and of the stream surfaces can be handled easily. Further, it can also be used for other visualization techniques, such as time surfaces and stream volumes. Finally, an effective method for the automatic placement of originating curves is presented.
IEEE Visualization	HyperSlice - Visualization of Scalar Functions of Many Variables.	Jarke J. van Wijk,Robert van Liere	1993	HyperSlice is a new method for the visualization of scalar functions of many variables. With this method the multi-dimensional function is presented in a simple and easy to understand way in which all dimensions are treated identically. The central concept is the representation of a multi-dimensional function as a matrix of orthogonal two-dimensional slices. These two-dimensional slices lend themselves very well to interaction via direct manipulation, due to a one to one relation between screen space and variable space. Several interaction techniques, for navigation, the location of maxima, and the use of user-defined paths, are presented.
IEEE Visualization	Computer Visualization of Long Genomic Sequences.	Dachywan Wu,James Roberge,Douglas J. Cork,Bao Gia Nguyen,Thom Grace	1993	Human beings find it difficult to analyze local and global oligonucleotide patterns in the linear primary sequences of a genome. In this paper, we present a family of iterated function systems (IFS) that can be used to generate a set of visual models of a DNA sequence. A new visualization function, the W-curve, that is derived from this IFS family is introduced. Using W-curves, a user can readily compare subsequences within a long genomic sequence --- or between genomic sequences --- and can visually evaluate the effect of local variations (mutations) upon the global genomic information content.
IEEE Visualization	Accelerating Volume Animation by Space-Leaping.	Roni Yagel,Z. Shi	1993	In this paper we present a method for speeding the process of volume animation. It exploits coherency between consecutive images to shorten the path rays take through the volume. Rays are provided with the information needed to leap over the empty space and commence volume traversal at the vicinity of meaningful data. The algorithm starts by projecting the volume onto a C-buffer (Coordinates-buffer) which stores the object-space coordinates of the first non-empty voxel visible from a pixel. Following a change in the viewing parameters, the C-buffer is transformed accordingly. Next, coordinates that possibly became hidden are discarded. The remaining values serve as an estimate of the point where the new rays should start their volume traversal. This method does not require 3D preprocessing and does not suffer from any image degradation. It can be combined with existing acceleration techniques and can support any ray traversal algorithm and material modeling scheme.
IEEE Visualization	"Proceedings IEEE Visualization '93"	Gregory M. Nielson,R. Daniel Bergeron	1993	"Proceedings IEEE Visualization '93"
IEEE Visualization	Case Study: New Techniques in the Design of Healthcare Facilities.	Tarek K. Alameldin,Mardelle Shepley	1994	The recent advent of computer graphics techniques has helped to bridge the gap between architectural concepts and actual buildings. Closing this gap is especially critical in healthcare facilities. In this paper, we present new techniques to support the design decision process and apply them to the design of a neonatal intensive care unit. Two issues are addressed: ergonometric accessibility and visual supervision of spaces. These two issues can be investigated utilizing new technologies that demonstrate that computers are more than a medium of communication in the field of architecture; the computer can make a significant contribution as a proactive design tool.
IEEE Visualization	Interactive Visualization via 3D User Interfaces.	Andries van Dam	1994	Interactive Visualization via 3D User Interfaces.
IEEE Visualization	"Challenges and Opportunities in Visualization for NASA's EOS Mission to Planet Earth - Panel."	Mike E. Botts,Jon D. Dykstra,Lee S. Elson,Steven J. Goodman,Meemong Lee	1994	"Challenges and Opportunities in Visualization for NASA's EOS Mission to Planet Earth - Panel."
IEEE Visualization	Integrated Control of Distributed Volume Visualization Through the World-Wide-Web.	Cheong S. Ang,David C. Martin,Michael D. Doyle	1994	"The World-Wide-Web (WWW) has created a new paradigm for online information retrieval by providing immediate and ubiquitous access to digital information of any type from data repositories located throughout the world. The web's development enables not only effective access for the generic user, but also more efficient and timely information exchange among scientists and researchers. We have extended the capabilities of the web to include access to three-dimensional volume data sets with integrated control of a distributed client-server volume visualization system. This paper provides a brief background on the World-Wide-Web, an overview of the extensions necessary to support these new data types and a description of an implementation of this approach in a WWW-compliant distributed visualization system"
IEEE Visualization	VolVis: A Diversified Volume Visualization System.	Ricardo S. Avila,Taosong He,Lichan Hong,Arie E. Kaufman,Hanspeter Pfister,Cláudio T. Silva,Lisa M. Sobierajski,Sidney W. Wang	1994	VolVis is a diversified, easy to use, extensible, high performance, and portable volume visualization system for scientists and engineers as well as for visualization developers and researchers. VolVis accepts as input 3D scalar volumetric data as well as 3D volume-sampled and classical geometric models. Interaction with the data is controlled by a variety of 3D input devices in an input device-independent environment. VolVis output includes navigation preview, static images, and animation sequences. A variety of volume rendering algorithms are supported, ranging from fast rough approximations, to compression-domain rendering, to accurate volumetric ray tracing and radiosity, and irregular grid rendering.
IEEE Visualization	Streamball Techniques for Flow Visualization.	Manfred Brill,Hans Hagen,Hans-Christian Rodrian,Wladimir Djatschin,Stanislav V. Klimenko	1994	We introduce the concept of streamballs for flow visualization. Streamballs are based upon implicit surface generation techniques adopted from the well-known metaballs. Their property to split or merge automatically in areas of significant divergence or convergence makes them an ideal tool for the visualization of arbitrary complex flow fields. Using convolution surfaces generated by continuous skeletons for streamball construction offers the possibility to visualize even tensor fields.
IEEE Visualization	Triangulation and Display of Rational Parametric Surfaces.	Chandrajit L. Bajaj,Andrew V. Royappa	1994	We present a comprehensive algorithm to construct a topologically correct triangulation of the real affine part of a rational parametric surface with few restrictions on the defining rational functions. The rational functions are allowed to be undefined on domain curves (pole curves) and at certain special points (base points), and the surface is allowed to have nodal or cuspidal self-intersections. We also recognize that for a complete display some real points on the parametric surface may be generated only by complex parameter values, and that some finite points on the surface may be generated only by infinite parameter values; we show how to compensate for these conditions. Our techniques for handling these problems have applications in scientific visualization, rendering non-standard NURBS, and in finite-element mesh generation.
IEEE Visualization	The Design and Implementation of the Cortex Visualization System.	Deb Banerjee,Chris Morley,Wayne Smith	1994	Cortex has been designed for interactive analysis and display of simulation data generated by CFD applications based on unstructured-grid solvers. Unlike post-processing visualization environments, Cortex is designed to work in co-processing mode with the CFD application. This significantly reduces data storage and data movement requirements for visualization and also allows users to interactively steer the application. Further, Cortex supports high-performance by running on massively parallel computers and workstation clusters.An important goal for Cortex is to provide visualization to a variety of solvers which differ in their solution methodologies and supported flow models. Coupled with the co-processing requirement, this has required the development of a well defined programming interface to the CFD solver that lets the the visualization system communicate efficiently with the solver, and requires minimal programming effort for porting to new solvers. Further, the requirement for targeting multiple solvers and application niches demands that the visualization system be rapidly and easily modifiable. Such flexibility is attained in Cortex by using the high-level, interpreted language Scheme for implementing user-interfaces and high-level visualization functions. By making the Scheme interpreter available from the Cortex text interface, the user can also customize and extend the visualization system.
IEEE Visualization	Vortex Tubes in Turbulent Flows: Identification, Representation, Reconstruction.	David C. Banks,Bart A. Singer	1994	"A new algorithm for identifying vortices in complex flows is presented. The scheme uses both the vorticity and pressure fields. A skelton line along the center of a vortex is produced by a two-step predictor-corrector scheme. The technique uses the vector field to move in the direction of the skeleton line and the scalar field to correct the location in the plane perpendicular to the skelton line. With an economical description of the vortex tube's cross-section, the skeleton compresses the representation of the flow by a factor of 4000 or more. We show how the reconstructed geometry of vortex tubes can be enhanced to help visualize helical motion."
IEEE Visualization	Information Workspaces for Large Scale Cognition.	Stuart K. Card	1994	Information Workspaces for Large Scale Cognition.
IEEE Visualization	Virtual Reality Performance for Virtual Geometry.	Robert A. Cross,Andrew J. Hanson	1994	"We describe the theoretical and practical visualization issues solved in the implementation of an interactive real-time four-dimensional geometry interface for the CAVE, an immersive virtual reality environment. While our specific task is to produce a ""virtual geometry"" experience by approximating physically correct rendering of manifolds embedded in four dimensions, the general principles exploited by our approach reflect requirements common to many immersive virtual reality applications, especially those involving volume rendering. Among the issues we address are the classification of rendering tasks, the specialized hardware support required to attain interactivity, specific techniques required to render 4D objects, and interactive methods appropriate for our 4D virtual world application."
IEEE Visualization	The Topology of Symmetric, Second-Order Tensor Fields.	Thierry Delmarcelle,Lambertus Hesselink	1994	We study the topology of symmetric, second-order tensor fields. The goal is to represent their complex structure by a simple set of carefully chosen points and lines analogous to vector field topology. We extract topological skeletons of the eigenvector fields, and we track their evolution over time. We study tensor topological transitions and correlate tensor and vector data.The basic constituents of tensor topology are the degenerate points, or points where eigenvalues are equal to each other. Degenerate points play a similar role as critical points in vector fields. We identify two kinds of elementary degenerate points, which we call wedges and trisectors. They can combine to form more familiar singularities---such as saddles, nodes, centers, or foci. However, these are generally unstable structures in tensor fields.Finally, we show a topological rule that puts a constraint on the topology of tensor fields defined across surfaces, extending to tensor fields the Pointcar&eacute;-Hopf theorem for vector fields.
IEEE Visualization	An Object Oriented Design for the Visualization of Multi-Variable Data Objects.	Jean-Marie Favre,James K. Hahn	1994	This paper presents an object-oriented system design supporting the composition of scientific data visualization techniques based on the definition of hierarchies of typed data objects and tools. Traditional visualization systems focus on creating graphical objects which often cannot be re-used for further processing. Our approach provides objects of different topological dimension to offer a natural way of describing the results of visualization mappings. Serial composition of data extraction tools is allowed, while each intermediate visualization object shares a common description and behavior. Visualization objects can be re-used, facilitating the data exploration process by expanding the available analysis and correlation functions provided. This design offers an open-ended architecture for the development of new visualization techniques. It promotes data and software re-use, eliminates the need for writing special purpose software and reduces processing requirements during interactive visualization sessions.
IEEE Visualization	User Modeling for Adaptive Visualization Systems.	Gitta Domik,Bernd Gutkauf	1994	Meaningful scientific visualizations benefit the interpretation of scientific data, concepts and processes. To ensure meaningful visualizations, the visualization system needs to adapt to desires, disabilities and abilities of the user, interpretation aim, resources (hardware, software) available, and the form and content of the data to be visualized. We suggest to describe these characteristics by four models: user model, problem domain/task model, resource model and data model. This paper makes suggestions for the generation of a user model as a basis for an adaptive visualization system.We propose to extract information about the user by involving the user in interactive computer tests and games. Relevant abilities tested are color preception, color memory, color ranking, mental rotation, and fine motor coordination.
IEEE Visualization	A Case Study on Visualization for Boundary Value Problems.	Gábor Domokos,Randy C. Paffenroth	1994	In this paper we present a method, and a software based on this method, making highly inter-active visualization possible for computational results on nonlinear BVPs associated with ODEs. The program PCR relies partly on computer graphics tools and partly on real-time computations, the combination of which not only helps the understanding of complex problems, it also permits the reduction of stored data by orders of magnitude. The method has been implemented on PCs (running on DOS) and on the Application Visualization System (AVS) for UNIX machines, this paper provides a brief introduction to the latter version besides describing the mathematical background of the method.
IEEE Visualization	Visualizing Flow Over Curvilinear Grid Surfaces Using Line Integral Convolution.	Lisa K. Forssell	1994	"Line Integral Convolution (LIC), introduced by Cabral and Leedom in Siggraph '93, is a powerful technique for imaging and animating vector fields. We extend the LIC paradigm in three ways:1. The existing technique is limited to vector fields over a regular Cartesian grid. We extend it to vector fields over parametric surfaces, specifically those found in curvilinear grids, used in computational fluid dynamics simulations.2. Periodic motion filters can be used to animate the flow visualization. When the flow lies on a parametric surface, however, the motion appears misleading. We explain why this problem arises and show how to adjust the LIC algorithm to handle it.3. We introduce a technique to visualize vector magnitude as well as vector direction. Cabral and Leedom have suggested a method for variable-speed animation, which is based on varying the frequency of the filter function. We develop a different technique based on kernel phase shifts which we have found to show substantially better results.Our implementation of these algorithms utilizes texture-mapping hardware to run in real time, which allows them to be included in interactive applications."
IEEE Visualization	Spiders: A New User Interface for Rotation and Visualization of N-dimensional Point Sets.	Kirk L. Duffin,William A. Barrett	1994	We present a new method for creating n-dimensional rotation matrices from manipulating the projections of n-dimensional data coordinate axes onto a viewing plane. A user interface for n-dimensional rotation is implemented. The interface is shown to have no rotational hysteresis.
IEEE Visualization	Nonpolygonal Isosurface Rendering for Large Volume Datasets.	James W. Durkin,John F. Hughes	1994	Surface-based rendering techniques, particularly those that extract a polygonal approximation of an isosurface, are widely used in volume visualization. As dataset size increases though, the computational demands of these methods can overwhelm typically available computing resources. Recent work on accelerating such techniques has focused on preprocessing the volume data or postprocessing the extracted polygonization. Our new algorithm concentrates instead on streamlining the surface extraction process itself so as to accelerate the rendering of large volumes. The technique shortens the conventional isosurface visualization pipeline by eliminating the intermediate polygonization. We compute the contribution of the isosurface within a volume cell to the resulting image directly from a simplified numerical description of the cell/surface intersection. Our approach also reduces the work in the remaining stages of the visualization process. By quantizing the volume data, we exploit precomputed and cached data at key processing steps to improve rendering efficiency. The resulting implementation provides comparatively fast renderings with reasonable image quality.
IEEE Visualization	Volume Rendering Methods for Computational Fluid Dynamics Visualization.	David S. Ebert,Roni Yagel,James N. Scott,Yair Kurzion	1994	This paper describes three alternative volume rendering approaches to visualizing computational fluid dynamics (CFD) data. One new approach uses realistic volumetric gas rendering techniques to produce photo-realistic images and animations from scalar CFD data. The second uses ray casting that is based on a simpler illumination model and is mainly centered around a versatile new tool for the design of transfer functions. The third method employs a simple illumination model and rapid rendering mechanisms to provide efficient preview capabilities. These tools provide a large range of volume rendering capabilities to be used by the CFD explorer to render rapidly for navigation through the data, to emphasize data features (e.g., shock waves) with a specific transfer function, or to present a realistic rendition of the model.
IEEE Visualization	Wavelet-Based Volume Morphing.	Taosong He,Sidney W. Wang,Arie E. Kaufman	1994	This paper presents a technique for performing volume morphing between two volumetric datasets in the wavelet domain. The idea is to decompose the volumetric datasets into a set of frequency bands, apply smooth interpolation to each band, and reconstruct to form the morphed model. In addition, a technique for establishing a suitable correspondence among object voxels is presented. The combination of these two techniques results in a smooth transition between the two datasets and produces morphed volume with fewer high frequency distortions than those obtained from spatial domain volume morphing.
IEEE Visualization	Visualization in the Information Highway.	Nahum D. Gershon	1994	Visualization in the Information Highway.
IEEE Visualization	Visualizing Flow with Quaternion Frames.	Andrew J. Hanson,Hui Ma	1994	"Flow fields, geodesics, and deformed volumes are natural sources of families of space curves that can be characterized by intrinsic geometric properties such as curvature, torsion, and Frenet frames. By expressing a curve's moving Frenet coordinate frame as an equivalent unit quaternion, we reduce the number of components that must be displayed from nine with six constraints to four with one constraint. We can then assign a color to each curve point by dotting its quaternion frame with a 4D light vector, or we can plot the frame values separately as a curve in the threesphere. As example, we examine twisted volumes used in topology to construct knots and tangles, a spherical volume deformation known as the Dirac string trick, and streamlines of 3D vector flow fields."
IEEE Visualization	A Lattice Model for Data Display.	William L. Hibbard,Charles R. Dyer,Brian E. Paul	1994	In order to develop a foundation for visualization, we develop lattice models for data objects and displays that focus on the fact that data objects are approximations to mathematical objects and real displays are approximations to ideal displays. These lattice models give us a way to quantize the information content of data and displays and to define conditions on the visualization mappings from data to displays. Mappings satisfy these conditions if and only if they are lattice isomorphisms. We show how to apply this result to scientific data and display models, and discuss how it might be applied to recursively defined data types appropriate for complex information processing.
IEEE Visualization	Visualizing multidimensional (multivariate) data and relations (Panel).	Alfred Inselberg,Georges G. Grinstein,Ted Mihalisin,Hans Hinterberger	1994	Visualizing multidimensional (multivariate) data and relations (Panel).
IEEE Visualization	Isosurface Generation by Using Extrema Graphs.	Takayuki Itoh,Koji Koyamada	1994	"A high-performance algorithm for generating isosurfaces is presented. In this algorithm, extrema points in a scalar field are first extracted. A graph is then generated in which the extrema points are taken as nodes. Each arc of the graph has a list of IDs of the cells that are intersected by the arc. A boundary cell list ordered according to cells' values is also generated. The graph and the list generated in this pre-process are used as a guide in searching for seed cells. Isosurfaces are generated from seed cells that are found in arcs of the graph. In this process isosurfaces appear to propagate themselves. The algorithm visits only cells that are intersected by an isosurface and cells whose IDs are included in cell lists. It is especially efficient when many isosurfaces are interactively generated in a huge volume. Some benchmark tests described in this paper show the efficiency of the algorithm."
IEEE Visualization	Case Study: Visualization of Mesoscale Flow Features in Ocean Basins.	Andreas Johannsen,Robert J. Moorhead	1994	Environmental issues such as global warming are an active area of international research and concern today. This case study describes various visualization paradigms that have been developed and applied in an attempt to elucidate the information provided by environmental models and observations. The ultimate goal is to accurately measure the existence of any long term climatological change. The global ocean is the starting point, since it is a major source and sink of heat within our global environment.
IEEE Visualization	The Crucial Difference Between Human and Machine Vision: Focal Attention.	Bela Julesz	1994	The Crucial Difference Between Human and Machine Vision: Focal Attention.
IEEE Visualization	UFAT - A Particle Tracer for Time-Dependent Flow Fields.	David A. Lane	1994	Time-dependent (unsteady) flow fields are commonly generated in Computational Fluid Dynamics (CFD) simulations; however, there are very few flow visualization systems that generate particle traces in unsteady flow fields. Most existing systems generate particle traces in time-independent flow fields. A particle tracing system has been developed to generate particle traces in unsteady flow fields. The system was used to visualize several 3D unsteady flow fields from real-world problems, and it has provided useful insights into the time-varying phenomena in the flow fields. In this paper, the design requirements and the architecture of the system are described. Some examples of particle traces computed by the system are also shown.
IEEE Visualization	Strata-Various: Multi-Layer Visualization of Dynamics in Software System Behavior.	Doug Kimelman,Bryan S. Rosenburg,Tova Roth	1994	"Current software visualization tools are inadequate for understanding, debugging, and tuning realistically complex applications. These tools often present only static structure, or they present dynamics from only a few of the many layers of a program and its underlying system. This paper introduces ""PV"", a prototype program visualization system which provides concurrent visual presentation of behavior from all layers, including: the program itself, user-level libraries, the operating system, and the hardware, as this behavior unfolds over time. PV juxtaposes views from different layers in order to facilitate visual correlation, and allows these views to be navigated in a coordinated fashion. This results in an extremely powerful mechanism for exploring application behavior. Experience is presented from actual use of PV in production settings with programmers facing real deadlines and serious performance problems."
IEEE Visualization	3D Visualization of Unsteady 2D Airplane Wake Vortices.	Kwan-Liu Ma,Z. C. Zheng	1994	Air flowing around the wing tips of an airplane forms horizontal tornado-like vortices that can be dangerous to following aircraft. The dynamics of such vortices, including ground and atmospheric effects, can be predicted by numerical simulation, allowing the safety and capacity of airports to be improved. In this payer, we introduce three-dimensional techniques for visualizing time-dependent, two-dimensional wake vorte computations, and the hazard strength of such vortices near the ground. We describe a vortex core tracing algorithm and a local tiling method to visualize the vortex evolution. The tiling method converts time-dependent, two-dimensional vortex cores into three-dimensional vortex tubes. Finally, a novel approach is used to calculate the induced rolling moment on the following airplane at each grid point within a region near the vortex tubes and thus allows three-dimensional visualization of the hazard strength of the vortices.
IEEE Visualization	An Annotation System for 3D Fluid Flow Visualization.	Maria M. Loughlin,John F. Hughes	1994	Annotation is a key activity of data analysis. However, current data analysis systems focus almost exclusively on visualization. We propose a system which integrates annotations into a visualization system. Annotations are embedded in 3D data space, using the Post-it1 metaphor. This embedding allows contextual-based information storage and retrieval, and facilitates information sharing in collaborative environments. We provide a traditional database filter and a Magic Lens2 filter to create specialized views of the data. The system is customized for fluid flow applications, with features which allow users to store parameters of visualization tools and sketch 3D volumes.
IEEE Visualization	Introducing Alpha Shapes for the Analysis of Path Integral Monte Carlo Results.	Patrick J. Moran,Marcus Wagner	1994	We present a new technique for the visualization and analysis of the results from Monte Carlo simulations based on &alpha;-complexes and &alpha;-shapes. The specific application we present in this article is the analysis of the quantum-mechanical behavior of hydrogen molecules and helium atoms on a surface at very low temperatures. Our technique is an improvement over existing techniques in two respects. First, our approach allows one to visualize the points on a random walk at varying levels of detail and interactively select the level of detail that is most appropriate. Second using &alpha;-shapes one can obtain quantitative measures of spatial properties of the system, such as the boundary length and interior area of clusters, that would be difficult to obtain otherwise.
IEEE Visualization	Case Study: Visualization of an Electric Power Transmission System.	Pramod M. Mahadev,Richard D. Christie	1994	Visualization techniques are applied to an electric power system transmission network to create a graphical picture of network power flows and voltages. A geographic data map is used. Apparent power flow is encoded as the width of an arrow, with direction from real power flow. Flows are superposed on flow limits. Contour plots and color coding failed for representing bus voltages. A two-color thermometer encoding worked well. The resulting visualization is a significant improvement over current user interface practice in the power industry.
IEEE Visualization	Case Study: Visualization and Data Analysis in Space and Atmospheric Science.	A. Mankofsky,E. P. Szuszczewicz,P. Blanchard,C. Goodrich,D. McNabb,R. Kulkarni,D. Kamins	1994	"In this paper we show how SAVS, a tool for visualization and data analysis in space and atmospheric science, can be used to quickly and easily address problems that would previously have been far more laborious to solve. Based on the popular AVS package, SAVS presents the user with an environment tailored specifically for the physical scientist. Thus there is minimal ""startup"" time, and the scientist can immediately concentrate on his science problem. The SAVS concept readily generalizes to many other fields of science and engineering."
IEEE Visualization	Piecewise-Linear Surface Approximation From Noisy Scattered Samples.	Michael Margaliot,Craig Gotsman	1994	We consider the problem of approximating a smooth surface f(x,y), based on n scattered samples {(xi, yi, zi)ni=1} where the sample values {zi} are contaminated with noise: zi = f(xi, yi) + &epsilon;i. We present an algorithm that generates a PLS (Piecewise Linear Surface) f1, defined on a triangulation of the sample locations V = {(xi, yi)ni=1}, approximating f well. Constructing the PLS involves specifying both the triangulation of V and the values of f1 at the points of V. We demonstrate that even when the sampling process is not noisy, a better approximation for f is obtained using our algorithm, compared to existing methods. This algorithm is useful for DTM (Digital Terrain Map) manipulation by polygon-based graphics engines for visualization applications.
IEEE Visualization	An Evaluation of Reconstruction Filters for Volume Rendering.	Stephen R. Marschner,Richard Lobb	1994	To render images from a three-dimensional array of sample values, it is necessary to interpolate between the samples. This paper is concerned with interpolation methods that are equivalent to convolving the samples with a reconstruction filter; this covers all commonly used schemes, including trilinear and cubic interpolation.We first outline the formal basis of interpolation in three-dimensional signal processing theory. We then propose numerical metrics that can be used to measure filter characteristics that are relevant to the appearance of images generated using that filter. We apply those metrics to several previously used filters and relate the results to isosurface images of the intrpolations. We show that the choice of interpolation scheme can have a dramatic effect on image quality, and we discuss the cost/benefit tradeoff inherent in choosing a filter.
IEEE Visualization	Approximation of Isosurface in the Marching Cube: Ambiguity Problem.	Sergey V. Matveyev	1994	The purpose of the present article is the consideration of the problem of ambiguity over the faces arising in the Marching Cube algorithm. The article shows that for unambiguous choice of the sequence of the points of intersection of the isosurface with edges confining the face it is sufficient to sort them along one of the coordinates. It also presents the solution of this problem inside the cube. The graph theory methods are used to approximate the isosurface inside the cell.
IEEE Visualization	A Library for Visualizing Combinatorial Structures.	Marc Najork,Marc H. Brown	1994	"This paper describes ANIM3D, a 3D animation library targeted at visualizing combinatorial structures. In particular, we are interested in algorithm animation. Constructing a new view for an algorithm typically takes dozens of design iterations, and can be very time-consuming. Our library eases the programmer's burden by providing high-level constructs for performing animations, and by offering an interpretive environment that eliminates the need for recompilations. This paper also illustrates ANIM3D's expressiveness by developing a 3D animation of Dijkstra's shortest-path algorithm in just 70 lines of code."
IEEE Visualization	Visualizing 3D Velocity Fields Near Contour Surfaces.	Nelson L. Max,Roger Crawfis,Charles Grant	1994	Vector field rendering is difficult in 3D because the vector icons overlap and hide each other. We propose four different techniques for visualizing vector fields only near surfaces. The first uses motion blurred particles in a thickened region around the surface. The second uses a voxel grid to contain integral curves of the vector field. The third uses many antialiased lines through the surface, and the fourth uses hairs sprouting from the surface and then bending in the direction of the vector field. All the methods use the graphics pipeline, allowing real time rotation and interaction, and the first two methods can animate the texture to move in the flow determined by the velocity field.
IEEE Visualization	Feature Detection from Vector Quantities in a Numerically Simulated Hypersonic Flow Field in Combination with Experimental Flow Visualization.	Hans-Georg Pagendarm,Birgit Walter	1994	In computational fluid dynamics visualization is a frequently used tool for data evaluation, understanding of flow characteristics, and qualitative comparison to flow visualizations originating from experiments. Building on an existing visualization software system, that allows for a careful selection of state-of-the-art visualization techniques and some extensions, it became possible to present various features of the data in a single image. The visualizations show vortex position and rotation as well as skin-friction lines, experimental oil-flow traces, and shock-wave positions. By adding experimental flow visualization, a comparison between numerical simulation and wind-tunnel flow becomes possible up to a high level of detail. Since some of the underlying algorithms are not yet described in detail in the visualization literature, some experiences gained from the implementation are illustrated.
IEEE Visualization	Mix&Match: A Construction Kit for Visualization.	Alex Pang,Naim Alper	1994	Mix&Match: A Construction Kit for Visualization.
IEEE Visualization	Case Study: Tokamak Plasma Turbulence Visualization.	Scott E. Parker,Ravi Samtaney	1994	One of the most fundamental issue in magnetic fusion research is the understanding of turbulent transport observed in present-day tokamak experiments. Plasma turbulence is very challenging from a theoretical point of view due to the nonlinearity and high dimensionality of the governing equations. Recent developments in algorithms along with the astounding advances in high performance computing now make first-principle particle simulations an important tool for improved understanding of such phenomena. Due to the five dimensional phase space (3 spatial, 2 velocity) and complex toroidal geometry, visualization is crucial for interpreting such simulation data. This paper discusses how visualization tools are currently used and what new physics has been elucidated, along with what can be learned about tokamak turbulence through the interplay between theory, simulation and visualization.
IEEE Visualization	Case Study: Integrating Spatial Data Display with Virtual Reconstruction.	Philip Peterson,Brian Hayden,F. David Fracchia	1994	In the process of archaeological excavation, a vast amount of data, much of it three-dimensional in nature, is recorded. In recent years, computer graphics techniques have been applied to the task of visualizing such data. In particular, data visualization has been used to accomplish the virtual reconstruction of site architecture and to enable the display of spatial data distributions using three-dimensional models of site terrain. In the case we present here, these two approaches are integrated in the modeling of a prehistoric pithouse. In order to better visualize artifact distributions in the context of site architecture, surface data is displayed as a layer in a virtual reconstruction viewable at interactive rates. This integration of data display with the architectural model has proven valuable in identifying correlations between distributions of different artifact categories and their spatial proximity to significant architectural features.
IEEE Visualization	Visualization and Geographic Information System Integration: What are the needs and the requirements, if any? (Panel).	Theresa-Marie Rhyne,William Ivey,Loey Knapp,Peter Kochevar,Tom Mace	1994	Visualization and Geographic Information System Integration: What are the needs and the requirements, if any? (Panel).
IEEE Visualization	A Visualization System on Every Desk - Keeping it Simple.	Steven F. Roth	1994	A Visualization System on Every Desk - Keeping it Simple.
IEEE Visualization	Case Study: Visualization of Volcanic Ash Clouds.	Mitchell Roth,Rick Guritz	1994	Ash clouds resulting from volcanic eruptions are a serious hazard to aviation safety. In Alaska alone, there are over 40 active volcanoes whose eruptions may affect more than 40,000 flights using the great circle polar routes each year. The clouds are especially problematic because they are invisible to radar and nearly impossible to distinguish from weather clouds. The Arctic Region Supercomputing Center and the Alaska Volcano Observatory have collaborated to develop a system for predicting and visualizing the movement of volcanic ash clouds when an eruption occurs. The output from the model is combined with a digtal elevation model to produce a realistic view of the ash cloud which may be examined interactively from any desired point of view at any time during the prediction period. This paper describes the visualization techniques employed in the system and includes a video animation of the 1989 Mount Redoubt eruption which caused complete engine failure on a 747 passenger jet.
IEEE Visualization	Visualization in Medicine: VIRTUAL Reality or ACTUAL Reality? (Panel).	Christian Roux,Jean-Louis Coatrieux,Jean-Louis Dillenseger,Elliot K. Fishman,Murray H. Loew,Hans-Peter Meinzer,Justin D. Pearlman	1994	Visualization in Medicine: VIRTUAL Reality or ACTUAL Reality? (Panel).
IEEE Visualization	A distributed, parallel, interactive volume rendering package.	John S. Rowlan,G. Edward Lent,Nihar Gokhale,Shannon Bradshaw	1994	This paper presents a parallel ray-casting volume rendering algorithm and its implementation on the massively parallel IBM SP-1 computer using the Chameleon message passing library. Though this algorithm takes advantage of many of the unique features of the SP-1 (e.g. high-speed switch, large memory per node, high-speed disk array, HIPPI display, et al), the use of Chameleon allows the code to be executed on any collection of workstations.The algorithm is image-ordered and distributes the data and the computational load to individual processors. After the volume data is distributed, all processors then perform local raytracing of their respective subvolumes concurrently. No interprocess communication takes place during the ray tracing process. After a subimage is generated by each processor, the final image is obtained by composing subimages between all the processers.The program itself is implemented as an interactive process through a GUI residing on a graphics work-station which is coupled to the parallel rendering algorithm via sockets. The paper highlights the Chameleon implementation, the GUI, some optimization improvements, static load balancing, and direct parallel display to a HIPPI framebuffer.
IEEE Visualization	Case Study: Volume Rendering of Pool Fire Data.	Holly E. Rushmeier,A. Hamins,M.-Y. Choi	1994	We describe how techniques from computer graphics are used to visualize pool fire data and computer radiative effects from pool fires. The basic tools are ray casting and accurate line integration using the RADCAL program. Example images in the visible and infrared band are shown which give qualitative insights about the fire data. Examples are given of irradiation calculations and novel methods to visualize the results of irradiation calculations.
IEEE Visualization	Case Study: Visualization of 3D Ultrasonic Data.	Georgios Sakas,Lars-Arne Schreyer,Marcus Grimm	1994	3D Ultrasound is one of the most interesting non-invasive, non-radiative tomographic techniques. Rendering 3D modells from such data is not straight-forward due to the noisy, fuzzy nature of ultrasound imaging containing a lot of artefacts. In this paper we first apply speckle, median and gaussian prefiltering to improve the image quality. Using several semi-automatic segmentation tools we isolate interesting features within a few minutes. Our improved surface-extraction procedure enables volume rendering of high quality within a few seconds on a normal work-station, thus making the complete system suitable for routine clinical applications.
IEEE Visualization	Implicit Modeling of Swept Surfaces and Volumes.	William J. Schroeder,William E. Lorensen,Steve Linthicum	1994	Swept surfaces and volumes are generated by moving a geometric model through space. Swept surfaces and volumes are important in many computer-aided design applications including geometric modeling, numerical cutter path generation, and spatial path planning. In this paper we describe a numerical algorithm to generate swept surfaces and volumes using implicit modeling techniques. The algorithm is applicable to any geometric representation for which a distance function can be computed. The algorithm also treats degenerate trajectories such as self-intersection and surface singularity. We show applications of this algorithm to maintainability design and robot path planning.
IEEE Visualization	Differential Volume Rendering: A Fast Volume Visualization Technique for Flow Animation.	Han-Wei Shen,Christopher R. Johnson	1994	We present a direct volume rendering algorithm to speed up volume animation for flow visualizations. Data coherency between consecutive simulation time steps is used to avoid casting rays from those pixels retaining color values assigned to the previous image. The algorithm calculates the differential information among a sequence of 3D volumetric simulation data. At each time step the differential information is used to compute the locations of pixels that need updating and a ray-casting method is utilized to produce the updated image. We illustrate the utility and speed of the differential volume rendering algorithm with simulatin data from computational bioelectric and fluid dynamics applications. We can achieve considerable disk-space savings and nearly real-time rendering of 3D flows using low-cost, single processor workstations* for models which contain hundreds of thousands of data points.
IEEE Visualization	The Future of Graphic User Interfaces.	Ben Shneiderman	1994	The Future of Graphic User Interfaces.
IEEE Visualization	Parallel Performance Measures for Volume Ray Casting.	Cláudio T. Silva,Arie E. Kaufman	1994	We describe a technique for achieving fast volume ray casting on parallel machines, using a load balancing scheme and an efficient pipelined approach to compositing. We propose a new model for measuring the amount of work one needs to perform in order to render a given volume, and use this model to obtain a better load balancing scheme for distributed memory machines. We also discuss in detail the design tradeoffs of our technique. In order to validate our model we have implemented it on the Intel iPSC/860 and the Intel Paragon, and conducted a detailed performance analysis.
IEEE Visualization	Fast Surface Rendering from Raster Data by Voxel Traversal Using Chessboard Distance.	Milos Srámek	1994	Fast Surface Rendering from Raster Data by Voxel Traversal Using Chessboard Distance.
IEEE Visualization	Case Study: Observing a Volume Rendered Fetus within a Pregnant Patient.	Andrei State,David T. Chen,Chris Tector,Andrew Brandt,Hong Chen,Ryutarou Ohbuchi,Michael Bajura,Henry Fuchs	1994	"Augmented reality systems with see-through head-mounted displays have been used primarily for applications that are possible with today's computational capabilities. We explore possibilities for a particular application---in-place, real-time 3D ultrasound visualization---without concern for such limitations. The question is not ""How well could we currently visualize the fetus in real time,"" but ""How well could we see the fetus if we had sufficient compute power?""Our video sequence shows a 3D fetus within a pregnant woman's abdomen---the way this would look to a HMD user. Technical problems in making the sequence are discussed. This experience exposed limitations of current augmented reality systems; it may help define the capabilities of future systems needed for applications as demanding as real-time medical visualization."
IEEE Visualization	Visualizing Data: Is Virtual Reality the Key? (Panel).	Linda M. Stone,Thomas Erickson,Benjamin B. Bederson,Peter Rothman,Raymond Muzzy	1994	Visualizing Data: Is Virtual Reality the Key? (Panel).
IEEE Visualization	GASP - A System for Visualizing Geometric Algorithms.	Ayellet Tal,David P. Dobkin	1994	"This paper describes a system, GASP, that facilitates the visualization of geometric algorithms. The user need not have any knowledge of computer graphics in order to quickly generate a visualization. The system is also intended to facilitate the task of implementing and debugging geometric algorithms. The viewer is provided with a comfortable user interface enhancing the exploration of an algorithm's functionality. We describe the underlying concepts of the system as well as a variety of examples which illustrate its use."
IEEE Visualization	Progressive Transmission of Scientific Data Using Biorthogonal Wavelet Transform.	Hai Tao,Robert J. Moorhead	1994	An important issue in scientific visualization systems is the management of data sets. Most data sets in scientific visualization, whether created by measurement or simulation, are usually voluminous. The goal of data management is to reduce the storage space and the access time of these data sets to speed up the visualization process. A new progressive transmission scheme using spline biorthogonal wavelet bases is proposed in this paper. By exploiting the properties of this set of wavelet bases, a fast algorithm involving only additions and subtractions is developed. Due to the multiresolutional nature of the wavelet transform, this scheme is compatible with hierarchical-structured rendering algorithms. The formula for reconstructing the functional values in a continuous volume space is given in a simple polynomial form. Lossless compression is possible, even when using floating-point numbers. This algorithm has been applied to data from a global ocean model. The lossless compression ratio is about 1.5:1. With a compression ratio of 50:1, the reconstructed data is still of good quality. Several other wavelet bases are compared with the spline biorthogonal wavelet bases. Finally, the reconstructed data is visualized using various algorithms and the results are demonstrated.
IEEE Visualization	Case Study: Severe Rainfall Events in Northwestern Peru (Visualization of Scattered Meteorological Data).	Lloyd Treinish	1994	The ordinarily arid climate of coastal Peru is disturbed every few years by a phenomenon called El Ni&ntilde;o, characterized by a warming in the Pacific Ocean. Severe rainstorms are one of the consequences of El Ni&ntilde;o, which cause great damage. An examination of daily data from 66 rainfall stations in the Chiura-Piura region of north-western Peru from late 1982 through mid-1983 (associated with an El Ni&ntilde;o episode) yields information on the mesoscale structure of these storms. These observational data are typical of a class that are scattered at irregular locations in two dimensions. The use of continuous realization techniques for qualitative visualization (e.g., surface deformation or contouring) requires an intermediate step to define a topological relationship between the locations of data to form a mesh structure. Several common methods are considered, and the results of their application to the study of the rainfall events are analyzed.
IEEE Visualization	Restorer: A Visualization Technique for Handling Missing Data.	Ray Twiddy,John Cavallo,Shahram M. Shiri	1994	Pseudocoloring is a frequently used technique in scientific visualization for mapping a color to a data value. When using pseudocolor and animation to visualize data that contain missing regions displayed as black or transparent, the missing regions popping in and out can distract the viewer from the more relevant information. Filling these gaps with interpolated data could lead to a misinterpretation of the data. This paper presents a method for combining pseudocoloring and grayscale in the same colormap. Valid data are mapped to colors in the colormap. The luminance values of the colors bounding areas of missing data are used in interpolating over these regions. The missing data are mapped to the grayscale portion of the colormap. This approach has the advantages of eliminating distracting gaps caused by missing data and distinguishing between those areas that represent valid data and those areas that do not. This approach was inspired by a technique used in the restoration of paintings.
IEEE Visualization	Validation, verification and evaluation (Panel).	Samuel P. Uselton,Geoff Dorn,Charbel Farhat,Michael W. Vannier,Kim Esbensen,Al Globus	1994	Validation, verification and evaluation (Panel).
IEEE Visualization	XmdvTool: Integrating Multiple Methods for Visualizing Multivariate Data.	Matthew O. Ward	1994	Much of the attention in visualization research has focussed on data rooted in physical phenomena, which is generally limited to three or four dimensions. However, many sources of data do not share this dimensional restriction. A critical problem in the analysis of such data is providing researchers with tools to gain insights into characteristics of the data, such as anomalies and patterns. Several visualization methods have been developed to address this problem, and each has its strengths and weaknesses. This paper describes a system named XmdvTool which integrates several of the most common methods for projecting multivariate data onto a two-dimensional screen. This integration allows users to explore their data in a variety of formats with ease. A view enhancement mechanism called an N-dimensional brush is also described. The brush allows users to gain insights into spatial relationships over N dimensions by highlighting data which falls within a user-specified subspace.
IEEE Visualization	Visualizing Polycrystalline Orientation Microstructures with Spherical Color Maps.	Boris Yamrom,John A. Sutliff,Andrew P. Woodfield	1994	Spherical color maps can be an effective tool in the microstructure visualization of polycrystals. Electron backscatter diffraction pattern analysis provides large arrays of the orientation data that can be visualized easily using the technique described in this paper. A combination of this technique with the traditional black and white scanning electron microscopy imaging will enable scientists to better understand the correlation between material properties and their polycrystalline structure.
IEEE Visualization	"Proceedings IEEE Visualization '94, Washington, DC, USA, October 17-21, 1994"	R. Daniel Bergeron,Arie E. Kaufman	1994	"Proceedings IEEE Visualization '94, Washington, DC, USA, October 17-21, 1994"
ICCV	Active Fixation Using Attentional Shifts, Affine Resampling, and Multiresolution Search.	A. Lynn Abbott,Bibo Zheng	1995	Presents a new approach for fixing two cameras at a single location in a 3D scene. Fixation requires the detection of binocular disparity for a single point of interest in the scene so that vergence control can reduce that disparity to zero. Most existing systems use area-based matching, since feature-based matching is computationally prohibitive. Unfortunately, the area-based approach does not perform well when confronted with steeply inclined surfaces, occlusions, repeating patterns or featureless image regions. The method presented in this paper utilizes attentional shifts and affine resampling to combat these problems. These are integrated with adaptive window-size control and coarse-to-fine correlation-based searching. The effectiveness of the approach for complex scenes is demonstrated with several stereo image pairs.
ICCV	Tracking and Recognizing Rigid and Non-Rigid Facial Motions Using Local Parametric Models of Image Motion.	Michael J. Black,Yaser Yacoob	1995	This paper explores the use of local parametrized models of image motion for recovering and recognizing the non-rigid and articulated motion of human faces. Parametric flow models (for example affine) are popular for estimating motion in rigid scenes. We observe that within local regions in space and time, such models not only accurately model non-rigid facial motions but also provide a concise description of the motion in terms of a small number of parameters. These parameters are intuitively related to the motion of facial features during facial expressions and we show how expressions such as anger, happiness, surprise, fear, disgust and sadness can be recognized from the local parametric motions in the presence of significant head motion. The motion tracking and expression recognition approach performs with high accuracy in extensive laboratory experiments involving 40 subjects as well as in television and movie sequences.
ICCV	Reconstructing Complex Surfaces from Multiple Stereo Views.	Pascal Fua	1995	We present a framework for 3D surface reconstruction that can be used to model fully 3 dimensional scenes from an arbitrary number of stereo views. Taken from vastly different viewpoints. This is a key step toward producing 3D world descriptions of complex scenes using stereo and is a very challenging problem: real world scenes tend to contain many 3D objects, they do not usually conform to the 2-1/2D assumption made by traditional algorithms, and one cannot take it for granted that the computed 3D points can easily be clustered into separate groups. By combining a particle based representation, robust fitting, and optimization of an image based objective function, we have been able to reconstruct surfaces without any a priori knowledge of their topology and in spite of the noisiness of the stereo data. Our current implementation goes through three steps-initializing a set of particles from the input 3D data, optimizing their location, and finally grouping them into global surfaces. Using several complex scenes containing multiple objects, we demonstrate its competence and ability to merge information and thus to go beyond what can be done with conventional stereo alone.
ICCV	Epipole and Fundamental Matrix Estimation Using Virtual Parallax.	Boubakeur Boufama-Seddik,Roger Mohr	1995	The paper addresses the problem of computing the fundamental matrix which describes a geometric relationship between a pair of stereo images: the epipolar geometry. We propose a novel method based on virtual parallax. Instead of computing directly the 3/spl times/3 fundamental matrix, we compute a homography with one epipole position, and show that this is equivalent to computing the fundamental matrix. Simple equations are derived by reducing the number of parameters to estimate. As a consequence, we obtain an accurate fundamental matrix of rank two with a stable linear computation. Experiments with simulated and real images validate our method and clearly show the improvement over existing methods.
ICCV	Seeing Behind the Scene: Analysis of Photometric Properties of Occluding Edges by the Reversed Projection Blurring Model.	Naoki Asada,Hisanaga Fujiwara,Takashi Matsuyama	1995	This paper analyzes photometric properties of occluding edges and proves that (1) we can observe surface edges on the farther object located close to the occluding edge even if they are occluded by the nearer object, (2) the image of an occluding edge coincides with that of a surface edge on the nearer object if the brightness of the farther object is uniform around the occluding edge. First, we propose a blurring model named the reversed projection blurring model to analyze photometric properties of blurring phenomena of an occluding edge. Using this model, the theoretical proof of the two properties mentioned above is given. Finally, experimental results in real world environments demonstrate the validity of our blurring model as well as the observability of the photometric properties of occluding edges.
ICCV	Visual Navigation Using a Single Camera.	Jean-Yves Bouguet,Pietro Perona	1995	We assess the usefulness of monocular recursive motion estimation techniques for vehicle navigation in the absence of a model for the environment. For this purpose we extend a recently proposed recursive motion estimator, the Essential filter, to handle scale estimation. We examine experimentally the accuracy with which the motion and position of the vehicle may be computed on an 8000 frame indoors sequence. The issues of sampling time frequency and number of necessary features in the environment are addressed systematically.
ICCV	Layered Representation of Motion Video Using Robust Maximum-Likelihood Estimation of Mixture Models and MDL Encoding.	Serge Ayer,Harpreet S. Sawhney	1995	Representing and modeling the motion and spatial support of multiple objects and surfaces from motion video sequences is an important intermediate step towards dynamic image understanding. One such representation, called layered representation, has recently been proposed. Although a number of algorithms have been developed for computing these representations, there has not been a consolidated effort into developing a precise mathematical formulation of the problem. This paper presents one such formulation based on maximum likelihood estimation (MLE) of mixture models and the minimum description length (MDL) encoding principle. The three major issues in layered motion representation are: (i) how many motion models adequately describe image motion, (ii) what are the motion model parameters, and (iii) what is the spatial support layer for each motion model.
ICCV	Scale-Space from Nonlinear Filters.	J. Andrew Bangham,Paul D. Ling,Richard Harvey	1995	Decomposition by extrema is put into the context of linear vision systems and scale-space. One dimensional discrete M- and N-sieves neither introduce new edges as the scale increases nor create new extrema. They share this property with diffusion based filters. Furthermore M- and N-sieve algorithms are extremely fast with order complexity n. Used to decompose an image, the resulting granularity is appropriate for pattern recognition.
ICCV	Nonlinear Manifold Learning for Visual Speech Recognition.	Christoph Bregler,Stephen M. Omohundro	1995	"A technique for representing and learning smooth nonlinear manifolds is presented and applied to several lip reading tasks. Given a set of points drawn from a smooth manifold in an abstract feature space, the technique is capable of determining the structure of the surface and of finding the closest manifold point to a given query point. We use this technique to learn the ""space of lips"" in a visual speech recognition task. The learned manifold is used for tracking and extracting the lips, for interpolating between frames in an image sequence and for providing features for recognition. We describe a system based on hidden Markov models and this learned lip manifold that significantly improves the performance of acoustic speech recognizers in degraded environments. We also present preliminary results on a purely visual lip reader."
ICCV	Region Tracking through Image Sequences.	Benedicte Bascle,Rachid Deriche	1995	The paper describes an approach to the tracking of complex shapes through image sequences, that combines deformable region models and deformable contours. A deformable region model is presented: its optimisation is based on texture correlation and is constrained by the use of a motion model, such as rigid, affine or homographic. The use of texture information (versus edge information) noticeably improves the tracking performances of deformable models in the presence of texture. Then the region contour is refined using an edge based deformable model in order to better deal with specularities, non planar objects and occlusions. The method is illustrated and validated by experimental results on real images.
ICCV	"Electronically Directed ""Focal"" Stereo."	Peter J. Burt,Lambert E. Wixson,Garbis Salgian	1995	"A key to developing computationally efficient stereo vision is the incorporation of intelligent control. Stereo is most effective when it is able to ""focus"" its analysis on regions and details of a scene that are important to the task at hand, while avoiding less important regions and unnecessary detail. The paper describes two methods for electronically ""focusing"" stereo measurement through simple image pre-processing. The first allows measurement sensitivity to be adjusted. The second allows the shape of the 3D region in which measurements are gathered to be matched to the shape of surfaces in the scene."
ICCV	Recognition Using Region Correspondences.	Ronen Basri,David W. Jacobs	1995	"Recognition systems attempt to recover information about the identity of the observed objects and their location in the environment. A fundamental problem in recognition is the following. Given a correspondence between some portions of an object model and some portions of an image, determine whether the image contains an instance of the object, and, in case it does, determine the transformation that relates the model to the image. The current approaches to this problem are divided into methods that use ``global'''' properties of the object (e.g., centroid and moments of inertia) and methods that use ``local'''' properties of the object (e.g., corners and line segments). Global properties are sensitive to occlusion and, specifically, to self occlusion. Local properties are difficult to locate reliably, and their matching involves intensive computation. A novel method for recognition that uses region information is presented. In our approach the model is divided into volumes, and the image is divided into regions. Given a match between subsets of volumes and regions (without any explicit correspondence between different pieces of the regions) the alignment transformation is computed. The method applies to planar objects under similarity, affine, and projective transformations and to projections of 3-D objects undergoing affine and projective transformations. The new approach combines many of the advantages of the previous two approaches, while avoiding some of their pitfalls. Like the global methods, our approach makes use of region information that reflects the true shape of the object. But like local methods, our approach can handle occlusion."
ICCV	Recognition of Human Body Motion Using Phase Space Constraints.	Lee W. Campbell,Aaron F. Bobick	1995	"A new method for representing and recognizing human body movements is presented. The basic idea is to identify sets of constraints that are diagnostic of a movement: expressed using body-centered coordinates such as joint angles and in force only during a particular movement. Assuming the availability of Cartesian tracking data, we develop techniques for a representation of movements defined by space curves in subspaces of a ""phase space."" The phase space has axes of joint angles and torso location and attitude, and the axes of the subspaces are subsets of the axes of the phase space. Using this representation we develop a system for learning new movements from ground truth data by searching for constraints. We then use the learned representation for recognizing movements in unsegmented data. We train and test the system on nine fundamental steps from classical ballet performed by two dancers; the system accurately recognizes the movements in the unsegmented stream of motion."
ICCV	Head-Eye Calibration.	Mengxiang Li,Demetrios Betsis	1995	We deal with the calibration problem of an active head-eye system, which consists of a pair of cameras mounted on a head with 13 degrees of freedom. The aim of the calibration is to establish relative positions of different 3D systems: between camera and neck, eye and neck, etc., so that we can keep track of the camera position in a fixed (calibration) reference system as a function of the visual parameters of the head-eye system. We formulate the problem and propose both closed-form and nonlinear optimization approaches to solve it. Experiments were carried out and comparison of results with other algorithms were made on both simulated and real data.
ICCV	Active Visual Navigation Using Non-Metric Structure.	Paul A. Beardsley,Ian D. Reid,Andrew Zisserman,David W. Murray	1995	Demonstrates a method of using nonmetric visual information derived from an uncalibrated active vision system to navigate an autonomous vehicle through free-space regions detected in a cluttered environment. The structure of 3-space is recovered modulo an affine transformation using an uncalibrated active stereo head carried by the vehicle. The plane at infinity, necessary for recovering affine structure from projective structure, is found in a novel manner by making controlled rotations of the head. The structure is composed of 3D points obtained by detecting and matching image corners through the stereo image sequence. Considerable care has been taken to ensure that the processing is reliable, robust and automatic. Driveable regions are determined from the projection of the affine structure onto a plane parallel to the ground determined using projective constructs. Two methods of negotiating the regions are explored. The first introduces metric information to allow control of a Euclidean vehicle. The second uses visual servoing of the active head to navigate in the affinely described free-space regions.
ICCV	Learning Geometric Hashing Functions for Model-Based Object Recognition.	George Bebis,Michael Georgiopoulos,Niels da Vitoria Lobo	1995	"A major problem associated with geometric hashing and methods which have emerged from it is the non-uniform distribution of invariants over the hash space. This problem can affect the performance of the method significantly. Finding a ""good"" geometric hash function which redistributes the invariants uniformly over the hash space is not easy. In this paper, a new approach is proposed for alleviating the above problem. It is based on the use of an ""elastic hash table"" which is implemented as a self-organizing feature map neural network (SOFM-NN). In contrast to existing approaches which try to redistribute the invariants over the hash bins, we proceed oppositely, spreading the hash bins over the invariants. During training, the SOFM-NN resembles an elastic net which deforms over the hash space. The objective of the deformation process is to spread more hash bins in hash space areas which are heavily occupied and less hash bins in lower density areas. The advantage of the proposed approach is that it is a process that adapts to the invariants through learning. Hence, it makes absolutely no assumptions about the statistical characteristics of the invariants and the geometric hash function is actually computed through learning. Furthermore, the well known ""topology preserving"" property of the SOFM-NN guarantees that the computed geometric hash function should be well behaved. Finally, the proposed approach is inherently parallelizable."
ICCV	Recovering Object Surfaces from Viewed Changes in Surface Texture Patterns.	Peter N. Belhumeur,Alan L. Yuille	1995	Explores the reconstruction of object surfaces from viewed changes in surface texture patterns. Our approach differs from those in the past in that instead of simply producing local estimates of the surface orientation, our algorithm recovers complete surfaces. Past approaches only found the surface orientation locally and, therefore, did not take advantage of the surface integrability constraint. Our algorithm does not assume that the surface texture pattern is isotropic, and it does not assume that the viewed surface is at some point fronto-parallel. Furthermore, our algorithm has mechanisms for handling texture boundaries and, consequently, does not produce erratic results in the regions abutting these boundaries. Results on real images are presented demonstrating the potential of our approach.
ICCV	Monocular Tracking of the Human Arm in 3D.	Enrico Di Bernardo,Luis Goncalves,Enrico Ursella,Pietro Perona	1995	Monocular Tracking of the Human Arm in 3D.
ICCV	Stereo in the Presence of Specular Reflection.	Dinkar N. Bhat,Shree K. Nayar	1995	The problem of accurate depth estimation using stereo in the presence of specular reflection is addressed. Specular reflection, a fundamental and ubiquitous reflection mechanism, is viewpoint dependent and can cause large intensity differences at corresponding points, resulting in significant depth errors. We analyze the physics of specular reflection and the geometry of stereopsis which led us to a relationship between stereo vergence, surface roughness, and the likelihood of a correct match. Given a lower bound on surface roughness, an optimal binocular stereo configuration can be determined which maximizes precision in depth estimation despite specular reflection. However, surface roughness is difficult to estimate in unstructured environments. Therefore, trinocular configurations, independent of surface roughness, are determined such that at each scene point visible to all sensors, at least one stereo pair can compute produce depth. We have developed a simple algorithm to reconstruct depth from the multiple stereo pairs.
ICCV	Geodesic Active Contours.	Vicent Caselles,Ron Kimmel,Guillermo Sapiro	1995	"A novel scheme for the detection of object boundaries is presented. The technique is based on active contours deforming according to intrinsic geometric measures of the image. The evolving contours naturally split and merge, allowing the simultaneous detection of several objects and both interior and exterior boundaries. The proposed approach is based on the relation between active contours and the computation of geodesics or minimal distance curves. The minimal distance curve lays in a Riemannian space whose metric as defined by the image content. This geodesic approach for object segmentation allows to connect classical ""snakes"" based on energy minimization and geometric active contours based on the theory of curve evolution. Previous models of geometric active contours are improved as showed by a number of examples. Formal results concerning existence, uniqueness, stability, and correctness of the evolution are presented as well."
ICCV	Fast Object Recognition in Noisy Images Using Simulated Annealing.	Margrit Betke,Nicholas C. Makris	1995	A fast simulated annealing algorithm is developed for automatic object recognition. The object recognition problem is addressed as the problem of best describing a match between a hypothesized object and an image. The normalized correlation coefficient is used as a measure of the match. Templates are generated on-line during the search by transforming model images. Simulated annealing reduces the search time by orders of magnitude with respect to an exhaustive search. The algorithm is applied to the problem of how landmarks, e.g., traffic signs, can be recognized by a navigating robot. We illustrate the performance of our algorithm with real-world images of complicated scenes with traffic signs. False positive matches occur only for templates with very small information content. To avoid false positive matches, we propose a method to select model images for robust object recognition by measuring the information content of the model images. The algorithm works well in noisy images for model images with high information content.
ICCV	Face Recognition from One Example View.	David Beymer,Tomaso Poggio	1995	"To create a pose-invariant face recognizer, one strategy is the view-based approach, which uses a set of real example views at different poses. But what if we only have one real view available, such as a scanned passport photo-can we still recognize faces under different poses? Given one real view at a known pose, it is still possible to use the view-based approach by exploiting prior knowledge of faces to generate virtual views, or views of the face as seen from different poses. To represent prior knowledge, we use 2D example views of prototype faces under different rotations. We develop example-based techniques for applying the rotation seen in the prototypes to essentially ""rotate"" the single real view which is available. Next, the combined set of one real and multiple virtual views is used as example views for a view-based, pose-invariant face recognizer. Oar experiments suggest that among the techniques for expressing prior knowledge of faces, 2D example-based approaches should be considered alongside the more standard 3D modeling techniques."
ICCV	On Multi-Feature Integration for Deformable Boundary Finding.	Amit Chakraborty,Marcel Worring,James S. Duncan	1995	Precise segmentation of underlying objects in an image is very important especially for biomedical image analysis. We present an integrated approach for boundary finding using region and curvature information along with the gradient. Unlike the previous methods, where smoothing is enforced by penalizing curvature, here the grey level curvature is used as an extra source of information. However, information fusion may not be useful unless used properly. To address that, we present results that highlight the pros and cons of using the various sources of information and indicate when one should get precedence over the others.
ICCV	Face Detection by Fuzzy Pattern Matching.	Qian Chen,Haiyuan Wu,Masahiko Yachida	1995	"The paper describes an approach to detect faces whose size and position are unknown in an image with a complex background. The candidates of faces are detected by finding out ""face like"" regions in the input image using the fuzzy pattern matching method. The perceptually uniform color space is used in our research in order to obtain reliable results. The skin color that is used to detect face like regions, is represented by a model developed by us called skin color distribution function. The skin color regions are then extracted by estimating a measure that describes how well the color of a pixel looks like the skin color for each pixel in the input image. The faces which appear in images are modeled as several 2 dimensional patterns. The face like regions are extracted by a fuzzy pattern matching approach using these face models. The face candidates are then verified by estimating how well the extracted facial features fit a face model which describes the geometrical relations among facial features."
ICCV	A Model of Figure-Ground Segregation from Kinetic Occlusion.	George T. Chou	1995	The paper presents a model of image segmentation that can distinguish foreground from background purely on the basis of motion information. The main processing steps involved are: detection of motion boundaries, and analysis of figure ground relationship. The proposed model utilizes the observation that in kinetic occlusion, motion boundaries typically display mixture motion information, and foreground surfaces tend to move with motion boundaries. Through distributed probabilistic modeling, these constraints can be embedded into computations with efficient network representations. The resulting networks use spatiotemporal Gabor filters as front ends, and are suitable for parallel distributed processing. We demonstrate the application of the model in the decomposition of moving images into surfaces according to depth.
ICCV	Motion from the Frontier of Curved Surfaces.	Roberto Cipolla,Kalle Åström,Peter J. Giblin	1995	The frontier of a curved surface is the envelope of contour generators showing the boundary, at least locally, of the visible region swept out under viewer motion. In general, the outlines of curved surfaces (apparent contours) from different viewpoints are generated by different contour generators on the surface and hence do not provide a constraint on viewer motion. We show that frontier points, however, have projections which correspond to a real point on the surface and can be used to constrain viewer motion by the epipolar constraint. We show how to recover viewer motion from frontier points for both continuous and discrete motion, calibrated and uncalibrated cameras. We present preliminary results of an iterative scheme to recover the epipolar line structure from real image sequences using only the outlines of curved surfaces. A statistical evaluation as also performed to estimate the stability of the solution.
ICCV	Surface Geometry from Cusps of Apparent Contours.	Roberto Cipolla,Gordon J. Fletcher,Peter J. Giblin	1995	It is known that the deformations of the apparent contours of a surface under perspective projection and viewer motion enable the recovery of the geometry of the surface, for example by utilising the epipolar parametrization. These methods break down with apparent contours that are singular i.e. with cusps. In this paper, we study this situation in detail and show how, nevertheless, the surface geometry (including the Gauss curvature and mean curvature of the surface) can be recovered by following the cusps. Indeed the formulae are much simpler in this case, and require lower spatio-temporal derivatives than in the general case of nonsingular apparent contours. We give a simulated example, and also show that following cusps does not by itself provide us with information on ego-motion.
ICCV	Improving Laser Triangulation Sensors Using Polarization.	James Clark,Emanuele Trucco,H.-F. Cheung	1995	We report a novel application of polarization based vision addressing the robustness of laser triangulation range sensors. Such sensors are based on the accurate detection of a pattern of laser light projected onto a scene, usually a point or line. Typical problems arise with highly specularly reflective surfaces, which can generate visible reflections of the light in various parts of the image. This can confuse the detection algorithms and lead to wrong range measurements. This paper demonstrates experimentally the feasibility of polarization based vision for disambiguating multiple specular inter reflections of the laser light. We concentrate on metal components as they have high interest for inspection in manufacturing, and show positive results with situations of various complexities.
ICCV	Auxiliary Variables for Deformable Models.	Laurent D. Cohen	1995	We present a mathematical formulation for curve and surface reconstruction algorithms by introduction of auxiliary variables. For deformable models and templates, two step iterative algorithms have been often used where, at each iteration, the model is first locally deformed according to the potential data attraction and then globally smoothed. We show how these approaches can be interpreted as the introduction of auxiliary variables and the minimization of a two variables energy. This permits us to transform an implicit data constraint defined by a non convex potential into an explicit convex reconstruction problem. We show some mathematical properties and results on this new auxiliary problem, in particular when the potential is a function of the distance to the closest feature point. We then illustrate our approach for some deformable models and templates and image restoration.
ICCV	Site Model Acquisition and Extension from Aerial Images.	Robert T. Collins,Yong-Qing Cheng,Christopher O. Jaynes,Frank Stolle,Xiaoguang Wang,Allen R. Hanson,Edward M. Riseman	1995	A system has been developed to acquire, extend and refine 3D geometric site models from aerial imagery. This system hypothesizes potential building roofs in an image, automatically locates supporting geometric evidence in other images, and determines the precise shape and position of the new buildings via multiimage triangulation. Model-to-image registration techniques are applied to align new, incoming images against the site model. Model extension and refinement procedures are then performed to add previously unseen buildings and to improve the geometric accuracy of the existing 3D building models.
ICCV	Real-Time Obstacle Avoidance Using Central Flow Divergence and Peripheral Flow.	David Coombs,Martin Herman,Tsai-Hong Hong,Marilyn Nashman	1995	"The lure of using motion vision as a fundamental element in the perception of space drives this effort to use flow features as the sole cues for robot mobility. Real-time estimates of image flow and flow divergence provide the robot's sense of space. The robot steers down a conceptual corridor comparing left and right peripheral flows. Large central flow divergence warns the robot of impending collisions at ""dead ends."" When this occurs, the robot turns around and resumes wandering. Behavior is generated by directly using flow-based information in the 2D image sequence; no 3D reconstruction is attempted. Active mechanical gate stabilization simplifies the visual interpretation problems by reducing camera rotation. By combining corridor following and dead-end deflection, the robot has wandered around the lab at 30 cm/s for as long as 20 minutes without collision. The ability to support this behavior in real-time with current equipment promises expanded capabilities as computational power increases in the future."
ICCV	A Multi-Body Factorization Method for Motion Analysis.	João Paulo Costeira,Takeo Kanade	1995	The structure from motion problem has been extensively studied in the field of computer vision. Yet, the bulk of the existing work assumes that the scene contains only a single moving object. The more realistic case where an unknown number of objects move in the scene has received little attention, especially for its theoretical treatment. We present a new method for separating and recovering the motion and shape of multiple independently moving objects in a sequence of images. The method does not require prior knowledge of the number of objects, nor is dependent on any grouping of features into an object at the image level. For this purpose, we introduce a mathematical construct of object shapes, called the shape interaction matrix, which is invariant to both the object motions and the selection of coordinate systems. This invariant structure is computable solely from the observed trajectories of image features without grouping them into individual objects. Once the structure is computed, it allows for segmenting features into objects by the process of transforming it into a canonical form, as well as recovering the shape and motion of each object.
ICCV	Learning-Based Hand Sign Recognition Using SHOSLIF-M.	Yuntao Cui,Daniel L. Swets,Juyang Weng	1995	We present a self-organizing framework called the SHOSLIF-M for learning and recognizing spatiotemporal events (or patterns) from intensity image sequences. The proposed framework consists of a multiclass, multivariate discriminant analysis to automatically select the most discriminating features (MDF), a space partition tree to achieve a logarithmic retrieval time complexity for a database of n items, and a general interpolation scheme to do view inference and generalization in the MDF space based on a small number of training samples. The system is tested to recognize 28 different hand signs. The experimental results show that the learned system can achieve a 96% recognition rate for test sequences that have not been used in the training phase.
ICCV	Better Optical Triangulation Through Spacetime Analysis.	Brian Curless,Marc Levoy	1995	The standard methods for extracting range data from optical triangulation scanners are accurate only for planar objects of uniform reflectance illuminated by an incoherent source. Using these methods, curved surfaces, discontinuous surfaces, and surfaces of varying reflectance cause systematic distortions of the range data. Coherent light sources such as lasers introduce speckle artifacts that further degrade the data. We present a new ranging method based on analyzing the time evolution of the structured light reflections. Using our spacetime analysis, we can correct for each of these artifacts, thereby attaining significantly higher accuracy using existing technology. We present results that demonstrate the validity of our method using a commercial laser stripe triangulation scanner.
ICCV	Adaptive Shape Evolution Using Blending.	Douglas DeCarlo,Dimitris N. Metaxas	1995	We propose a shape representation scheme which allows two shapes to be combined into a single model. The desired regions of the two shapes are selected, and then merged together forming a blended shape. For reconstruction, blending is incorporated into a deformable model framework. The model automatically adapts to the data, blending when necessary. Hierarchical blending allows multiple blends of a shape to occur forming an evolution from the initial shape of a sphere to the final shape. Blending also allows the insertion of a hole between arbitrary locations. The models used are globally defined, making the recovered shape a natural symbolic description. We present reconstruction experiments involving shapes of various topologies.
ICCV	Real-Time X-Ray Inspection of 3D Defects in Circuit Board Patterns.	Hideaki Doi,Yoko Suzuki,Yasuhiko Hara,Tadashi Iida,Yasuhiro Fujishita,Koichi Karasaki	1995	Real-Time X-Ray Inspection of 3D Defects in Circuit Board Patterns.
ICCV	COSMOS-A Representation Scheme for Free-Form Surfaces.	Chitra Dorai,Anil K. Jain	1995	We address the problem of representing and recognizing arbitrarily curved 3D rigid objects when: the objects may vary in shape and complexity, and no restrictive assumptions are made about the types of surfaces on the object. We propose a new and general surface representation scheme for recognizing objects with free form (sculpted) surfaces from range data. In this scheme, an object is described concisely in terms of maximal surface patches of constant shape index. These maximal patches are mapped onto the unit sphere via their orientations, and aggregated via shape spectral functions. Properties such as surface area, curvedness and connectivity that capture local and global information are also built into the representation. The scheme yields not only a meaningful and rich surface description useful for the recoverability of the object, but also a set of powerful indexing primitives for object matching. We demonstrate the generality and the effectiveness of our scheme using real range images of complex objects. We also present results on the categorization of object views based on a novel shape spectral matching technique.
ICCV	Indexing Visual Representations through the Complexity Map.	Benoit Dubuc,Steven W. Zucker	1995	In differential geometry curves are characterized as mappings from an interval to the plane. In topology curves are characterized as a Hausdorff space with certain countability properties. Neither of these definitions captures the role that curves play in vision, however, in which curves can denote simple objects (such as a straight line), or complicated objects (such as a jumble of string). The difference between these situations is in part a measure of their complexity, and in part a measure of their dimensionality. Note that the map defining such curves is unknown, as is the proper way to represent them. We propose a formal complexity theory of curves appropriate for computational vision in general, and for problems like separating straight lines from jumbles in particular. The theory is applied to the problem of perceptual grouping.
ICCV	Facial Expression Recognition Using a Dynamic Model and Motion Energy.	Irfan A. Essa,Alex Pentland	1995	"Previous efforts at facial expression recognition have been based on the Facial Action Coding System (FACS), a representation developed in order to allow human psychologists to code expression from static facial ""mugshots."" We develop new more accurate representations for facial expression by building a video database of facial expressions and then probabilistically characterizing the facial muscle activation associated with each expression using a detailed physical model of the skin and muscles. This produces a muscle based representation of facial motion, which is then used to recognize facial expressions in two different ways. The first method uses the physics based model directly, by recognizing expressions through comparison of estimated muscle activations. The second method uses the physics based model to generate spatio temporal motion energy templates of the whole face for each different expression. These simple, biologically plausible motion energy ""templates"" are then used for recognition. Both methods show substantially greater accuracy at expression recognition than has been previously achieved."
ICCV	Transfer of Fixation for an Active Stereo Platform via Affine Structure Recovery.	Stuart M. Fairley,Ian D. Reid,David W. Murray	1995	"This paper describes an algorithm for stereo tracking using 3D affine transfer of a body-centred fixation point. Transfer is based on corners detected in the image and matched over time and in stereo. The paper presents a method of basing the transfer on all the available data, providing immunity to noise and poor conditioning. The paper also shows an implementation at video rates on a four axis active camera platform. Graceful degradation in the presence of insufficient data and fixed latency tracking in parallel with the structure calculation provide robust performance. Recovered trajectories are shown in an approximately Euclidean frame while structure transfer is demonstrated by the evolution of the target's convex hull."
ICCV	On the Geometry and Algebra of the Point and Line Correspondences Between N Images.	Olivier D. Faugeras,Bernard Mourrain	1995	We explore the geometric and algebraic relations that exist between correspondences of points and lines in an arbitrary number of images. We propose to use the formalism of the Grassmann-Cayley algebra as the simplest way to make both geometric and algebraic statements in a very synthetic and effective way (i.e. allowing actual computation if needed). We have a fairly complete picture of the situation in the case of points; there are only three types of algebraic relations which are satisfied by the coordinates of the images of a 3-D point: bilinear relations arising when we consider pairs of images among the N and which are the well-known epipolar constraints, trilinear relations arising when we consider triples of images among the N, and quadrilinear relations arising when we consider four-tuples of images among the N. In the case of lines, we show how the traditional perspective projection equation can be suitably generalized and that in the case of three images there exist two independent trilinear relations between the coordinates of the images of a 3-D line.
ICCV	3D-2D Projective Registration of Free-Form Curves and Surfaces.	Jacques Feldmar,Nicholas Ayache,Fabienne Betting	1995	Some medical interventions require knowing the correspondence between an MRI/CT pre-operative image and the actual position of the patient. Examples occur in neurosurgery, radiotherapy, interventional radiology, but also in video surgery (laparoscopy). We present in this article three new techniques for performing this task without artificial markers. We find the 3D-2D projective transformation (composition of a rigid displacement and a perspective projection) which maps a 3D object onto a 2D image of this object. Depending on the object model (curve or surface), and on the 2D image acquisition system (X-Ray, video), the techniques are different but the framework is common. It does not depend on the initial relative positions of the objects and deals with the occlusions and the outliers. Results are presented on real medical data to demonstrate the validity of our approach.
ICCV	Global Rigidity Constraints in Image Displacement Fields.	Cornelia Fermüller,Yiannis Aloimonos	1995	Image displacement fields-optical flow fields, stereo disparity fields, normal flow fields-due to rigid motion possess a global geometric structure which is independent of the scene in view. Motion vectors of certain lengths and directions are constrained to lie on the imaging surface at particular loci whose location and form depends solely on the 3D motion parameters. If optical flow fields or stereo disparity fields are considered, then equal vectors are shown to lie on conic sections. Similarly, for normal motion fields, equal vectors lie within regions whose boundaries also constitute conics. By studying various properties of these curves and regions and their relationships, a characterization of the structure of rigid motion fields is given. The goal of this paper is to introduce a concept underlying the global structure of image displacement fields. This concept gives rise to various constraints that could form the basis of algorithms for the recovery of visual information from multiple views.
ICCV	Color Constancy in Diagonal Chromaticity Space.	Graham D. Finlayson	1995	"Simple constraints on the sets of possible surface reflectances and illuminants are exploited in a new color constancy algorithm that builds upon Forsyth's (1990) theory of color constancy. The goal defined for a color constancy algorithm is to discount variations in the color and intensity of the incident illumination and thereby extract illumination-independent descriptors of surface colors from images. Forsyth's method is based on two constraints: first, the surface colors under a canonical illuminant all fall within an established maximal convex gamut of possible colors and second that a diagonal matrix accurately maps colors between illuminants. These constraints taken together turn out to be very effective in solving for color constancy; however, other strong assumptions about the scenes are required for the method to work-the illumination must be uniform, the surfaces must be planar, and there can be no specularities. We show that these restrictions are necessary only because Forsyth sets out to recover the intensity of descriptors. At the outset we abandon 3-dimensional descriptor recovery in favor of recovering only orientation (i.e. 2 dimensions). Intensity information is factored out of the problem by mapping 3-dimensional (r, g, b) camera responses onto 2-dimensional chromaticities; specifically (r/b, g/b). We show that this ""diagonal chromaticity space"" has two important properties: first, gamut convexity is preserved and second illumination change is still described by a diagonal matrix. It follows that Forsyth's algorithm can be directly applied to the recover chromaticity descriptors and from these the 3D descriptor orientations can be derived. The basic algorithm is then extended to include a maximal gamut constraint on the set of illuminants that is analogous to the gamut constraint on surface colors. The diagonal chromaticity space facilitates the expression of the illumination constraint in the algorithm. Tests on real images show that the algorithm provides good color constancy."
ICCV	Color Constancy under Varying Illumination.	Graham D. Finlayson,Brian V. Funt,Kobus Barnard	1995	Illumination is rarely constant in intensity or color throughout a scene. Multiple light sources with different spectra-sun and sky, direct and interreflected light-are the norm. Nonetheless, almost all color constancy algorithms assume that the spectrum of the incident illumination remains constant across the scene. We assume the converse, that illumination does vary, in developing a new algorithm for color constancy. Rather than creating difficulties, varying illumination is in fact a very powerful constraint. Indeed tests of our algorithm using real images of an office scene show excellent results.
ICCV	Bayesian Decision Theory, the Maximum Local Mass Estimate, and Color Constancy.	William T. Freeman,David H. Brainard	1995	Vision algorithms are often developed in a Bayesian framework. Two estimators are commonly used: maximum a posteriori (MAP), and minimum mean squared error (MMSE). We argue that neither is appropriate for perception problems. The MAP estimator makes insufficient use of structure in the posterior probability. The squared error penalty of the MMSE estimator does not reflect typical penalties. We describe a new estimator, which we call maximum local mass (MLM) [10, 26, 65], which integrates the local probability density. The MLM method is sensitive to local structure of the posterior probability, which MAP is not. The new method uses an optimality criterion that is appropriate for perception tasks: it finds the most probable approximately correct answer. For the case of low observation noise, we provide an efficient approximation. We apply this new estimator to color constancy. An unknown illuminant falls on surfaces of unknown colors. We seek to estimate both the illuminant spectrum and the surface spectra from photosensor responses which depend on the product of the unknown spectra. In simulations, we show that the MLM method performs better than the MAP estimator, and better than two standard color constancy algorithms. The MLM method may prove useful in other vision problems as well.
ICCV	Polymorphic Grouping for Image Segmentation.	Claudia Fuchs,Wolfgang Förstner	1995	The paper describes a new approach to image segmentation. It accepts the inherent deficiencies occuring when extracting low-level features and when dealing with the complexity of real scenes. Image segmentation therefore is understood as deriving a rich symbolic description useful for tasks such as stereo or object recognition in outdoor scenes. The approach is based on a polymorphic scheme for simultaneously extracting points, lines and segments in a topologically consistent manner, together with their mutual relations derived from the feature adjacency graph (FAG) thereby performing several grouping steps which gradually use more and more specific domain knowledge to achieve an optimal image description. The heart of the approach is (1) a detailed analysis of the FAG and (2) a robust estimation for validating the found geometric hypotheses. The analysis of the FAG, derived from the exoskeleton of the features, allows to detect inconsistencies of the extracted features with the ideal image model, a cell-complex. The FAG is used for finding hypotheses about incidence relations and geometric hypotheses, such as collinearity or parallelity, also between non-neighbored points and lines. The M-type robust estimation is used for simultaneously eliminating wrong hypotheses on geometric relationships. It uses a new argument for the weighting function.
ICCV	Combining Color and Geometry for the Active, Visual Recognition of Shadows.	Gareth Funka-Lea,Ruzena Bajcsy	1995	"Shadows are a frequent occurrence, but they cannot be infallibly recognized until a scene's geometry and lighting are known. We present a number of cues which together strongly suggest the identification of a shadow and which can be examined with low cost. The techniques are: a color image segmentation method that recovers single material surfaces as single image regions irregardless of the surface partially in shadow, a method to recover the penumbra and umbra of shadow; a method for determining whether some object could be obstructing a light source. The last cue requires the examination of well understood shadows in the scene. Our observer is equipped with an extendable probe for casting its own shadows. Actively obtained shadows allow the observer to experimentally determine the location of the light sources in the scene. The system has been tested both indoors and out."
ICCV	Surface Orientation and Curvature from Differential Texture Distortion.	Jonas Gårding	1995	"A unified differential geometric framework for estimation of local surface shape and orientation from projective texture distortion is proposed, based on a differential version of the texture stationarity assumption introduced by Malik and Rosenholtz. This framework allows the information content of the gradient of any texture descriptor defined in a local coordinate frame to be characterized in a very compact form. The analysis encompasses both full affine texture descriptors and the classical ""texture gradients"". For estimation of local surface orientation and curvature from uncertain observations of affine texture distortion, the proposed framework allows the dimensionality of the search space to be reduced from five to one."
ICCV	Optical Flow and Deformable Objects.	Andrea Giachetti,Vincent Torre	1995	When a plane undergoes a deformation that can be represented by a planar linear vector field, the projected vector field on the image plane of an optical device is at most quadratic. This 2D motion field has one singular point, with eigenvalues identical to those of the singular point describing the deformation. As a consequence, the nature of the singular point of the deformation is a projective invariant. When the plane moves and experiences a linear deformation at the same time, the associated 2D motion field is still quadratic with at most 3 singular points. In the case of a normal rototranslation, i.e. when the angular velocity is normal to the plane, and of a linear deformation, the 2D motion field has at most one singular point and substantial information on the rigid motion and on the deformation can be recovered from it. Experiments with simulated deformations and real deformable objects show that the proposed analysis can provide accurate results and information on more general 3D deformations.
ICCV	Calibration-Free Visual Control Using Projective Invariance.	Gregory D. Hager	1995	Much of the previous work on hand-eye coordination has emphasized the reconstructive aspects of vision. Recently, techniques that avoid explicit reconstruction by placing visual feedback into a control loop have been developed. When properly defined, these methods lead to calibration insensitive hand-eye coordination. Recent work on projective geometry as applied to vision is used to extend this paradigm in two ways. First, it is shown how results from projective geometry can be used to perform online calibration. Second, results on projective invariance are used to define setpoints for visual control that are independent of viewing location. These ideas are illustrated through a number of examples and have been tested on an implemented system.
ICCV	A Linear Method for Reconstruction from Lines and Points.	Richard I. Hartley	1995	Discusses the basic role of the trifocal tensor in scene reconstruction. This 3/spl times/3/spl times/3 tensor plays a role in the analysis of scenes from three views analogous to the role played by the fundamental matrix in the two-view case. In particular, the trifocal tensor maybe computed by a linear algorithm from a set of 13 line correspondences in three views. It is further shown in this paper to be essentially identical to a set of coefficients introduced by Shashua (1994) to effect point transfer in the three-view case. This observation means that the 13-line algorithm may be extended to allow for the computation of the trifocal tensor given any mixture of sufficiently many line and point correspondences. From the trifocal tensor, the camera image matrices may be computed, and the scene may be reconstructed. For unrelated uncalibrated cameras, this reconstruction is unique up to projectivity. Thus, projective reconstruction of a set of lines and points may be reconstructed linearly from three views.
ICCV	In Defence of the 8-Point Algorithm.	Richard I. Hartley	1995	The fundamental matrix is a basic tool in the analysis of scenes taken with two uncalibrated cameras, and the 8 point algorithm is a frequently cited method for computing the fundamental matrix from a set of 8 or more point matches. It has the advantage of simplicity of implementation. The prevailing view is, however, that it is extremely susceptible to noise and hence virtually useless for most purposes. The paper challenges that view, by showing that by preceding the algorithm with a very simple normalization (translation and scaling) of the coordinates of the matched points, results are obtained comparable with the best iterative algorithms. This improved performance is justified by theory and verified by extensive experiments on real images.
ICCV	Segmented Shape Descriptions from 3-View Stereo.	Parag Havaldar,Gérard G. Medioni	1995	"We address the recovery of segmented, 3-D descriptions of an object from intensity images. We use three views of an object from slightly different viewpoints as our input. For each image we extract a hierarchy of groups based on proximity, parallelism and symmetry in a robust manner. The groups in the three images are matched by computing the epipolar geometry. For each set of matched groups from the three images, we then label the contours of the groups as ""true"" or ""limb"" edges. Using the information about groups, the label associated with their contours and projective properties of subclasses of Generalized Cylinders, we infer the 3-D structure of these groups. The proposed method not only allows robust shape recovery but also produces segmented parts. Our approach can also deal with groups generated as a result of texture or shadows on the object. We present results on real images of moderately complex objects."
ICCV	The Illumination-Invariant Recognition of Texture in Color Texture.	Glenn Healey,Lizhi Wang	1995	The Illumination-Invariant Recognition of Texture in Color Texture.
ICCV	Reconstruction from Image Sequences by Means of Relative Depths.	Anders Heyden	1995	The paper deals with the problem of reconstructing the locations of n points in space from m different images without camera calibration. It shows how these problems can be put into a similar theoretical framework. A new concept, the reduced fundamental matrix, is introduced. It contains just 4 parameters and can be used to predict locations of points in the images and to make reconstruction. We also introduce the concept of reduced fundamental tensor which describes the relations between points in 3 images. It has 15 components and depends on 9 parameters. Necessary and sufficient conditions for a tensor to be a reduced fundamental tensor are derived. This framework can be generalised to a sequence of images. The dependencies between the different representations are investigated. Furthermore a canonical form of the camera matrices in a sequence are presented.
ICCV	Object Pose: Links Between Paraperspective and Perspective.	Radu Horaud,Stéphane Christy,Fadi Dornaika,Bart Lamiroy	1995	D.F. Dementhon and L.S. Davis (1995) proposed a method for determining the pose of a 3D object with respect to a camera from 3D to 2D point correspondences. The method consists of iteratively improving the pose computed with a weak perspective camera model to converge at the limit, to a pose estimation computed with a perspective camera model. We show that the method of Dementhon and Davis can be extended to paraperspective. The iterative paraperspective pose algorithm that we describe in detail has interesting properties both in terms of speed and rate of convergence. Moreover, we introduce a simple way of taking into account the orthogonality constraint associated with the rotation matrix and we define the optimal experimental setup to be used in the presence of camera calibration errors.
ICCV	Statistical Learning, Localization, and Identification of Objects.	Joachim Hornegger,Heinrich Niemann	1995	This work describes a statistical approach to deal with learning and recognition problems in the field of computer vision. An abstract theoretical framework is provided, which is suitable for automatic model generation from examples, identification, and localization of objects. Both, the learning and localization stage are formalized as parameter estimation tasks. The statistical learning phase is unsupervised with respect to the matching of model and scene features. The general mathematical description yields algorithms which can even treat parameter estimation problems from projected data. The experiments show that this probabilistic approach is suitable for solving 2D and 3D object recognition problems using grey-level images. The method can also be applied to 3D image processing issues using range images, i.e. 3D input data.
ICCV	Computing Visual Correspondence: Incorporating the Probability of a False Match.	Daniel P. Huttenlocher,Eric W. Jaquith	1995	We describe a method for computing visual correspondence which employs a formal model of the probability of a false match. This model estimates the chance that the best match for each point could have occurred at random. The model is effective at identifying points in one image for which there is no corresponding point in the other image, as occurs at depth boundaries in stereo and at motion boundaries in optical flow. More generally, the model can be used to identify points where the best match is of poor quality, as occurs in regions of uniform texture. We describe the similarity measure used in the method and present the formal model of a false match. We also show examples of using the method to compute stereo disparity.
ICCV	Closed-World Tracking.	Stephen S. Intille,Aaron F. Bobick	1995	"A new approach to tracking weakly modeled objects in a semantically rich domain is presented. We define a closed-world as a space-time region of an image sequence in which the complete taxonomy of objects is known, and in which each pixel should be explained as belonging to one of those objects. Given contextual object information, context-specific features can be dynamically selected as the basis for tracking. A context-specific feature is one that has been chosen based upon the context to maximize the chance of successful tracking between frames. Our work is motivated by the goal of video annotation-the semi-automatic generation of symbolic descriptions of action taking place in a contextually-rich dynamic scene. We describe how contextual knowledge in the ""football domain"" can be applied to closed-world football player tracking and present the details of our implementation. We include tracking results based on hundreds of images that demonstrate the wide range of tracking situations the algorithm successfully handles as well as a few examples of where the algorithm fails."
ICCV	Mosaic Based Representations of Video Sequences and Their Applications.	Michal Irani,P. Anandan,Steven C. Hsu	1995	Recently, there has been a growing interest in the use of mosaic images to represent the information contained in video sequences. The paper systematically investigates how to go beyond thinking of the mosaic simply as a visualization device, but rather as a basis for efficient representation of video sequences. We describe two different types of mosaics called the static and the dynamic mosaic that are suitable for different needs and scenarios. We discuss a series of extensions to these basic mosaics to provide representations at multiple spatial and temporal resolutions and to handle 3D scene information. We describe techniques for the basic elements of the mosaic construction process, namely alignment, integration, and residual analysis. We describe several applications of mosaic representations including video compression, enhancement, enhanced visualization, and other applications in video indexing, search, and manipulation.
ICCV	Saliency Maps and Attention Selection in Scale and Spatial Coordinates: An Information Theoretic Approach.	Martin Jägersand	1995	Information measures with respect to spatial locations and scales of objects in an image are important to image processing and interpretation. It allows us to focus attention on relevant data, saving effort and reducing false positives. In particular, the information content of a man-made scene is typically confined to a small set of scales. We devise a scale space based measure of image information. Kullback contrasts between successive resolution lengths gives the differential information gain. Experiments show that this measure gives a clear indication of characteristic lengths in a variety of real world images and is superior to power spectrum based measurements. Decomposing the expected information gain into spatial coordinates gives us a saliency map for use by an attention selector. We combine the scale and spatial decompositions into a single information measure, giving both the spatial extent and scale range of interest. The information measure has an efficient implementation, and thus can be used routinely in early vision processing.
ICCV	Model-Based Matching of Line Drawings by Linear Combinations of Prototypes.	Michael J. Jones,Tomaso Poggio	1995	"We describe a technique for finding pixelwise correspondences between two images by using models of objects of the same class to guide the search. The object models are ""learned"" from example images (also called prototypes) of an object class. The models consist of a linear combination of prototypes. The flow fields giving pixelwise correspondences between a base prototype and each of the other prototypes must be given. A novel image of an object of the same class is matched to a model by minimizing an error between the novel image and the current guess for the closest model image. Currently, the algorithm applies to line drawings of objects. An extension to real grey level images is discussed."
ICCV	Structure and Motion Estimation from Dynamic Silhouettes under Perspective Projection.	Tanuja Joshi,Narendra Ahuja,Jean Ponce	1995	Addresses the problem of estimating the structure and motion of a smooth curved object from its silhouettes observed over time by a trinocular stereo rig under perspective projection. We first construct a model for the local structure along the silhouette for each frame in the temporal sequence. Successive local models are then integrated into a global surface description by estimating the motion between successive time instants. The algorithm tracks certain surface features (parabolic points) and image features (silhouette inflections and frontier points) which are used to bootstrap the motion estimation process. The entire silhouette along with the reconstructed local structure are then used to refine the initial motion estimate. We have implemented the proposed approach and report results on real images.
ICCV	3D Human Body Model Acquisition from Multiple Views.	Ioannis A. Kakadiaris,Dimitris N. Metaxas	1995	"We present a novel motion-based approach for the part determination and shape estimation of a human's body parts. The novelty of the technique is that neither a prior model of the human body is employed nor prior body part segmentation is assumed. We present a human body part identification strategy (HBPIS) that recovers all the body parts of a moving human based on the spatiotemporal analysis of its deforming silhouette. We formalize the process of simultaneous part determination and 2D shape estimation by employing the supervisory control theory of discrete event systems. In addition, in order to acquire the 3D shape of the body parts, we present a new algorithm which selectively integrates the (segmented by the HBPIS) apparent contours, from three mutually orthogonal views. The effectiveness of the approach is demonstrated through a series of experiments, where a subject performs a set of movements according to a protocol that reveals the structure of the human body."
ICCV	A Robot System that Observes and Replicates Grasping Tasks.	Sing Bang Kang,Katsushi Ikeuchi	1995	To alleviate the problem of overwhelming complexity in grasp synthesis and path planning associated with robot task planning, we adopt the approach of teaching the robot by demonstrating in front of it. The system has four components: the observation system, the grasping task recognition module, the task translator and the robot system. The observation system comprises an active multibaseline stereo system and a dataglove. The data stream recorded is then used to track object motion; this paper illustrates how complimentary sensory data can be used for this purpose. The data stream is also interpreted by the grasping task recognition module, which produces higher levels of abstraction to describe both the motion and actions taken in the task. The resulting information are provided to the task translator which creates commands for the robot system to replicate the observed task. In this paper we describe how these components work with special emphasis on the observation system. The robot system that we use to perform the grasping tasks comprises the PUMA 560 arm and the Utah/MIT hand.
ICCV	A Multibaseline Stereo System with Active Illumination and Real-Time Image Acquisition.	Sing Bing Kang,Jon A. Webb,C. Lawrence Zitnick,Takeo Kanade	1995	We describe our implementation of a parallel depth recovery scheme for a four-camera multibaseline stereo in a convergent configuration. Our system is capable of image capture at video rate. This is critical in applications that require three-dimensional tracking. We obtain dense stereo depth data by projecting a light pattern of frequency modulated sinusoidally varying intensity onto the scene, thus increasing the local discriminability at each pixel and facilitating matches. In addition, we make most of the camera view areas by converging them at a volume of interest. Results show that we are able to extract stereo depth data that are, on the average, less than 1 mm in error at distances between 1.5 to 3.5 m away from the cameras.
ICCV	Unsupervised Parallel Image Classificiation Using a Hierarchical Markovian Model.	Zoltan Kato,Josiane Zerubia,Marc Berthod	1995	The paper deals with the problem of unsupervised classification of images modeled by Markov random fields (MRF). If the model parameters are known then we have various methods to solve the segmentation problem (simulated annealing, ICM, etc...). However, when they are not known, the problem becomes more difficult. One has to estimate the hidden label field parameters from the only observable image. Our approach consists of extending a recent iterative method of estimation, called iterative conditional estimation (ICE) to a hierarchical Markovian model. The idea resembles the estimation-maximization (EM) algorithm as we recursively look at the maximum a posteriori (MAP) estimate of the label field given the estimated parameters then we look at the maximum likelihood (ML) estimate of the parameters given a tentative labeling obtained at the previous step. We propose unsupervised image classification algorithms using a hierarchical model. The only parameter supposed to be known is the number of regions, all the other parameters are estimated. The presented algorithms have been implemented on a Connection Machine CM200. Comparative tests have been done on noisy synthetic and real images (remote sensing).
ICCV	Annular Symmetry Operators: A Method for Locating and Describing Objects.	Michael F. Kelly,Martin D. Levine	1995	"We present a machine vision system in which segmentation is computed in conjunction with a structural description of objects in the scene. It is assumed that contrast edges capture all relevant object information. The principles which dictate how edge features are grouped to infer objects are based upon detecting SYMMETRICAL ENCLOSING edge configurations. These are detected using ANNULAR OPERATORS applied at multiple scales to edge data which have been extracted at multiple scales from a gray level image. The subsequent grouping of symmetry points results in a set of PARTS which make it possible to identify the LOCATION of objects within an image. These parts are used as a basis for constructing coarse graph-based DESCRIPTORS for the PERCEPTUALLY SIGNIFICANT objects found in the scene. Results are presented to illustrate the method's performance on several images."
ICCV	Gradient Flows and Geometric Active Contour Models.	Satyanad Kichenassamy,Arun Kumar,Peter J. Olver,Allen Tannenbaum,Anthony J. Yezzi	1995	In this paper, we analyze the geometric active contour models discussed previously from a curve evolution point of view and propose some modifications based on gradient flows relative to certain new feature-based Riemannian metrics. This leads to a novel snake paradigm in which the feature of interest may be considered to lie at the bottom of a potential well. Thus the snake is attracted very naturally and efficiently to the desired feature. Moreover, we consider some 3-D active surface models based on these ideas.
ICCV	Direct Estimation of Affine Image Deformations Using Visual Front-End Operations with Automatic Scale Selection.	Tony Lindeberg	1995	This article deals with the problem of estimating deformations of brightness patterns using visual front-end operations. Estimating such deformations constitutes an important subtask in several computer vision problems relating to image correspondence and shape estimation. The following subjects are treated: The problem of decomposing affine flow fields into simpler components is analysed in detail. A canonical parametrization is presented based on singular value decomposition, which naturally separates the rotationally invariant components of the flow field from the rotationally variant ones. A novel mechanism is presented for automatic selection of scale levels when estimating local affine deformations. This mechanism is expressed within a multiscale framework where disparity estimates are computed in a hierarchical coarse-to-fine manner and corrected using iterative techniques. Then, deformation estimates are selected from the scales that minimize a certain normalized residual over scales. Finally, the descriptors so obtained serve as initial data for computing refined estimates of the local deformations.
ICCV	3D Surface Reconstruction from Stereoscopic Image Sequences.	Reinhard Koch	1995	3D Surface Reconstruction from Stereoscopic Image Sequences.
ICCV	Towards a Unified IU Environment: Coordination of Existing IU Tools with the IUE.	Charles A. Kohl,Jeffrey J. Hunter,Cynthia L. Loiselle	1995	"Progress in the field of computer vision/image understanding (IU) has long been hampered by the lack of standard software environments for research and application development. The Image Understanding Environment (IUE) is being implemented to provide a freely-distributed standard software environment that is appropriate for a wide range of research and development activities. Nevertheless, we recognize that the IUE will not serve everyone's individual needs and that conversion to the IUE could be an expensive process for many sites. Therefore, we are investing significant effort to make the IUE an open environment that can work in coordination with existing IU systems and tools. This paper introduces the work being performed at Amerinex AI to address the processing and data representation factors involved in this coordination."
ICCV	Multiscale Detection of Curvilinear Structures in 2D and 3D Image Data.	Thomas Koller,Guido Gerig,Gábor Székely,Daniel Dettwiler	1995	Multiscale Detection of Curvilinear Structures in 2D and 3D Image Data.
ICCV	3D Pose Estimation by Fitting Image Gradients Directly to Polyhedral Models.	Henner Kollnig,Hans-Hellmut Nagel	1995	Addresses the problem of pose estimation and tracking of vehicles in image sequences from traffic scenes recorded by a stationary camera. In a new algorithm, the vehicle pose is estimated by directly fitting image gradients to polyhedral vehicle models without an edge segment extraction process. The new approach is significantly more robust than approaches that rely on feature extraction because the new approach exploits more information from the image data. We can track vehicles that are partially occluded by textured objects, e.g. foliage, where classical approaches based on edge segment extraction fail. Results from various experiments with real-world traffic scenes are presented.
ICCV	Gabor Wavelets for 3-D Object Recognition.	Xing Wu,Bir Bhanu	1995	This paper presents a model-based object recognition approach that uses a hierarchical Gabor wavelet representation. The key idea is to use magnitude, phase and frequency measures of Gabor wavelet representation in an innovative flexible matching approach that can provide robust recognition. A Gabor grid a topology-preserving map, efficiently encodes both signal energy and structural information of an object in a sparse multi-resolution representation. The Gabor grid subsamples the Gabor wavelet decomposition of an object model and is deformed to allow the indexed object model match with the image data. Flexible matching between the model and the image minimizes a cost function based on local similarity and geometric distortion of the Gabor grid. Grid erosion and repairing is performed whenever a collapsed grid, due to object occlusion, is detected. The results on infrared imagery are presented. Where objects undergo rotation, translation, scale, occlusion and aspect variations under changing environmental conditions.
ICCV	Texture Segmentation and Shape in the Same Image.	John Krumm,Steven A. Shafer	1995	"Uniformly textured surfaces in 3D scenes provide important cues for image understanding. Texture can be used for both segmentation and for 3D shape inference. Unfortunately, virtually all current algorithms are based on assumptions that make it impossible to do texture segmentation and shape-from-texture in the same image. Texture segmentation algorithms rely on an absence of 3D effects that tend to distort the texture. Shape-from-texture algorithms depend on these effects, relying instead on the texture being already segmented. To really understand texture in images, texture segmentation and shape-from-texture must be viewed as a combined problem to be solved simultaneously. We present a solution to this problem with a region-growing algorithm that explicitly accounts for perspective distortions of otherwise uniform texture. We use the image spectrogram to compute local surface normals, which are in turn used to ""frontalize"" the texture. These frontalized texture patches are then subjected to a region-growing algorithm based on similarity in the local frequency domain and a minimum description length criteria. We show results of our algorithm on real texture images taken in the lab and outdoors."
ICCV	Vision Based Hand Modeling and Tracking for Virtual Teleconferencing and Telecollaboration.	James J. Kuch,Thomas S. Huang	1995	"The authors present a hand model that simultaneously satisfies both the synthesis and analysis requirements of model based compression. The model can be fitted to any person's hand and can be done using a single camera. Once the model is fitted to a real human hand, it is then used in several tracking scenarios in order to verify its effectiveness. With successful tracking achieved, the model is ready to be incorporated into a virtual environment or model based compression scheme such as sign language communication over telephone lines or virtual teleconferences over computer networks at very low bit rates and at very high image quality."
ICCV	Reflectance Function Estimation and Shape Recovery from Image Sequence of a Rotating Object.	Jiping Lu,James J. Little	1995	We describe a technique for surface recovery of a rotating object illuminated under a collinear light source (where the light source lies on or near the optical axis). We show that the surface reflectance function can be directly estimated from the image sequence without any assumption on the reflectance property of the object surface. From the image sequence, the 3D locations of some singular surface points are calculated and their brightness values are extracted for the estimation of the reflectance function. We also show that the surface can be recovered by using shading information in two images of the rotating object. Iteratively using the first-order Taylor series approximation and the estimated reflectance function, the depth and orientation of the surface can be recovered simultaneously. The experimental results on real image sequences of both matte and specular surfaces demonstrate that the technique is feasible and robust.
ICCV	An Integrated Stereo-Based Approach to Automatic Vehicle Guidance.	Quang-Tuan Luong,Joseph Weber,Daphne Koller,Jitendra Malik	1995	Proposes a new approach for vision-based longitudinal and lateral vehicle control. The novel feature of this approach is the use of binocular vision. We integrate two modules consisting of a new, domain-specific, efficient binocular stereo algorithm, and a lane marker detection algorithm, and show that the integration results in a improved performance for each of the modules. Longitudinal control is supported by detecting and measuring the distances to leading vehicles using binocular stereo. The knowledge of the camera geometry with respect to the locally planar road is used to map the images of the road plane in the two camera views into alignment. This allows us to separate image features into those lying in the road plane, e.g. lane markers, and those due to other objects which are dynamically integrated into an obstacle map. Therefore, in contrast with the previous work, we can cope with the difficulties arising from occlusion of lane markers by other vehicles. The detection and measurement of the lane markers provides us with the positional parameters and the road curvature which are needed for lateral vehicle control. Moreover, this information is also used to update the camera geometry with respect to the road, therefore allowing us to cope with the problem of vibrations and road inclination to obtain consistent results from binocular stereo.
ICCV	Affine Surface Reconstruction by Purposive Viewpoint Control.	Kiriakos N. Kutulakos	1995	"We present an approach for building an affine representation of an unknown curved object viewed under orthographic projection from images of its occluding contour. It is based on the observation that the projection of a point on a curved, featureless surface can be computed along a special viewing direction that does not belong to the point's tangent plane. We show that by circumnavigating the object on the tangent plane of selected surface points, we can (1) compute two orthogonal projections of every point projecting to the occluding contour during this motion, and (2) compute the affine coordinates of these points. Our approach demonstrates that affine shape of curved objects can be computed directly, i.e., without Euclidean calibration or image velocity and acceleration measurements."
ICCV	Hierarchical Statistical Models for the Fusion of Multiresolution Image Data.	Jean-Marc Laferté,Fabrice Heitz,Patrick Pérez,Eric Fabre	1995	"This paper presents a class of nonlinear hierarchical algorithms for the fusion of multiresolution image data in low-level vision. The approach combines nonlinear causal Markov models defined on hierarchical graph structures, with standard bayesian estimation theory. Two random processes defined on simple hierarchical graphs (quadtrees or ""ternary graphs"") are introduced to represent the multiresolution observations at hand and the hidden labels to be estimated. An optimal algorithm (inspired from the Viterbi algorithm) is developed to compute the bayesian estimates on the hierarchical graph structures. Estimates are obtained within two passes on the graph structure. This algorithm is non-iterative and yields a per pixel computational complexity which is independent of image size. This approach is compared to the multiscale algorithm proposed by (Bouman et al., 1994) for single-resolution image segmentation (that we have extended for multiresolution data fusion)."
ICCV	A Unified Approach to Coding and Interpreting Face Images.	Andreas Lanitis,Christopher J. Taylor,Timothy F. Cootes	1995	Face images are difficult to interpret because they are highly variable. Sources of variability include individual appearance, 3D pose, facial expression and lighting. We describe a compact parametrised model of facial appearance which takes into account all these sources of variability. The model represents both shape and grey-level appearance and is created by performing a statistical analysis over a training set of face images. A robust multi-resolution search algorithm is used to fit the model to faces in new images. This allows the main facial features to be located and a set of shape and grey-level appearance parameters to be recovered. A good approximation to a given face can be reconstructed using less than 100 of these parameters. This representation can be used for tasks such as image coding, person identification, pose recovery, gender recognition and expression recognition. The system performs well on all the tasks listed above.
ICCV	Determining Wet Surfaces from Dry.	Howard B. Mall Jr.,Niels da Vitoria Lobo	1995	Wet surfaces are ubiquitous in our visual experience. Autonomous machines with vision systems will need to identify wet surfaces from dry. Wet surfaces (especially rough, absorbent ones) appear darker when wet. This paper presents the Lekner and Dorf (1988) model for describing the darkening caused by wetting. We explain how to use this optics model to transform intensity values of a region of an image to make that region appear wet. We also show how the model can be reversed in order to make a wet part of an image appear dry. It is also shown that this technique can be used to identify wet regions. This identification is contrasted with darkening caused by shadows. Comparisons of the gray-level histograms of these real images show the validity of this approach for distinguishing wet surfaces from dry.
ICCV	Finding Faces in Cluttered Scenes Using Labeled Random Graph Matching.	Thomas K. Leung,Michael C. Burl,Pietro Perona	1995	Finding Faces in Cluttered Scenes Using Labeled Random Graph Matching.
ICCV	On Representation and Matching of Multi-Coloured Objects.	Jiri Matas,R. Marik,Josef Kittler	1995	A new representation for objects with multiple colours-the colour adjacency graph (CAG)-is proposed. Each node of the CAG represents a single chromatic component of the image defined as a set of pixels forming a unimodal cluster in the chromatic scattergram. Edges encode information about adjacency of colour components and their reflectance ratio. The CAG is related to both the histogram and region adjacency graph representations. It is shown to be preserving and combining the best features of these two approaches while avoiding their drawbacks. The proposed approach is tested on a range of difficult object recognition and localisation problems involving complex imagery of non-rigid 3D objects under varied viewing conditions with excellent results.
ICCV	Automatic Recognition of Human Facial Expressions.	Katsuhiro Matsuno,Chil-Woo Lee,Satoshi Kimura,Saburo Tsuji	1995	The paper presents a new idea for detecting an unknown human face in input imagery and recognizing his/her facial expression represented in the deformation of the two dimensional net, called potential net. The method deals with the facial information, faceness and expressions, as an overall pattern of the net activated by edges in a single input image of face, rather than from changes in the shape of the facial organs or their geometrical relationships. We build models of facial expressions from the deformation patterns in the potential net for face images in the training set of different expressions and then project them into emotion space. Expression of an unknown subject can be recognized from the projection of the net for the image into the emotion space. The potential net is further used to model the common human face. The mosaic method representing energy in the net is used as a template for finding candidates for the face area and the candidates are verified their faceness by projecting them into emotion space in order to select the finalist. Precise location of the face is determined by the histogram analysis of vertical and horizontal projections of edges.
ICCV	A Recursive Filter for Phase Velocity Assisted Shape-Based Tracking of Cardiac Non-Rigid Motion.	John C. McEachen II,François G. Meyer,R. Todd Constable,Arye Nehorai,James S. Duncan	1995	"A framework for tracking pointwise periodic non-rigid motion of the heart's left ventricular (LV) wall is presented which incorporates information from two different magnetic resonance imaging (MRI) techniques. New developments in phase-contrast cine MR imaging have produced spatial maps of instantaneous velocity that heave proven accuracy within the myocardium, or wall, of the heart. This information is combined with shape-based matching techniques to provide improved estimates of trajectories, especially in regions where shape information is limited. These raw trajectories act as input to a recursive least squares (RLS) filter which applies the constraints of temporal periodicity and spatial smoothness for the final estimate. The results of the RLS filter are compared with the motion of actual implanted markers. Comparisons are also made between exclusively shape-based filtered and phase-contrast enhanced trajectory estimates using both phantom and actual canine heart MR images."
ICCV	Topologically Adaptable Snakes.	Tim McInerney,Demetri Terzopoulos	1995	The paper presents a typologically adaptable snakes model for image segmentation and object representation. The model is embedded in the framework of domain subdivision using simplicial decomposition. This framework extends the geometric and topological adaptability of snakes while retaining all of the features of traditional snakes, such as user interaction, and overcoming many of the limitations of traditional snakes. By superposing a simplicial grid over the image domain and using this grid to iteratively reparameterize the deforming snakes model, the model is able to flow into complex shapes, even shapes with significant protrusions or branches, and to dynamically change topology as necessitated by the data. Snakes can be created and can split into multiple parts or seamlessly merge into other snakes. The model can also be easily converted to and from the traditional parametric snakes model representation. We apply a 2D model to various synthetic and real images in order to segment objects with complicated shapes and topologies.
ICCV	A Unifying Framework for Structure and Motion Recovery from Image Sequences.	Philip F. McLauchlan,David W. Murray	1995	The paper proposes a statistical framework that enables 3D structure and motion to be computed optimally from an image sequence, on the assumption that feature measurement errors are independent and Gaussian distributed. The analysis and results demonstrate that computing both camera/scene motion and 3D structure is essential to computing either with any accuracy. Having computed optimal estimates of structure and motion over a small number of initial images, a recursive version of the algorithm (previously reported) recomputes sub optimal estimates given new image data. The algorithm is designed explicitly for real time implementation, and the complexity is proportional to the number of tracked features. 3D projective, affine and Euclidean models of structure and motion recovery have been implemented, incorporating both point and line features into the computation. The framework can handle any feature type and camera model that may be encapsulated as a projection equation from scene to image.
ICCV	Rigidity Checking of 3D Point Correspondences under Perspective Projection.	Daniel P. McReynolds,David G. Lowe	1995	"An algorithm is described which rapidly verifies the potential rigidity of three dimensional point correspondences from a pair of two dimensional views under perspective projection. The output of the algorithm is a simple yes or no answer to the question ``Could these corresponding points from two views be the projection of a rigid configuration?'''' Potential applications include 3D object recognition from a single previous view and correspondence matching for stereo or motion over widely separated views. Our analysis begins with the observation that it is often the case that two views cannot provide an accurate structure-from-motion estimate because of ambiguity and ill- conditioning. However, it is argued that an accurate yes/no answer to the rigidity question is possible and experimental results support this assertion with as few as six pairs of corresponding points over a wide range of scene structures and viewing geometries. Rigidity checking verifies point correspondences by using 3D recovery equations as a matching condition. The proposed algorithm improves upon other methods that fall under this approach because it works with as few as six corresponding points under full perspective projection, handles correspondences from widely separated views, makes full use of the disparity of the correspondences, and is integrated with a linear algorithm for 3D recovery due to Kontsevich. The rigidity decision is based on the residual error of an integrated pair of linear and nonlinear structure-from-motion estimators. Results are given for experiments with synthetic and real image data. A complete implementation of this algorithm is being made publicly available."
ICCV	Task-Oriented Generation of Visual Sensing Strategies.	Jun Miura,Katsushi Ikeuchi	1995	In vision-guided robotic operations, vision is used for extracting necessary information for achieving the task. Since visual sensing is usually performed with limited resources, visual sensing strategies should be planned so that only necessary information is obtained efficiently. This paper describes a method of systematically generating visual sensing strategies based on knowledge of the task to be performed. The generation of the appropriate visual sensing strategy entails knowing what information to extract, where to get it, and how to get it. This is facilitated by the knowledge of the task, which describes what objects are involved in the operation, and how they are assembled. Our method has been implemented using a laser range finder as the sensor. Experimental results show the feasibility of the method, and point out the importance of task-oriented evaluation of visual sensing strategies.
ICCV	Probabilistic Visual Learning for Object Detection.	Baback Moghaddam,Alex Pentland	1995	We present an unsupervised technique for visual learning which is based on density estimation in high-dimensional spaces using an eigenspace decomposition. Two types of density estimates are derived for modeling the training data: a multivariate Gaussian (for a unimodal distributions) and a multivariate Mixture-of-Gaussians model (for multimodal distributions). These probability densities are then used to formulate a maximum-likelihood estimation framework for visual search and target detection for automatic object recognition. This learning technique is tested in experiments with modeling and subsequent detection of human faces and non-rigid objects such as hands.
ICCV	Volumetric Deformable Models with Parameter Functions: A New Approach to the 3D Motion Analysis of the LV from MRI-SPAMM.	Jinah Park,Dimitris N. Metaxas,Leon Axel	1995	"We present a new method for analyzing the 3D motion of the heart's left ventricle (LV) from tagged magnetic resonance imaging (MRI) data. Our technique is based on the development of a new class of volumetric physics-based deformable models whose parameters are functions and can capture the local shape variation of an object. These parameters require no complex post-processing in order to be used by a physician. These volumetric models allow the accurate estimation of the shape and motion of the inner and outer walls of the LV as well as within the walls. We also present a new technique for calculating forces exerted by tagged MRI data to material points of the deformable model. Furthermore, by plotting the variations over time of the extracted LV model parameters from normal heart data we are able to quantitatively analyze and compare the epicardial and endocardial motion."
ICCV	Shape Extraction for Curves Using Geometry-Driven Diffusion and Functional Optimization.	Eric J. Pauwels,Peter Fiddelaers,Luc J. Van Gool	1995	In this paper we show how both geometry-driven diffusion and optimization of the Mumford-Shah functional can be used to develop a type of curve-evolution that is able to preserve salient features of closed curves (such as corners and straight line segments), while simultaneously suppressing noise and irrelevant details. The idea is to characterize the curve by means of its angle-function (i.e. the angle between the tangent and a fixed axis) and to apply the appropriate dynamics to this one-dimensional representation. We show how constrained evolution equations can be used to keep the corresponding curve closed at all times.
ICCV	Validation of 3D Registration Methods Based on Points and Frames.	Xavier Pennec,Jean-Philippe Thirion	1995	Validation of 3D Registration Methods Based on Points and Frames.
ICCV	Determining Facial Expressions in Real-Time.	Yael Moses,David Reynard,Andrew Blake	1995	We suggest an approach to describing and tracking the deformation of facial features. We concentrate on the mouth since its shape is important in detecting emotion. However, we believe that our system could be extended to deal with other facial features. In our system, the mouth is described by a valley contour which is based between the lips. This contour is shown to exist independently of illumination, viewpoint, identity, and expression. We present a real time mouth tracking system that follows this valley. It is shown to be robust to changes in identity, illumination and viewpoint. A simple classification algorithm was found to be sufficient to discriminate between 5 different mouth shapes, with a 100% recognition rate.
ICCV	Matching of 3D Curves Using Semi-Differential Invariants.	Tomás Pajdla,Luc J. Van Gool	1995	Matching of 3D Curves Using Semi-Differential Invariants.
ICCV	Automatic Generation of GRBF Networks for Visual Learning.	Shayan Mukherjee,Shree K. Nayar	1995	Learning can often be viewed as the problem of mapping from an input space to an output space. Examples of these mappings are used to construct a continuous function that approximates given data and generalizes for intermediate instances. Generalized Radial Basis Function (GRBF) networks are used to formulate this approximating function. A novel method is introduced to construct an optimal GRBF network for a given mapping and error bound using the integral wavelet transform. Simple one-dimensional examples are used to demonstrate how the optimal network is superior to one constructed using standard ad hoc optimization techniques. The paper concludes with an application of optimal GRBF networks to object recognition and pose estimation. The results of this application are favorable.
ICCV	Recognizing 3D Objects Using Photometric Invariant.	Kenji Nagao	1995	We describe an efficient algorithm for recognizing 3D objects by combining photometric, and geometric invariants. A photometric property is derived, that is invariant to the changes of illumination and to relative object motion with respect to the camera and/or the lighting source in 3D space. We argue that conventional color constancy algorithms can not be used in the recognition of 3D objects. Further we show recognition does not require a full constancy of colors, rather, it only needs something that remains unchanged under the varying light conditions and poses of the objects. Combining the derived color invariant and the spatial constraints on the object surfaces, we identify corresponding positions in the model and the data space coordinates, using centroid invariance of corresponding groups of feature positions. Tests are given to show the stability and efficiency of our approach to 3D object recognition.
ICCV	Structure and Semi-Fluid Motion Analysis of Stereoscopic Satellite Images for Cloud Tracking.	Kannappan Palaniappan,Chandra Kambhamettu,Frederick Hasler,Dmitry B. Goldgof	1995	Time-varying multispectral observations of clouds from meteorological satellites are used to estimate cloud-top heights (structure) and cloud winds (semi-fluid motion). Stereo image pairs over several time steps were acquired by two geostationary satellites with synchronized scanning instruments. Cloud-top height estimation from these image pairs is performed using an improved automatic stereo analysis algorithm on a massively parallel Maspar computer with 16 K processors. A new category of motion behavior known as semi-fluid motion is described for modeling cloud motions and an automatic algorithm for extracting semi-fluid motion is developed to track cloud winds. The time sequential dense estimates of cloud-top height depth maps in conjunction with intensity data are used to estimate local semi-fluid motion parameters for cloud tracking. Both stereo disparities and motion correspondences are estimated to sub-pixel accuracy. The Interactive Image SpreadSheet (IISS) is a new versatile visualization tool that was enhanced to analyze and visualize the results of the stereo analysis and semi-fluid motion estimation algorithms. Experimental results using time-varying data of the visible channel from two satellites in geosynchronous orbit is presented for the Hurricane Frederic.
ICCV	Results Using Random Field Models for the Segmentation of Color Images of Natural Scenes.	Dileep Kumar Panjwani,Glenn Healey	1995	We present results using a Markov random field color texture model for the unsupervised segmentation of images of outdoor scenes. The color random field model describes textured regions in terms of spatial interaction within color bands and between different color bands. The model is used by a segmentation algorithm based on agglomerative hierarchical clustering. At the heart of the clustering is a step wise optimal merging process that at each iteration maximizes a global performance functional. The test for stopping the clustering is based on changes in the likelihood of the image. We provide experimental results that demonstrate the performance of the segmentation algorithm on color images of natural scenes. Most of the processing during segmentation is local making the algorithm amenable to high performance parallel implementation.
ICCV	Real-Time Focus Range Sensor.	Shree K. Nayar,Masahiro Watanabe,Minori Noguchi	1995	Structures of dynamic scenes can only be recovered using a real-time range sensor. Depth-from-defocus offers a direct solution to fast and dense range estimation. It is computationally efficient as it circumvents the correspondence problem faced by stereo and feature tracking in structure-from-motion. However, accurate depth estimation requires theoretical and practical solutions to a variety of problems including the recovery of textureless surfaces, precise blur estimation, and magnification variations caused by defocusing. Both textured and textureless surfaces are recovered using an illumination pattern that is projected via the same optical path used to acquire images. The illumination pattern is optimized to ensure maximum accuracy and spatial resolution in the computed depth. The relative blurring in two images is computed using a narrow-band linear operator that is designed by considering all the optical, sensing and computational elements of the depth-from-defocus system. Defocus-invariant magnification is achieved by the use of an additional aperture in the imaging optics. A prototype focus range sensor has been developed that produces up to 512/spl times/480 depth estimates at 30 Hz with an accuracy better than 0.3%. Several experimental results are included to demonstrate the performance of the sensor.
ICCV	Deformable Velcro(tm) Surfaces.	Walter M. Neuenschwander,Pascal Fua,Gábor Székely,Olaf Kübler	1995	Deformable Velcro(tm) Surfaces.
ICCV	Surface Reconstruction: GNCs and MFA.	Mads Nielsen	1995	"The reconstruction of noise corrupted surfaces can be inferred by methodologies such as Bayesian estimation and minimum description length. Both of these imply a formulation where the reconstruction minimizes a functional. Often this functional is non convex and the minimum cannot be found by simple gradient methods. The paper concerns functionals with quadratic data term, criteria for such functionals to be convex, and the variational approach of minimizing non convex functionals. Initial convexity of the approximating functional is considered to be a critical point. Two fully automatic methods of generating convex functionals are presented. They are based on Gaussian convolution and are compared to the Blake-Zisserman graduated non convexity (GNC) (A. Blake, A. Zisserman, 1987) and G.L. Bilbro et al. (1992) and D. Geiger and F. Girosi's (1991) mean field annealing (MFA) of the weak membrane."
ICCV	The Nonparametric Approach for Camera Calibration.	MaoLin Qiu,Song De Ma	1995	The Nonparametric Approach for Camera Calibration.
ICCV	Invariant of a Pair of Non-Coplanar Conies in Space: Definition, Geometric Interpretation and Computation.	Long Quan	1995	The joint invariants of a pair of coplanar conics has been widely used in recent vision literature. In this paper, the algebraic invariant of a pair of non-coplanar conics in space is concerned. The algebraic invariant of a pair of non-coplanar conics is first derived from the invariant algebra of a pair of quaternary quadratic forms by using the dual representation of space conics. Then, this algebraic invariant is geometrically interpreted in terms of cross-ratios. Finally, an analytical procedure for projective reconstruction of a space conic from two uncalibrated images is developed and the correspondence conditions of the conics between two views are also explicited.
ICCV	Detecting Kinetic Occlusion.	Sourabh A. Niyogi	1995	Visual motion boundaries provide a powerful cue for the perceptual organization of scenes. Motion boundaries are present when surfaces in motion occlude one another. Conventional approaches to motion analysis have relied on assumptions of data conservation and smoothness, which has made analysis of motion boundaries difficult. We show that a common source of motion boundary, kinetic occlusion, can be detected using spatiotemporal junction analysis. Junction analysis is accomplished by utilizing distributed representations of motion used in models of human visual motion sensing. By detecting changes in the direction of motion in these representations, spatiotemporal junctions are detected in a manner which differentiates accretion from deletion. We demonstrate successful occlusion detection on spatiotemporal imagery containing occluding surfaces in motion.
ICCV	A Snake for Model-Based Segmentation.	Petia Radeva,Joan Serrat,Enric Martí	1995	Despite the promising results of numerous applications, the hitherto proposed snake techniques share some common problems: snake attraction by spurious edge points, snake degeneration (shrinking and flattening), convergence and stability of the deformation process, snake initialization and local determination of the parameters of elasticity. We argue here that these problems can be solved only when all the snake aspects are considered. The snakes proposed here implement a new potential field and external force in order to provide a deformation convergence, attraction by both near and far edges as well as snake behaviour selective according to the edge orientation. Furthermore, we conclude that in the case of model-based segmentation, the internal force should include structural information about the expected snake shape. Experiments using this kind of snakes for segmenting bones in complex hand radiographs show a significant improvement.
ICCV	A Theory of Specular Surface Geometry.	Michael Oren,Shree K. Nayar	1995	A theoretical framework is introduced for the perception of specular surface geometry. When an observer moves in three-dimensional space, real scene features, such as surface markings, remain stationary with respect to the surfaces they belong to. In contrast, a virtual feature, which is the specular reflection of a real feature, travels on the surface. Based on the notion of caustics, a novel feature classification algorithm is developed that distinguishes real and virtual features from their image trajectories that result from observer motion. Next, using support functions of curves, a closed-form relation is derived between the image trajectory of a virtual feature and the geometry of the specular surface it travels on. It is shown that in the 2D case where camera motion and the surface profile are coplanar, the profile is uniquely recovered by tracking just two unknown virtual features. Finally, these results are generalized to the case of arbitrary 3D surface profiles that are travelled by virtual features when camera motion is not confined to a plane. An algorithm is developed that uniquely recovers 3D surface profiles using a single virtual feature tracked from the occluding boundary of the object. All theoretical derivations and proposed algorithms are substantiated by experiments.
ICCV	Object Indexing Using an Iconic Sparse Distributed Memory.	Rajesh P. N. Rao,Dana H. Ballard	1995	"A general-purpose object indexing technique is described that combines the virtues of principal component analysis with the favorable matching properties of high-dimensional spaces to achieve high-precision recognition. An object is represented by a set of high-dimensional iconic feature vectors comprised of the responses of derivatives of Gaussian filters at a range of orientations and scales. Since these filters can be shown to form the eigenvectors of arbitrary images containing both natural and man-made structures, they are well-suited for indexing in disparate domains. The indexing algorithm uses an active vision system in conjunction with a modified form of Kanerva's (1988, 1993) sparse distributed memory which facilitates interpolation between views and provides a convenient platform for learning the association between an object's appearance and its identity. The robustness of the indexing method was experimentally confirmed by subjecting the method to a range of viewing conditions and the accuracy was verified using a well-known model database containing a number of complex 3D objects under varying pose."
ICCV	Model-Based Tracking of Self-Occluding Articulated Objects.	James M. Rehg,Takeo Kanade	1995	Computer sensing of hand and limb motion is an important problem for applications in human computer interaction and computer graphics. We describe a framework for local trading of self occluding motion, in which one part of an object obstructs the visibility of another. Our approach uses a kinematic model to predict occlusions and windowed templates to track partially occluded objects. We present offline 3D tracking results for hand motion with significant self occlusion.
ICCV	Weakly-Calibrated Stereo Perception for Rover Navigation.	Luc Robert,Michel Buffa,Martial Hebert	1995	Presents a vision system for autonomous navigation based on stereo perception without 3D reconstruction. This approach uses weakly calibrated stereo images, i.e. images for which only the epipolar geometry is known. The vision system first rectifies the images, matches selected points between the two images, and then computes the relative elevation of the points relative to a reference plane as well as the images of their projections on this plane. We have integrated this vision module into a complete navigation system. In this system, the relative elevation is used as a shape indicator in order to compute appropriate steering directions everytime a new stereo pair is processed. We have conducted initial experiments in unstructured, outdoor environments with an wheeled rover.
ICCV	Trilinearity of Three Perspective Views and its Associated Tensor.	Amnon Shashua,Michael Werman	1995	"It has been established that certain trilinear forms of three perspective views give rise to a tensor of 27 intrinsic coefficients. We show in this paper that a permutation of the the trilinear coefficients produces three homography matrices (projective transformations of planes) of three distinct intrinsic planes, respectively. This, in turn, yields the result that 3D invariants are recovered directly-simply by appropriate arrangement of the tensor's coefficients. On a secondary level, we show new relations between fundamental matrix, epipoles, Euclidean structure and the trilinear tensor. On the practical side, the new results extend the existing envelope of methods of 3D recovery from 2D views-for example, new linear methods that cut through the epipolar geometry, and new methods for computing epipolar geometry using redundancy available across many views."
ICCV	A Model-Based Integrated Approach to Track Myocardial Deformation Using Displacement and Velocity Constraints.	Pengcheng Shi,Glynn P. Robinson,R. Todd Constable,Albert J. Sinusas,James S. Duncan	1995	Accurate estimation of heart wall dense field motion and deformation could help to better understand the physiological processes associated with ischemic heart diseases, and to provide significant improvement in patient treatment. We present a new method of estimating left ventricular deformation which integrates instantaneous velocity information obtained within the mid-wall region with shape information found on the boundaries of the left ventricle. Velocity information is obtained from phase contrast magnetic resonance images, and boundary information is obtained from shape-based motion tracking of the endo- and cardial boundaries. The integration takes place within a continuum biomechanical heart model which is embedded in a finite element framework. We also employ a feedback mechanism to improve tracking accuracy. The integration of the two disparate but complementary sources overcomes some of the limitations of previous work in the field which concentrates on motion estimation from a single image-derived source.
ICCV	Probabilistic 3D Object Recognition.	Ilan Shimshoni,Jean Ponce	1995	A probabilistic 3D object recognition algorithm is presented. In order to guide the recognition process the probability that match hypotheses between image features and model features are correct is computed. A model is developed which uses the probabilistic peaking effect of measured angles and ratios of lengths by tracing iso angle and iso ratio curves on the viewing sphere. The model also accounts for various types of uncertainty in the input such as incomplete and inexact edge detection. For each match hypothesis the pose of the object and the pose uncertainty which is due to the uncertainty in vertex position are recovered. This is used to find sets of hypotheses which reinforce each other by matching features of the same object with compatible uncertainty subsets. A probabalistic expression is used to rank these hypothesis sets. The hypothesis sets with the highest rank are output. The algorithm has been fully implemented, and tested on real images.
ICCV	A Comparison of Projective Reconstruction Methods for Pairs of Views.	Charlie Rothwell,Gabriela Csurka,Olivier D. Faugeras	1995	Recently, different approaches for uncalibrated stereo have been suggested which permit projective reconstruction from multiple views. These use weak calibration which is represented by the epipolar geometry, and so no knowledge of the intrinsic or extrinsic camera parameters is required. We consider projective reconstructions from pairs of views, and compare a number of the available methods. Consequently we conclude which methods are most likely to be of use in applications that are dependent on 3D uncalibrated reconstructions.
ICCV	An Integral Approach to Free-Formed Object Modeling.	Heung-Yeung Shum,Martial Hebert,Katsushi Ikeuchi,Raj Reddy	1995	Presents a new approach to free-formed object modeling from multiple range images. In most conventional approaches, successive views are registered sequentially. In contrast to the sequential approaches, we propose an integral approach which reconstructs statistically optimal object models by simultaneously aggregating all data from multiple views into a weighted least-squares (WLS) formulation. The integral approach has two components. First, a global resampling algorithm constructs partial representations of the object from individual views so that correspondences can be established among different views. The global resampling algorithm is based on the spherical attribute image (SAI) previously introduced in the context of object representation and recognition. Second, a weighted least-squares algorithm integrates resampled partial representations of multiple views, using the technique of principal component analysis with missing data (PCAMD). Experiments using real range images show that our approach is robust against noise and mismatches, and generates accurate object models.
ICCV	Locating Objects Using the Hausdorff Distance.	William Rucklidge	1995	The Hausdorff distance is a measure defined between two point sets representing a model and an image. In the past, it has been used to search images for instances of a model that has been translated or translated and scaled by finding transformations that bring a large number of model features close to image features, and vice versa. The Hausdorff distance is reliable even when the image contains multiple objects, noise, spurious features, and occlusions. We apply it to the task of locating an affine transformation of a model in an image; this corresponds to determining the pose of a planar object that has undergone weak perspective projection. We develop a rasterised approach to the search and a number of techniques that allow us to quickly locate all transformations of the model that satisfy two quality criteria; we can also quickly locate only the best transformation. We discuss an implementation of this approach, and present some examples of its use.
ICCV	Steerable Wedge Filters.	Eero P. Simoncelli,Hany Farid	1995	"Steerable filters, as developed by Freeman and Adelson (1991), are a class of rotation-invariant linear operators that may be used to analyze local orientation patterns in imagery. The most common examples of such operators are directional derivatives of Gaussians and their 2D Hilbert transforms. The inherent symmetry of these filters produces an orientation response that is periodic with period /spl pi/, even when the underlying image structure does not have such symmetry. This problem may be alleviated by reconsidering the full class of steerable filters. We develop a family of even- and odd-symmetric steerable filters that have a spatially asymmetric ""wedge-like"" shape and are optimally localized in their orientation response. Unlike the original steerable filters, these filters are not based on directional derivatives and the Hilbert transform relationship is imposed on their angular components. We demonstrate the ability of these filters to properly represent oriented structures."
ICCV	Perceptual Organization in an Interactive Sketch Editing Application.	Eric Saund,Thomas P. Moran	1995	"The paper shows how techniques from computational vision can be deployed to support interactive sketch editing. While conventional computer supported drawing tools give users access to visible marks or image objects at a single level of abstraction, a human user's visual system rapidly constructs complex groupings and associations among image elements according to his or her immediate purposes. We have been exploring perceptually supported sketch editors in which computer vision algorithms run continuously, behind the scenes, to afford users efficient access to emergent visual objects in a drawing. We employ a flexible image interpretation architecture based on token grouping in a multiscale blackboard data structure. This organization supports multiple perceptual interpretations of line drawing data, domain specific knowledge bases for interpreting visual structures, and natural gesture based selection of visual objects."
ICCV	Combining Color and Geometric Information for the Illumination Invariant Recognition of 3D Objects.	David Slater,Glenn Healey	1995	Combining Color and Geometric Information for the Illumination Invariant Recognition of 3D Objects.
ICCV	Model-Based 2D&3D Dominant Motion Estimation for Mosaicing and Video Representation.	Harpreet S. Sawhney,Serge Ayer,Monika Gorkani	1995	Model-Based 2D&3D Dominant Motion Estimation for Mosaicing and Video Representation.
ICCV	ASSET-2: Real-Time Motion Segmentation and Shape Tracking.	Stephen M. Smith	1995	"The paper describes how image sequences taken by a moving video camera may be processed to detect and track moving objects against a moving background in real-time. The motion segmentation and shape tracking system as known as ASSET-2-A Scene Segmenter Establishing Tracking, Version 2. Motion is found by tracking image features, and segmentation is based on first-order (i.e., six parameter) flow fields. Shape tracking is performed using two dimensional radial map representation. The system runs in real-time, and is accurate and reliable. It requires no camera calibration and no knowledge of the camera's motion."
ICCV	Dynamic Rigid Motion Estimation from Weak Perspective.	Stefano Soatto,Pietro Perona	1995	"""Weak perspective"" represents a simplified projection model that approximates the imaging process when the scene is viewed under a small viewing angle and its depth relief is small relative to its distance from the viewer. We study how to generate dynamic models for estimating rigid 3D motion from weak perspective. A crucial feature in dynamic visual motion estimation is to decouple structure from motion in the estimation model. The reasons are both geometric-to achieve global observability of the model-and practical, for a structure independent motion estimator allows us to deal with occlusions and appearance of new features in a principled way. It is also possible to push the decoupling even further, and isolate the motion parameters that are affected by the so called ""bas relief ambiguity"" from the ones that are not. We present a novel method for reducing the order of the estimator by decoupling portions of the state space from the time evolution of the measurement constraint. We use this method to construct an estimator of full rigid motion (modulo a scaling factor) on a six dimensional state space, an approximate estimator for a four dimensional subset of the motion space, and a reduced filter with only two states. The latter two are immune to the bas relief ambiguity. We compare strengths and weaknesses of each of the schemes on real and synthetic image sequences."
ICCV	Limitations of Markov Random Fields as Models of Textured Images of Real Surfaces.	Athanasios Speis,Glenn Healey	1995	We investigate to what extent textures can be distinguished using conditional Markov fields and small samples. We establish that the least square (LS) estimator is the only reasonable choice for this task and we prove its asymptotic consistency and normality for a general class of random fields that include Gaussian Markov fields as a special case. The performance of this estimator when applied to textured images of real surfaces is poor if small boxes are used (20/spl times/20 or less). We investigate the nature of this problem by comparing the behavior predicted by the rigorous theory to the one that has been experimentally observed. Our analysis reveal that 20/spl times/20 samples contain enough information to distinguish between the textures in our experiments and that the poor performance mentioned above should be attributed to the fact that conditional Markov fields do not provide accurate models for textured images of many real surfaces. A more general model that exploits more efficiently the information contained in small samples is also suggested.
ICCV	Complete Scene Structure from Four Point Correspondences.	Steven M. Seitz,Charles R. Dyer	1995	A technique is presented for computing 3D scene structure from point and line features in monocular image sequences. Unlike previous methods, the technique guarantees the completeness of the recovered scene, ensuring that every scene feature that is detected in each image is reconstructed. The approach relies on the presence of four or more reference features whose correspondences are known in all the images. Under an orthographic or affine camera model, the parallax of the reference features provides constraints that simplify the recovery of the rest of the visible scene. An efficient recursive algorithm is described that uses a unified framework for point and line features. The algorithm integrates the tasks of feature correspondence and structure recovery, ensuring that all reconstructible features are tracked. In addition, the algorithm is immune to outliers and feature drift, two weaknesses of existing structure from motion techniques. Experimental results are presented for real images.
ICCV	Optimal Subpixel Matching of Contour Chains and Segments.	Bruno Serra,Marc Berthod	1995	This paper introduces a new general purpose algorithm that allows the optimal geometric match between contours to be determined, that is the transformation yielding a minimal deformation is obtained. The algorithm relies only on the geometric properties of the contours and does not call for any other constraint, so that it is particularly suitable when no parameterization of title deformation is available or desirable. Contour deformation is explicitly incorporated in the computation, allowing for a thorough use of all geometric information available. Moreover, no discretization is involved in the computation, resulting in two main advantages: first, the algorithm is robust to differences in the segmentation of contours and allows the matching of polygonal approximations of contours with very little loss of precision, second, subpixel precision matching can be achieved.
ICCV	Accurate Internal Camera Calibration Using Rotation, with Analysis of Sources of Error.	Gideon P. Stein	1995	Describes a simple and accurate method for internal camera calibration based on tracking image features through a sequence of images while the camera undergoes pure rotation. A special calibration object is not required and the method can therefore be used both for laboratory calibration and for self calibration in autonomous robots. Experimental results with real images show that focal length and aspect ratio can be found to within 0.15 percent, and lens distortion error can be reduced to a fraction of a pixel. The location of the principal point and the location of the center of radial distortion can each be found to within a few pixels. We perform a simple analysis to show to what extent the various technical details affect the accuracy of the results. We show that having pure rotation is important if the features are derived from objects close to the camera. In the basic method accurate angle measurement is important. The need to accurately measure the angles can be eliminated by rotating the camera through a complete circle while taking an overlapping sequence of images and using the constraint that the sum of the angles must equal 960 degrees.
ICCV	Robot Aerobics: Four Easy Steps to a More Flexible Calibration.	Daniel E. Stevenson,Margaret M. Fleck	1995	Presents a method for calibrating intrinsic and extrinsic camera parameters. This algorithm can easily be modified by other users to suit their particular calibration needs, without requiring a high-precision calibration target or complicated linear algebra. The algorithm uses controlled motions and a single light source to simulate calibration targets in convenient 3D locations. These convenient calibration targets enable us to simplify the calibration algorithm and gather dense data for lens distortion. Dense data makes the distortion correction more accurate than traditional low-order polynomial fits, and allows us to calibrate wide-angle lenses (
ICCV	Expected Performance of Robust Estimators Near Discontinuities.	Charles V. Stewart	1995	"In extracting a polynomial surface patch near an intensity or range discontinuity, a robust estimator must tolerate not only the truly random bad data (""random outliers""), but also the coherently structured points (""pseudo outliers"") that belong to a different surface. To characterize the performance of least median of squares, M estimators, Hough transforms, RANSAC, and MINPRAN on data containing both random and pseudo outliers, we develop two analytical measures, ""pseudo outlier bias"" and ""pseudo outlier breakdown"". Using these measures, we find that each robust estimator has surprisingly poor performance, even under the best possible circumstances, implying that present estimators should be used with care and new estimators should be developed."
ICCV	Motion Estimation with Quadtree Splines.	Richard Szeliski,Heung-Yeung Shum	1995	This paper presents a motion estimation algorithm based on a new multiresolution representation, the quadtree spline. This representation describes the motion field as a collection of smoothly connected patches of varying size, where the patch size is automatically adapted to the complexity of the underlying motion. The topology of the patches is determined by a quadtree data structure, and both split and merge techniques are developed for estimating this spatial subdivision. The quadtree spline is implemented using another novel representation, the adaptive hierarchical basis spline, and combines the advantages of adaptively-sized correlation windows with the speedups obtained with hierarchical basis preconditioners. Results are presented on some standard motion sequences.
ICCV	A Geometric Criterion for Shape-Based Non-Rigid Correspondence.	"Hemant Tagare,Don O'Shea,Anand Rangarajan"	1995	A geometric criterion is developed for establishing shape based non rigid correspondence between plane curves. Unlike previous efforts, the criterion does not use rigid invariants of shape. Instead, shapes are compared non rigidly from the vantage point of the correspondence. Geometric invariants are proposed for curves whose shapes can be exactly matched by a non rigid correspondence. The invariants are based on angular deviations of convex and concave segments of the curves. Examples of correspondences between curves obtained from medical images are provided.
ICCV	Curve and Surface Smoothing without Shrinkage.	Gabriel Taubin	1995	For a number of computational purposes, including visualization of scientific data and registration of multimodal medical data, smooth curves must be approximated by polygonal curves, and surfaces by polyhedral surfaces. An inherent problem of these approximation algorithms is that the resulting curves and surfaces appear faceted. Boundary-following and iso-surface construction algorithms are typical examples. To reduce the apparent faceting, smoothing methods are used. In this paper, we introduce a new method for smoothing piecewise linear shapes of arbitrary dimension and topology. This new method is in fact a linear low-pass filter that removes high-curvature variations, and does not produce shrinkage. Its computational complexity is linear in the number of edges or faces of the shape, and the required storage is linear in the number of vertices.
ICCV	Estimating the Tensor of Curvature of a Surface from a Polyhedral Approximation.	Gabriel Taubin	1995	Estimating principal curvatures and principal directions of a surface from a polyhedral approximation with a large number of small faces, such as those produced by iso-surface construction algorithms, has become a basic step in many computer vision algorithms, particularly in those targeted at medical applications. We describe a method to estimate the tensor of curvature of a surface at the vertices of a polyhedral approximation. Principal curvatures and principal directions are obtained by computing in closed form the eigenvalues and eigenvectors of certain 3/spl times/3 symmetric matrices defined by integral formulas, and closely related to the matrix representation of the tensor of curvature. The resulting algorithm is linear, both in time and in space, as a function of the number of vertices and faces of the polyhedral surface.
ICCV	Image Segmentation by Reaction-Diffusion Bubbles.	Hüseyin Tek,Benjamin B. Kimia	1995	Figure-ground segmentation is a fundamental problem in computer vision. The main difficulty is the integration of low-level, pixel-based local image features to obtain global object-based descriptions. Active contours in the form of snakes, balloons, and level-set modeling techniques have been proposed that satisfactorily address this question for certain applications. However, these methods require manual initialization, do not always perform well near sharp protrusions or indentations, or often cross gaps. We propose an approach inspired by these methods and a shock-based representation of shape in terms of parts, protrusions, and bends. Since initially it is not clear where the objects or their parts are, parts are hypothesized in the form of fourth order shocks randomly initialized in homogeneous areas of images. These shocks then form evolving contours, or bubbles, which grow, shrink, merge, split and disappear to capture the objects in the image. In the homogeneous areas of the image bubbles deform by a reaction-diffusion process. In the inhomogeneous areas, indicated by differential properties computed from low-level processes such as edge-detection, texture, optical-flow and stereo, etc., bubbles do not deform. As such, the randomly initialized bubbles integrate low-level information and, in the process, segment the figures from the ground.
ICCV	Animat Vision: Active Vision in Artificial Animals.	Demetri Terzopoulos,Tamer F. Rabie	1995	We propose and demonstrate a new paradigm for active vision research that draws upon recent advances in the fields of artificial life and computer graphics. A software alternative to the prevailing hardware vision mindset, animat vision prescribes artificial animals, or animats, situated in physics-based virtual worlds as autonomous virtual robots possessing active perception systems. To be operative in its world, an animat must autonomously control its eyes and muscle-actuated body, applying computer vision algorithms to continuously analyze the retinal image streams acquired by its eyes in order to locomote purposefully through its world. We describe an initial animat vision implementation within lifelike artificial fishes inhabiting a physics-based, virtual marine world. Emulating the appearance, motion, and behavior of real fishes in their natural habitats, these animats are capable of spatially nonuniform retinal imaging, foveation, retinal image stabilization, color object recognition, and perceptually-guided navigation. These capabilities allow them to pursue moving targets such as fellow artificial fishes. Animat vision offers a fertile approach to the development, implementation, and evaluation of computational theories that profess sensorimotor competence for animal or robotic situated agents.
ICCV	Recovering 3D Motion of Multiple Objects Using Adaptive Hough Transform.	Tina Yu Tian,Mubarak Shah	1995	"Presents a method to determine the 3D motion of multiple objects from two perspective views. In our method, segmentation is determined based on a 3D rigidity constraint. We divide the input image into overlapping patches, and for each sample of the translation parameter space, we compute the rotation parameters of patches using a least-squares fit. Every patch votes for a sample in the translation and rotation parameter space. For a patch containing multiple motions, we use an M-estimator to compute rotation parameters of a dominant motion. We use the adaptive Hough transform to refine the relevant parameter space in a ""coarse-to-fine"" fashion. Applications of the proposed method to both synthetic and real images are demonstrated with promising results."
ICCV	Computation of Coherent Optical Flow by Using Multiple Constraints.	Massimo Tistarelli	1995	The optical flow constitutes one of the most widely adopted representations to define and characterize the evolution of image features over time. In order to compute the velocity field, it is necessary to define a set of constraints on the temporal change of image features. We consider the implications in using multiple constraints arising from multiple data points. The first step is the analysis of differential constraints and how they can be applied, locally, to compute the image velocity. This analysis allows to relate each constraint to a particular gray level pattern. This approach is extended to multiple image points, allowing also the characterization of the temporal behaviour of the image features and to detect erroneous measurements due to occlusions, depth discontinuities or shadows. Several experiments are presented from real image sequences.
ICCV	Robust Detection of Degenerate Configurations for the Fundamental Matrix.	Philip H. S. Torr,Andrew Zisserman,Stephen J. Maybank	1995	New methods are reported for the detection of multiple solutions (degeneracy) when estimating the fundamental matrix, with specific emphasis on robustness in the presence of data contamination (outliers). The fundamental matrix can be used as a first step in the recovery of structure from motion. If the set of correspondences is degenerate then this structure cannot be accurately recovered and many solutions will explain the data equally well. It is essential that we are alerted to such eventualities. However, current feature matchers are very prone to mismatching, giving a high rate of contamination within the data. Such contamination can make a degenerate data set appear non degenerate, thus the need for robust methods becomes apparent. The paper presents such methods with a particular emphasis on providing a method that will work on real imagery and with an automated (non perfect) feature detector and matcher. It is demonstrated that proper modelling of degeneracy in the presence of outliers enables the detection of outliers which would otherwise be missed. Results using real image sequences are presented. All processing, point matching, degeneracy detection and outlier detection is automatic.
ICCV	Matching Constraints and the Joint Image.	Bill Triggs	1995	"The paper studies the geometry of multi image perspective projection and the matching constraints that this induces on image measurements. The combined image projections define a 3D joint image subspace of the space of combined homogeneous image coordinates. This is a complete projective replica of the 3D world in image coordinates. Its location encodes the imaging geometry and is captured by the 4 index joint image Grassmannian tensor. Projective reconstruction in the joint image is a canonical process requiring only a simple rescaling of image coordinates. Reconstruction in world coordinates amounts to a choice of basis in the joint image. The matching constraints are multilinear tensorial equations in image coordinates that tell whether tokens in different images could be the projections of a single world token. For 2D images of 3D points there are exactly three basic types: the epipolar constraint, A. Shashua's (1995) trilinear one, and a new quadrilinear 4 image one. For images of lines, R. Hartley's (1994) trilinear constraint is the only type. The coefficients of the matching constraints are tensors built directly from the joint image Grassmannian. Their complex algebraic interdependency is captured by quadratic structural simplicity constraints on the Grassmannian."
ICCV	Towards an Active Visual Observer.	Tomas Uhlin,Peter Nordlund,Atsuto Maki,Jan-Olof Eklundh	1995	We present a binocular active vision system that can attend to and fixate a moving target. Our system has an open and expandable design and it forms the first steps of a long term effort towards developing an active observer using vision to interact with the environment, in particular capable of figure-ground segmentation. We also present partial real-time implementations of this system and show their performance in real-world situations together with motor control. In pursuit we particularly focus on occlusions of other targets, both stationary and moving, and integrate three cues, ego-motion, target motion and target disparity, to obtain an overall robust behavior. An active vision system must be open, expandable, and operate with whatever data are available momentarily. It must also be equipped with means and methods to direct and change its attention. This system is therefore equipped with motion detection for changing attention and pursuit for maintaining attention, both of which run concurrently.
ICCV	Motion Analysis with a Camera with Unknown, and Possibly Varying Intrinsic Parameters.	Thierry Viéville,Olivier D. Faugeras	1995	In the present paper we address the problem of computing structure and motion, given a set point correspondences in a monocular image sequence, considering small motions when the camera is not calibrated. We first set the equations defining the calibration, rigid motion and scene structure. We then review the motion equation, the structure from equation and the depth evolution equation, including the particular case of planar structures, considering a discrete displacement between two frames. A step further, we develop the first order expansion of these equations and analyse the observability of the related infinitesimal quantities. It is shown that we obtain a complete correspondence between these equations and the equation derived in the discrete case. However, in the case of infinitesimal displacements, the projection of the translation (focus of expansion or epipole) is clearly separated from the rotational component of the motion. This is an important advantage of the present approach. Using this last property, we propose a mechanism of image stabilization in which the rotational disparity is iteratively canceled. This allows a better estimation of the focus of expansion, and simplifies different aspects of the analysis of the equations: structure from motion equation, analysis of ambiguity, geometrical interpretation of the motion equation.
ICCV	Invariant-Based Recognition of Complex Curved 3D Objects from Image Contours.	B. Vijayakumar,David J. Kriegman,Jean Ponce	1995	To recognize three-dimensional objects bounded by smooth curved surfaces from monocular image contours, viewpoint-dependent image features must be related to object geometry. Contour bitangents and inflections along with associated parallel tangents points are the projection of surface points that lie on the occluding contour for a five-parameter family of scaled orthographic projection viewpoints. An invariant representation can be computed from these image features and seen for modeling and recognizing objects. Modeling is achieved by moving an object in front of a camera to obtain a curve of possible invariants. The relative camera-object motion is not required, and 3D models are not utilized. At recognition time, invariants computed from a single image are used to index the model database. Using the matched features, independent qualitative and quantitative verification procedures eliminate potential false matches. Examples from an implementation are presented.
ICCV	Alignment by Maximization of Mutual Information .	Paul A. Viola,William M. Wells III	1995	A new information-theoretic approach is presented for finding the pose of an object in an image. The technique does not require information about the surface properties of the object, besides its shape, and is robust with respect to variations of illumination. In our derivation, few assumptions are made about the nature of the imaging process. As a result, the algorithms are quite general and can foreseeably be used in a wide variety of imaging situations. Experiments are presented that demonstrate the approach in registering magnetic resonance images, aligning a complex 3D object model to real scenes including clutter and occlusion, tracking a human head in a video sequence and aligning a view-based 2D object model to real images. The method is based on a formulation of the mutual information between the model and the image. As applied in this paper, the technique is intensity-based, rather than feature-based. It works well in domains where edge or gradient-magnitude based methods have difficulty, yet it is more robust then traditional correlation. Additionally, it has an efficient implementation that is based on stochastic approximation.
ICCV	Shape from Shading with Interreflections under Proximal Light Source: 3D Shape Reconstruction of Unfolded Book Surface from a Scanner Image.	Toshikazu Wada,Hiroyuki Ukida,Takashi Matsuyama	1995	We address the problem to recover the 3D shape of an unfolded book surface from the shading information in a scanner image. From a technical point of view, this shape from shading problem in real world environments is characterized by (1) proximal light source, (2) interreflections, (3) moving light source, (4) specular reflection, and (5) nonuniform albedo distribution. Taking all these factors into account, we first formulate the problem based on an iterative nonlinear optimization scheme. Then we introduce piecewise polynomial models of the 3D shape. Image restoration experiments for a real book surface demonstrated that geometric and photometric distortions are almost completely removed by the proposed method.
ICCV	Region Correspondence by Inexact Attributed Planar Graph Matching.	Caihua Wang,Keiichi Abe	1995	An efficient graph matching approach is proposed for finding region correspondence between two images of the same scene but taken from different viewpoints. Regions and their relations in an image are represented with region adjacency graph (RAG), which is a kind of attributed planar graph. The problem to find an optimal region correspondence, which matches the regions in two images with maximal similarity in region features and region relations, is formulated into the problem to find the optimal inexact matching between two RAGs. The properties specific to planar graph and that of the region adjacency relations are utilized to invent an efficient algorithm to solve the problem. Experimental results on various kinds of images show the effectiveness of the method.
ICCV	Rigid Body Segmentation and Shape Description from Dense Optical Flow under Weak Perspective.	Joseph Weber,Jitendra Malik	1995	We present an algorithm for identifying and tracking independently moving rigid objects from optical flow. The proposed method uses the fact that each distinct object has a unique epipolar constraint associated with its motion. This is in contrast to using local optical flow information for segmentation. Thus motion discontinuities based on self-occlusion are distinguished from those due to separate objects. The use of epipolar geometry allows for the determination of individual motion parameters for each object as well as the recovery of relative depth for each point on the object. The segmentation problem is formulated as a scene partitioning problem and a statistic-based algorithm which uses only nearest neighbor interactions and a finite number of iterations is developed. A Kalman filter based approach is used for tracking motion parameters with time. The algorithm assumes an affine camera where perspective effects are limited to changes in overall scale. No camera calibration parameters are required.
ICCV	The Study of 3D-from-2D Using Elimination.	Michael Werman,Amnon Shashua	1995	The Study of 3D-from-2D Using Elimination.
ICCV	Rendering Real-World Objects Using View Interpolation.	Tomás Werner,Roger D. Hersch,Václav Hlavác	1995	Presents a new approach to rendering arbitrary views of real-world 3D objects of complex shapes. We propose to represent an object by a sparse set of corresponding 2D views, and to construct any other view as a combination of these reference views. We show that this combination can be linear, assuming proximity of the views, and we suggest how the visibility of constructed points can be determined. Our approach makes it possible to avoid difficult 3D reconstruction, assuming only rendering is required. Moreover, almost no calibration of views is needed. We present preliminary results on real objects, indicating that the approach is feasible.
ICCV	Algorithms for Implicit Deformable Models.	Ross T. Whitaker	1995	This paper presents a framework for implicit deformable models and a pair of new algorithms for solving the nonlinear partial differential equations that result from this framework. Implicit models offer a useful alternative to parametric models, particularly when dealing with the deformation of higher-dimensional objects. The basic expressions for the evolution of implicit models are relatively straightforward; they follow as a direct consequence of the chain rule for differentiation. More challenging, however, is the development of algorithms that are stable and efficient. The first algorithm is a viscosity approximation which gives solutions over a dense set in the range, providing a means of calculating the solutions of embedded families of contours simultaneously. The second algorithm incorporates sparse solutions for a discrete set of contours. This sparse-field method requires a fraction of the computation compared to the first but offers solutions only for a finite number of contours. Results from 3d medical data as well as video images are shown.
ICCV	Closing the Loop on Multiple Motions.	Charles Wiles,Michael Brady	1995	We describe a number of advances in the analysis of road scenes when the scene contains multiple moving objects and is observed by a single nonsteerable camera mounted on the front of a vehicle. Our structure from motion approach to scene segmentation derives front the observed motions of independently moving objects and requires no prior knowledge. We describe a hierarchy of camera models for the analysis of the scene, the simpler models handle degeneracies that occur in the more complex models. The major technical contribution is the recursive computation of feature clusters, which are fed forward over time. This closed loop feature tracking generates extended feature trajectories which significantly improve the discriminating power of scene segmentation.
ICCV	A Quantitative Analysis of View Degeneracy and its use for Active Focal Length control.	David Wilkes,Sven J. Dickinson,John K. Tsotsos	1995	We quantify the observation by Kender and Freudenstein (1987) that degenerate views occupy a significant fraction of the viewing sphere surrounding an object. This demonstrates that systems for recognition must explicitly account for the possibility of view degeneracy. We show that view degeneracy cannot be detected from a single camera viewpoint. As a result, systems designed to recognize objects from a single arbitrary viewpoint must be able to function in spite of possible undetected degeneracies, or else operate with imaging parameters that cause acceptably low probabilities of degeneracy. To address this need, we give a prescription for active control of focal length that allows a principled tradeoff between the camera field of view and probability of view degeneracy.
ICCV	Stochastic Completion Fields: A Neural Model of Illusory Contour Shape and Salience.	Lance R. Williams,David W. Jacobs	1995	We describe an algorithm and representation level theory of illusory contour shape and salience. Unlike previous theories, our model is derived from a single assumption-namely, that the prior probability distribution of boundary completion shape can be modeled by a random walk in a lattice whose points are positions and orientations in the image plane (i.e. the space which one can reasonably assume is represented by neurons of the mammalian visual cortex). Our model does not employ numerical relaxation or other explicit minimization, but instead relies on the fact that the probability that a particle following a random walk will pass through a given position and orientation on a path joining two boundary fragments can be computed directly as the product of two vector-field convolutions. We show that for the random walk we define, the maximum likelihood paths are curves of least energy, that is, on average, random walks follow paths commonly assumed to model the shape of illusory contours. A computer model is demonstrated on numerous illusory contour stimuli from the literature.
ICCV	A State-Based Technique for the Summarization and Recognition of Gesture.	Andrew D. Wilson,Aaron F. Bobick	1995	A State-Based Technique for the Summarization and Recognition of Gesture.
ICCV	Relational Matching with Dynamic Graph Structures.	Richard C. Wilson,Edwin R. Hancock	1995	The paper describes a novel approach to relational matching problems in machine vision. Rather than matching static scene descriptions, the approach adopts an active representation of the data to be matched. This representation is iteratively reconfigured to increase its degree of topological congruency with the model relational structure in a reconstructive matching process. The active reconfiguration of relational structures is controlled by a MAP update process. The final restored graph representation is optimal in the sense that it has maximum a posteriori probability with respect to the available attributes for the objects under match. The benefits of the technique are demonstrated experimentally on the matching of cluttered synthetic aperture radar data to a model in the form of a digital map. The operational limits of the method are established in a simulation study.
ICCV	Hypergeometric Filters for Optical Flow and Affine Matching.	Yalin Xiong,Steven A. Shafer	1995	"This paper proposes new ""hypergeometric"" filters for the problem of image matching under the translational and affine model. This new set of filters has the following advantages: (1) High-precision registration of two images under the translational and affine model. Because the window effects are eliminated, we are able to achieve superb performance in both translational and affine matching. (2) Affine matching without exhaustive search or image warping. Due to the recursiveness of the filters in the spatial domain, we are able to analytically express the relation between filter outputs and the six affine parameters. This analytical relation enables us to directly compute these affine parameters. (3) Generality. The approach we demonstrate here can be applied to a broad class of matching problems, as long as the transformation between the two image patches can be mathematically represented in the frequency domain."
ICCV	Region Competition: Unifying Snakes, Region Growing, Energy/Bayes/MDL for Multi-band Image Segmentation.	Song Chun Zhu,Tai Sing Lee,Alan L. Yuille	1995	We present a novel statistical and variational approach to image segmentation based on a new algorithm named region competition. This algorithm is derived by minimizing a generalized Bayes/MDL (Minimum Description Length) criterion using the variational principle. We show that existing techniques in early vision such as, snake/balloon models, region growing, and Bayes/MDL are addressing different aspects of the same problem and they can be unified within a common statistical framework which combines their advantages. We analyze how to optimize the precision of the resulting boundary location by studying the statistical properties of the region competition algorithm and discuss what are good initial conditions for the algorithm. Our method is generalized to color and texture segmentation and is demonstrated on grey level images, color images and texture images.
ICCV	FORMS: A Flexible Object Recognition and Modelling System.	Song Chun Zhu,Alan L. Yuille	1995	We briefly describe a generic statistical framework for representing the shapes of animate objects using principal component analysis and stochastic shape grammars. Such a representation scheme gives a formalism for solving the inverse problem-object recognition. Then we show: how these representations can be extracted from 2D silhouettes by a novel method for skeleton extraction and shape segmentation; how a similarity metric can be defined on this shape space; and how we can perform recognition in a bottom up/top down loop. The system is demonstrated to be stable in the presence of noise, the absence of parts, the presence of additional parts, and considerable variations in articulation and viewpoint. Successful categorization is demonstrated on a dataset of seventeen categories of animate objects.
ICCV	Class-Based Grouping in Perspective Images.	Andrew Zisserman,Joseph L. Mundy,David A. Forsyth,Jane Liu,Nic Pillow,Charlie Rothwell,Sven Utcke	1995	In any object recognition system a major and primary task is to associate those image features, within an image of a complex scene, that arise from an individual object. The key idea here is that a geometric class defined in 3D induces relationships in the image which must hold between points on the image outline (the perspective projection of the object). The resulting image constraints enable both identification and grouping of image features belonging to objects of that class. The classes include surfaces of revolution, canal surfaces (pipes) and polyhedra. Recognition proceeds by first recognising an object as belonging to one of the classes (for example a surface of revolution) and subsequently identifying the object (for example as a particular vase). This differs from conventional object recognition systems where recognition is generally targetted at particular objects. These classes also support the computation of 3D invariant descriptions including symmetry axes, canonical coordinate frames and projective signatures. The constraints and grouping methods are viewpoint invariant, and proceed with no information on object pose. We demonstrate the effectiveness of this class-based grouping on real, cluttered scenes using grouping algorithms developed for rotationally symmetric surfaces, canal-surfaces and polyhedra.
ICCV	Estimating Motion and Structure from Correspondences of Line Segments between Two Perspective Images.	Zhengyou Zhang	1995	Presents an algorithm for determining 3D motion and structure from correspondences of line segments between two perspective images. To our knowledge, the paper is the first investigation of use of line segments in motion and structure from motion. Classical methods use their geometric abstraction, namely straight lines, but then three images are necessary for the motion and structure determination process. We show that two views are in general sufficient when we use line segments. The assumption we use is that two matched line segments contain the projection of a common part of the corresponding line segment in space. Indeed this is what we use to match line segments between different views. Both synthetic and real data have been used to test the proposed algorithm, and excellent results have been obtained with real data containing a relatively large set of line segments. The results are comparable with those obtained using stereo calibration.
ICCV	Shape and Model from Specular Motion.	Jiang Yu Zheng,Yoshihiro Fukagawa,Norihiro Abe	1995	This work investigates visual characteristics of specular surfaces during rotation, and gives approaches to qualitatively identify and quantitatively recover shapes of these surfaces. Continuous images are taken when an object rotates. We look at specularly reflected patterns on the surfaces and their motion in the EPIs parallel to the rotation plane, from which estimation of each surface point and construction of the object model are carried out. We find very simple and direct methods to fulfill this objective; linear equations for multiple lights illumination, and a 1st-order differential equation for single light illumination. The motion of specular reflection has nice global characteristics in EPI. The surface types range from very shiny metal surfaces to surfaces with only week specular reflectance. We give both simulation and experiments on real objects.
IEEE Visualization	An Extended Data-Flow Architecture for Data Analysis and Visualization.	Greg Abram,Lloyd Treinish	1995	"Modular visualization environments utilizing a data-flow execution model have become quite popular in recent years, especially those that incorporate visual programming tools. However, simplistic implementations of such an execution model are quite limited when applied to problems of realistic complexity, which negate the intuitive advantage of data-flow systems. This situation can be resolved by extending the execution model to incorporate a more complete and efficient programming infrastructure while still preserving the virtues of ""pure data-flow"". This approach has been used for the implementation of a general-purpose software package, IBM Visualization Data Explorer."
IEEE Visualization	Unsteady Flow Volumes.	Barry G. Becker,Nelson L. Max,David A. Lane	1995	Flow volumes [1] are extended for use in unsteady (time-dependent) flows. The resulting unsteady flow volumes are the 3 dimensional analog of streaklines. There are few examples where methods other than particle tracing have been used to visualize time varying flows. Since particle paths can become convoluted in time there are additional considerations to be made when extending any visualization technique to unsteady flows. We will present some solutions to the problems which occur in subdivision, rendering, and system design. We will apply the unsteady flow volumes to a variety of field types including moving multi-zoned curvilinear grids.
IEEE Visualization	Interval Set: A Volume Rendering Technique Generalizing Isosurface Extraction.	Baining Guo	1995	A scalar volume V = { (x, f(x)) | x in R } is described by a function f(x) defined over some region R of the three-dimensional space. This paper presents a simple technique for rendering interval sets of the form I(a, b) = { (x, f(x)) | a
IEEE Visualization	A Rule-Based Tool for Assisting Colormap Selection.	Lawrence D. Bergman,Bernice E. Rogowitz,Lloyd Treinish	1995	"The paper presents an interactive approach for guiding the user's select of colormaps in visualization. PRAVDAColor, implemented as a module in the IBM Visualization Data Explorer, provides the user a selection of appropriate colormaps given the data type and spatial frequency, the user's task, and properties of the human perceptual system."
IEEE Visualization	Visualization of Biological Sequence Similarity Search Results.	Ed Huai-hsin Chi,Phillip Barry,Elizabeth Shoop,John V. Carlis,Ernest Retzel,John Riedl	1995	Biological sequence similarity analysis presents visualization challenges, primarily because of the massive amounts of discrete, multi-dimensional data. Genomic data generated by molecular biologists is analyzed by algorithms that search for similarity to known sequences in large genomic databases. The output from these algorithms can be several thousand pages of text, and is difficult to analyze because of its length and complexity. We developed and implemented a novel graphical representation for sequence similarity search results, which visually reveals features that are difficult to find in textual reports. The method opens new possibilities in the interpretation of this discrete, multi-dimensional data by enabling interactive investigation of the graphical representation.
IEEE Visualization	Interactive Realism for Visualization Using Ray Tracing.	Robert A. Cross	1995	"Visual realism is necessary for many virtual reality applications. In order to convince the user that the virtual environment is real, the scene presented should faithfully model the expected actual environment. A highly accurate, fully modeled, interactive environment is thus seen as ""virtually real.""This paper addresses the problem of interactive visual realism and discusses a possible solution: a hybrid rendering paradigm that ties distributed graphics hardware and ray tracing systems together for use in interactive, high visual realism applications.This new paradigm is examined in the context of a working rendering system. This system is capable of producing images of higher fidelity than possible through the use of graphics hardware alone, able both to render images at speeds useful for interactive systems and to progressively refine static, high quality snapshots."
IEEE Visualization	Using Spatial Access Methods to Support the Visualization of Environmental Data.	Charles Falkenberg,Ravi Kulkarni	1995	Using Spatial Access Methods to Support the Visualization of Environmental Data.
IEEE Visualization	Vector Plots for Irregular Grids.	Don Dovey	1995	"A standard method for visualizing vector fields consists of drawing many small ``glyphs'' to represent the field. This paper extends the technique from regular to curvilinear and unstructured grids. In order to achieve a uniform density of vector glyphs on nonuniformly spaced grids, the paper describes two approaches to resampling the grid data. One of the methods, an element-based resampling, can be used to visualize vector fields at arbitrary surfaces within three-dimensional grids."
IEEE Visualization	A Visualization Tool for Studying the Development of the Moss Physcomitrella patens.	F. David Fracchia,Neil W. Ashton	1995	A Visualization Tool for Studying the Development of the Moss Physcomitrella patens.
IEEE Visualization	Interval Volume: A Solid Fitting Technique for Volumetric Data Display and Analysis.	Issei Fujishiro,Yuji Maeda,Hiroshi Sato	1995	"Proposes as a generalization of isosurfaces, the 'interval volume', which is a new type of geometric model representing 3D subvolumes with field values belonging to a closed interval. A dominant surface fitting algorithm called 'marching cubes' is extended to obtain a solid fitting algorithm, which extracts from a given volumetric dataset a high-resolution polyhedral solid data structure of the interval volume. Rendering methods for the interval volume and principal related operations are also presented. The effectiveness of this approach is illustrated with 4D simulated data from atomic collision research."
IEEE Visualization	Direct Rendering of Laplacian Pyramid Compressed Volume Data.	Mohammad H. Ghavamnia,Xue D. Yang	1995	Volume rendering generates 2D images by ray tracing 3D volume data. This technique imposes considerable demands on storage space as the data set grows in size. In this paper we describe a method to render compressed volume data directly to reduce the memory requirements of the rendering process. The volume data was compressed by a technique called Laplacian pyramid. A compression ratio of 10:1 was achieved by uniform quantization over the Laplacian pyramid. The quality of the images obtained by this technique is virtually indistinguishable from that of the images generated from the uncompressed volume data. A significant improvement in computational performance was achieved by using a cache algorithm to temporarily retain the reconstructed voxels to be used by the adjacent rays.
IEEE Visualization	Subverting Structure: Data-Driven Diagram Generation.	Gene Golovchinsky,Klaus Reichenberger,Thomas Kamps	1995	Diagrams are data representations that convey information predominantly through combinations of graphical elements rather than through other channels such as text or interaction. We have implemented a prototype called AVE (Automatic Visualization Environment) that generates diagrams automatically based on a generative theory of diagram design. According to this theory, diagrams are constructed based on the data to be visualized rather than by selection from a predefined set of diagrams. This approach can be applied to knowledge represented by semantic networks. In this paper we give a brief introduction to the underlying theory, then describe the implementation and finally discuss strategies for extending the algorithm.
IEEE Visualization	Astronomers and their Shady Algorithms.	Richard Gooch	1995	The vast quantities of data which may be produced by modern radio telescopes have outstripped conventional visualisation techniques available to astronomers. While research in other areas of visualisation finds some application in astronomy, problems peculiar to the field require new techniques. This paper presents a brief overview of some of the problems of visualisation for astronomy and compares different shading algorithms.
IEEE Visualization	Visualization of High Speed Aerodynamic Configuration Design.	Monika Hannemann,Helmut Sobieczky	1995	Three topics of aerodynamic research at DLR are chosen to illustrate the need for visualization. These include aircraft configuration design variations, adaptation devices and unsteady flow simulation in the transonic, the supersonic and the hypersonic speed regime call for the combined use of a geometry generator, a powerful graphic system and video technology. Projects currently under investigation are illustrated and generic case studies are presented.
IEEE Visualization	Space Walking.	Andrew J. Hanson,Hui Ma	1995	"We propose an interactive method for exploring topological spaces based on the natural local geometry of the space. Examples of spaces appropriate for this visualization approach occur in abundance in mathematical visualization, surface and volume visualization problems, and scientific applications such as general relativity. Our approach is based on using a controller to choose a direction in which to ""walk"" a manifold along a local geodesic path. The method automatically generates orientation changes that produce a maximal viewable region with each step of the walk. The proposed interaction framework has many natural properties to help the user develop a useful cognitive map of a space and is well-suited to haptic interfaces that may be incorporated into desktop virtual reality systems."
IEEE Visualization	Voxel Based Object Simplification.	Taosong He,Lichan Hong,Arie E. Kaufman,Amitabh Varshney,Sidney W. Wang	1995	Presents a simple, robust and practical method for object simplification for applications where gradual elimination of high-frequency details is desired. This is accomplished by sampling and low-pass filtering the object into multi-resolution volume buffers and applying the marching cubes algorithm to generate a multi-resolution triangle-mesh hierarchy. Our method simplifies the genus of objects and can also help existing object simplification algorithms achieve better results. At each level of detail, a multi-layered mesh can be used for an optional and efficient antialiased rendering.
IEEE Visualization	Qualitative Analysis of Invariant Tori in a Dynamical System.	Daryl H. Hepting,Gianne Derks,Kossi D. Edoh,Robert D. Russell	1995	Invariant tori are examples of invariant manifolds in dynamical systems. Usual tools in dynamical systems such as analysis and numerical simulations alone are often not sufficient to understand the complicated mechanisms that cause changes in these manifolds. Computer-graphical visualization is a natural and powerful addition to these tools used for the qualitative study of dynamical systems, especially for the study of invariant manifolds. The dynamics of two linearly coupled oscillators is the focus of this case study. With little or no coupling between the oscillators, an invariant torus is present but it breaks down for strong coupling. Visualization has been employed to gain a qualitative understanding of this breakdown process. The visualization has allowed key features of the tori to be recognized, and it has proven to be indispensable in developing and testing hypotheses about the tori.
IEEE Visualization	Interactive Visualization of Mixed Scalar and Vector Fields.	Lichan Hong,Xiaoyang Mao,Arie E. Kaufman	1995	This paper describes an approach for interactive visualization of mixed scalar and vector fields, in which vector icons are generated from pre-voxelized icon templates and volume-rendered together with the volumetric scalar data. This approach displays simultaneously the global structure of the scalar field and the detailed features of the vector field. Interactive visualization is achieved with incremental image update, by re-rendering only a small portion of the image wherever and whenever a change occurs. This technique supports a set of interactive visualization tools, including change of vector field visualization parameters, real-time animation of vector icons advected within the scalar field, a zooming lens, and a local probe. Subject Terms: data visualisation; interactive systems; graphical user interfaces; rendering (computer graphics); interactive visualization; scalar fields; vector fields; vector icons; pre-voxelized icon templates; volume-rendered; volumetric scalar data; incremental image update; real-time animation; zooming lens; local probe
IEEE Visualization	On Enhancing the Speed of Splatting with Indexing.	Insung Ihm,Rae Kyoung Lee	1995	Splatting is an object-space direct volume rendering algorithm that produces images of high quality, but is computationally expensive like many other volume rendering algorithms. This paper presents a new technique that enhances the speed of splatting without trading off image quality. This new method reduces rendering time by employing a simple indexing mechanism which allows to visit and splat only the voxels of interest. It is shown that this algorithm is suitable for the dynamic situation in which viewing parameters and opacity transfer functions change interactively. We report experimental results on several test data sets of useful size and complexity, and discuss the cost/benefit trade-off of our method.
IEEE Visualization	Legibility Enhancement for Information Visualisation.	Rob Ingram,Steve Benford	1995	"Navigation in computer generated information spaces may be difficult, resulting in users getting ""lost in hyperspace."" This work aims to build on research from the area of city planning to try to solve this problem. We will introduce the concepts of legibility and cognitive maps and the five features of urban landscape with which they are associated. Following this will be descriptions of techniques and algorithms which we have developed to allow these features to be introduced to three dimensional spaces for information visualisation. Next we will describe a specific application of these techniques in the visualisation of the World Wide Web and conclude with a look at future development of the system."
IEEE Visualization	Enhancing Transparent Skin Surfaces with Ridge and Valley Lines.	Victoria Interrante,Henry Fuchs,Stephen M. Pizer	1995	There are many applications that can benefit from the simultaneous display of multiple layers of data. The objective in these cases is to render the layered surfaces in such a way that the outer structures can be seen and seen through at the same time. This paper focuses on the particular application of radiation therapy treatment planning, in which physicians need to understand the three-dimensional distribution of radiation dose in the context of patient anatomy. We describe a promising technique for communicating the shape and position of the transparent skin surface while at the same time minimally occluding underlying isointensity dose surfaces and anatomical objects: adding a sparse, opaque texture comprised of a small set of carefully-chosen lines. We explain the perceptual motivation for explicitly drawing ridge and valley curves on a transparent surface, describe straightforward mathematical techniques for detecting and rendering these lines, and propose a small number of reasonably effective methods for selectively emphasizing the most perceptually relevant lines in the display.
IEEE Visualization	An Integrated Approach for Steering, Visualization, and Analysis of Atmospheric Simulations.	Yves Jean,Thomas Kindler,William Ribarsky,Weiming Gu,Greg Eisenhauer,Karsten Schwan,Fred Alyea	1995	An Integrated Approach for Steering, Visualization, and Analysis of Atmospheric Simulations.
IEEE Visualization	Tensor Product Surfaces Guided by Minimal Surface Area Triangulations.	John K. Johnstone,Kenneth R. Sloan	1995	We present a method for constructing tensor product Bezier surfaces from contour (cross-section) data. Minimal area triangulations are used to guide the surface construction, and the final surface reflects the optimality of the triangulation. The resulting surface differs from the initial triangulation in two important ways: it is smooth (as opposed to the piecewise planar triangulation), and it is in tensor product form (as opposed to the irregular triangular mesh). The surface reconstruction is efficient because we do not require an exact minimal surface. The triangulations are used as strong hints, but no more than that. The method requires the computation of both open and closed isoparametric curves of the surface, using triangulations as a guide. These isoparametric curves form a tensor product Bezier surface. We show how to control sampling density by filling and pruning isoparametric curves, for accuracy and economy. A rectangular grid of points is produced that is compatible with the expected format for a tensor product surface interpolation, so that a host of well-supported methods are available to generate and manipulate the surface.
IEEE Visualization	Recursive Pattern: A Technique for Visualizing Very Large Amounts of Data.	Daniel A. Keim,Mihael Ankerst,Hans-Peter Kriegel	1995	"An important goal of visualization technology is to support the exploration and analysis of very large amounts of data. In this paper, we propose a new visualization technique called `recursive pattern' which has been developed for visualizing large amounts of multidimensional data. The technique is based on a generic recursive scheme which generalizes a wide range of pixel-oriented arrangements for displaying large data sets. By instantiating the technique with adequate data- and application-dependent parameters, the user may largely influence the structure of the resulting visualizations. Since the technique uses one pixel for presenting each data value, the amount of data which can be displayed is only limited by the resolution of current display technology and by the limitations of human perceptibility. Beside describing the basic idea of the `recursive pattern' technique, we provide several examples of useful parameter settings for the various recursion levels. We further show that our `recursive pattern ' technique is particularly advantageous for the large class of data sets which have a natural order according to one dimension (e.g. time series data). We demonstrate the usefulness of our technique by using a stock market application."
IEEE Visualization	Optimization of Time-Dependent Particle Tracing Using Tetrahedral Decomposition.	David N. Kenwright,David A. Lane	1995	An efficient algorithm is presented for computing particle paths, streak lines and time lines in time-dependent flows with moving curvilinear grids. The integration, velocity interpolation, and step size control are all performed in physical space which avoids the need to transform the velocity field into computational space. This leads to higher accuracy because there are no Jacobian matrix approximations, and expensive matrix inversions are eliminated. Integration accuracy is maintained using an adaptive step size control scheme which is regulated by the path line curvature. The problem of point location and interpolation in physical space is simplified by decomposing hexahedral cells into tetrahedral cells. This enables the point location to be done analytically and substantially faster than with a Newton-Raphson iterative method. Results presented show this algorithm is up to six times faster than particle tracers which operate on hexahedral cells and produces almost identical traces.
IEEE Visualization	Visualization for Aerodynamic Design of Helicopter Rotor Blades.	G. David Kerlick	1995	Visualization, animation, and simulation techniques are applied to the problem of rotor design for helicopters. Periodic unsteady experimental velocity data (laser Doppler velocimetry or LDV) in two dimensions and velocity data derived from simulated vortex systems in three dimensions are compared using the same visual tools. Animations show the development of rotor wake systems and induced velocities over time. Modified particle trace integration schemes are used to calculate steady streamlines and unsteady particle paths for both kinds of data. In an extension of this work, a virtual environment (VE) system was used to view the wake vortex system and an interactive probe was used to explore the induced velocity field. Future work will enable interactive visual debugging and simulation steering.
IEEE Visualization	An Architecture for Retaining and Analyzing Visual Explorations of Databases.	John Peter Lee,Georges G. Grinstein	1995	"A software architecture is presented to integrate a database management system with data visualization. One of it's primary objectives, the retention of user-data interactions, is detailed. By storing all queries over the data along with high-level descriptions of the query result and associated visualization, the process by wich a database is explored can be analyzed. This approach can lead to contributions in the development of user models as ""data explorers"", metadata models for scientific databases, intelligent assistants, and data exploration services. We describe the underlying elements of this approach, specifically the visual database exploration model and the metadata objects that support the model."
IEEE Visualization	High-Speed Volume Rendering Using Redundant Block Compression.	Günter Knittel	1995	We present a novel volume rendering method which offers high rendering speed on standard workstations. It is based on a lossy data compression scheme which drastically reduces the memory bandwidth and computing requirements of perspective raycasting. Starting from classified and shaded data sets, we use Block Truncation Coding or Color Cell Compression to compress a block of 12 voxels into 32 bits. All blocks of the data set are processed redundantly, yielding a data structure which avoids multiple memory accesses per raypoint. As a side effect, the tri-linear interpolation of data coded in such a way is very much simplified. These techniques allow us to perform walk-throughs at interactive frame rates. Furthermore, the algorithm provides depth-cueing and the semi-transparent display of different materials. The algorithm achieves a sustained frame generation rate of about 2Hz for large data sets (~200^3) at an acceptable image quality on an SGI Indy workstation. A number of examples are shown.
IEEE Visualization	Enhanced Spot Noise for Vector Field Visualization.	Wim C. de Leeuw,Jarke J. van Wijk	1995	Spot noise is a technique for texture synthesis, which is very useful for vector field visualization. This paper describes improvements and extensions of the basic principle of spot noise. First, better visualization of highly curved vector fields with spot noise is achieved, by adapting the shape of the spots to the local velocity field. Second, filtering of spots is proposed to eliminate undesired low frequency components from the spot noise texture. Third, methods are described to utilize graphics hardware to generate the texture, and to produce variable viewpoint animations of spot noise on surfaces. And fourth, the synthesis of spot noise on grids with highly irregular cell sizes is described.
IEEE Visualization	Virtual GIS: A Real-Time 3D Geographic Information System.	David Koller,Peter Lindstrom,William Ribarsky,Larry F. Hodges,Nickolas Faust,Gregory A. Turner	1995	Virtual GIS: A Real-Time 3D Geographic Information System.
IEEE Visualization	Marching Through the Visible Man.	William E. Lorensen	1995	"The National Library of Medicine is creating a digital atlas of the human body. This project, called the Visible Human, has already produced computed tomography, magnetic resonance imaging and physical cross-sections of a human male cadaver. This paper describes a methodology and results for extracting surfaces from the Visible Male's CT data. We use surface connectivity and isosurface extraction techniques to create polygonal models of the skin, bone, muscle and bowels. We also report early experiments with the physical cross-sections."
IEEE Visualization	3D Computational Steering with Parametrized Geometric Objects.	Jurriaan D. Mulder,Jarke J. van Wijk	1995	Computational Steering is the ultimate goal of interactive simulation: researchers change parameters of their simulation and immediately receive feedback on the effect. We present a general and flexible graphics tool that is part of an environment for Computational Steering developed at CWI. It enables the researcher to interactively develop his own interface with the simulation. This interface is constructed with 3D Parametrized Geometric Objects. The properties of the objects are parametrized to output data and input parameters of the simulation. The objects visualize the output of the simulation while the researcher can steer the simulation by direct manipulation of the objects. Several applications of 3D Computational Steering are presented.
IEEE Visualization	Pictorial Statistics - Visualization of High Dimensional Statistical Distributions.	Andreas Müller	1995	A general framework for visualization of statistical properties of high-dimensional pattern samples and the related computational steps are introduced. These procedures are exemplified on applications in anthropometrical research (shape information in faces) but can be easily generalized to various other morphometrical questions and data sets with pattern structure, e.g., data stemming from sensor arrays. Presently, the visualization techniques illustrated concentrate on (higher) moments of first order. It is suggested, how moments of second order can be visualized by animations and how this approach can be used in the context of comparative visualization.
IEEE Visualization	Splatting of Curvilinear Volumes.	Xiaoyang Mao,Lichan Hong,Arie E. Kaufman	1995	The paper presents a splatting algorithm for volume rendering of curvilinear grids. A stochastic sampling technique called Poisson sphere/ellipsoid sampling is employed to adaptively resample a curvilinear grid with a set of randomly distributed points whose energy support extents are well approximated by spheres and ellipsoids. Filter kernels corresponding to these spheres and ellipsoids are used to generate the volume rendered image of the curvilinear grid with a conventional footprint evaluation algorithm. Experimental results show that our approach can be regarded as an alternative to existing fast volume rendering techniques of curvilinear grids.
IEEE Visualization	Turbulent Flow Visualization in Computational and Experimental Hydraulics.	A. E. Mynett,I. Ari Sadarjoen,A. J. S. Hin	1995	Many practical problems in open channel hydraulics that were traditionally investigated in hydraulic model experiments, are nowadays being solved by using computational fluid dynamics. However, in order to interpret computational results, there is a clear preference among scientist and engineers for visualization in analogy with experimental techniques. One such technique, particle tracing, enables a dynamic (Lagrangian) interpretation of a statically (Eulerian) computed vector field. However, quite often the emphasis in particle tracing is only on the mean flow properties, while effects due to dispersion and mixing are often not accounted for. Hence turbulent flow characteristics have to be incorporated in a visualization system for practical hydraulic engineering problems. The particle tracing technique presented in this case study has been specifically developed to combine both mean and fluctuating velocity vectors, thus simulating stochastic perturbations around mean flow conditions. A number of cases are presented in this paper that demonstrate the practical applicability of advanced visualization techniques in realistic engineering studies.
IEEE Visualization	Interactive Maximum Projection Volume Rendering.	Michael D. McCool,Wolfgang Heidrich,John Stevens	1995	Interactive Maximum Projection Volume Rendering.
IEEE Visualization	An Illustrated Analysis of Sonification for Scientific Visualisation.	Rosane Minghim,A. Robin Forrest	1995	This paper presents an analysis of progress in the use of sound as a tool in support of visualisation and gives an insight into its development and future needs. Special emphasis is given to the use of sound in Scientific and Engineering Applications. A system developed to support surface data presentation and interaction by using sound is presented and discussed.
IEEE Visualization	Visualizing the Tracking and Diving Behavior of Marine Mammals.	Guy W. Oliver	1995	Visualizing the Tracking and Diving Behavior of Marine Mammals.
IEEE Visualization	Flow Visualization in a Hypersonic Fin/Ramp Flow.	Hans-Georg Pagendarm,Thomas Gerhold	1995	A recent study of a flow detail of an engine intake of future ground to orbit transport systems provided extremely complex data from numerical flow simulation and experimental flow visualization. The data posed a challenging problem to flow visualization, computational flow imaging (CFI), and the comparison of experimental imaging techniques versus computational imaging techniques. Some new visualization techniques have been implemented to provide compact representations of the complex features in the data. It turned out to be most useful to combine various specialized techniques for an icon-like representation of phenomena in a single image in order to study interaction of flow features. Some lessons were learned by simulating experimental visualization techniques on the numerical data.
IEEE Visualization	Iconic Techniques for Feature Visualization.	Frank J. Post,Theo van Walsum,Frits H. Post,Deborah Silver	1995	Iconic Techniques for Feature Visualization.
IEEE Visualization	Propositional n-traces: Visualizing a Problem in Philosophic Logic.	Nathalie Prevost,Ray E. Jennings,Loki Jorgenson,F. David Fracchia	1995	Propositional n-traces: Visualizing a Problem in Philosophic Logic.
IEEE Visualization	Interactive 3D Visualization of Actual Anatomy and Simulated Chemical Time-Course Data for Fish.	Penny Rheingans,John Nichols	1995	Interactive 3D Visualization of Actual Anatomy and Simulated Chemical Time-Course Data for Fish.
IEEE Visualization	Sweeping Simplices: A Fast Iso-Surface Extraction Algorithm for Unstructured Grids.	Han-Wei Shen,Christopher R. Johnson	1995	"We present an algorithm that accelerates the extraction of iso-surfaces from unstructured grids by avoiding the traversal of the entire set of cells in the volume. The algorithm consists of a sweep algorithm and a data decomposition scheme. The sweep algorithm incrementally locates intersected elements, and the data decomposition scheme restricts the algorithm's worst-case performance. For data sets consisting of hundreds of thousands of elements, our algorithm can reduce the cell traversal time more than 90% over the naive iso-surface extraction algorithm, thus facilitating interactive probing of scalar fields for large-scale problems on unstructured three-dimensional grids."
IEEE Visualization	Fast Normal Estimation Using Surface Characteristics.	Byeong-Seok Shin,Yeong-Gil Shin	1995	To visualize the volume data acquired from computation or sampling, it is necessary to estimate normals at the points corresponding to object surfaces. Volume data does not holds the geometric information for the surface comprising points, so it is necessary to calculate normals using local information at each point. The existing normal estimation methods have some problems of estimating incorrect normals at discontinuous, aliased or noisy points. Yagel et al solved some of these problems using their context-sensitive method. However this method requires too much processing time and it loses some information on detailed parts of the object surfaces. This paper proposes the surface-characteristic-sensitive normal estimation method which applies different operators according to characteristics of each surface for the normal calculation. This method has the same advantages of the context-sensitive method, and also some other advantages such as the less processing time and the reduction of the information loss on detailed parts.
IEEE Visualization	Automatic Generation of Triangular Irregular Networks Using Greedy Cuts.	Cláudio T. Silva,Joseph S. B. Mitchell,Arie E. Kaufman	1995	"Proposes a new approach to the automatic generation of triangular irregular networks (TINs) from dense terrain models. We have developed and implemented an algorithm based on the greedy principle used to compute minimum-link paths in polygons. Our algorithm works by taking greedy cuts (""bites"") out of a simple closed polygon that bounds the yet-to-be triangulated region. The algorithm starts with a large polygon, bounding the whole extent of the terrain to be triangulated, and works its way inward, performing at each step one of three basic operations: ear cutting, greedy biting, and edge splitting. We give experimental evidence that our method is competitive with current algorithms and has the potential to be faster and to generate many fewer triangles. Also, it is able to keep the structural terrain fidelity at almost no extra cost in running time and it requires very little memory beyond that for the input height array."
IEEE Visualization	A Hardware Acceleration Method for Volumetric Ray Tracing.	Lisa M. Sobierajski,Ricardo S. Avila	1995	In this paper we present an acceleration method for volumetric ray tracing which utilizes standard graphics hardware without compromising image accuracy. The graphics hardware is employed to identify those segments of each ray that could possibly contribute to the final image. A volumetric ray tracing algorithm is then used to compute the final image, traversing only the identified segments of the rays. This technique can be used to render volumetric isosurfaces as well as translucent volumes. In addition, this method can accelerate the traversal of shadow rays when performing recursive ray tracing.
IEEE Visualization	Fast Multiresolution Surface Meshing.	Oliver G. Staadt,Markus H. Gross,Roger Gatti	1995	Fast Multiresolution Surface Meshing.
IEEE Visualization	Automated Generation of Visual Simulation Databases Using Remote Sensing and GIS.	Martin Suter,Daniel Nüesch	1995	"This paper reports on the development of a strategy to generate databases used for real-time interactive landscape visualization. The database construction from real world data is intended to be as automated as possible. The primary sources of information are remote sensing imagery recorded by Landsat's Thematic Mapper (TM) and digital elevation models (DEM). Additional datasets (traffic networks and buildings) are added to extend the database. In a first step the TM images are geocoded and then segmented into areas of different land coverage. During the visual simulation highly detailed photo textures are applied onto the terrain based on the classification results to increase the apparent amount of detail. The data processing and integration is carried out using custom image processing and geographic information systems (GIS) software. Finally, a sample visual simulation application is implemented. Emphasis is put on practical implementation to test the feasibility of the approach as a whole."
IEEE Visualization	Fast Algorithms for Visualizing Fluid Motion in Steady Flow on Unstructured Grids.	Shyh-Kuang Ueng,K. Sikorski,Kwan-Liu Ma	1995	The plotting of streamlines is an effective way of visualizing fluid motion in steady flows. Additional information about the flowfield, such as local rotation and expansion, can be shown by drawing in the form of a ribbon or tube. In this paper, we present efficient algorithms for the construction of streamlines, streamribbons and streamtubes on unstructured grids. A specialized version of the Runge-Kutta method has been developed to speed up the integration of particle pathes. We have also derived close-form solutions for calculating angular rotation rate and radius to construct streamribbons and streamtubes, respectively. According to our analysis and test results, these formulations are two to four times better in performance than previous numerical methods. As a large number of traces are calculated, the improved performance could be significant.
IEEE Visualization	Defining, Computing, and Visualizing Molecular Interfaces.	Amitabh Varshney,Frederick P. Brooks Jr.,David C. Richardson,William V. Wright,Dinesh Manocha	1995	Defining, Computing, and Visualizing Molecular Interfaces.
IEEE Visualization	A Model and a System for Data-Parallel Program Visualization.	Thomas A. Wagner,R. Daniel Bergeron	1995	Parallel program visualization and debugging require new techniques for gathering and displaying execution trace and profile data. Interaction with the program during execution is also required to facilitate parallel debugging. We discuss the difficulties associated with runtime user/program interaction and how the data-parallel programming paradigm facilitates much more liberal runtime interaction than typical MIMD-based models. We present a model for data-parallel program visualization that addresses both data collection/interaction and visualization issues. We follow our model presentation with the design and implementation of a subset of our visualization model. We discuss our preliminary findings and propose future research directions.
IEEE Visualization	High Dimensional Brushing for Interactive Exploration of Multivariate Data.	Matthew O. Ward,Allen R. Martin	1995	High Dimensional Brushing for Interactive Exploration of Multivariate Data.
IEEE Visualization	Compression Domain Rendering of Time-Resolved Volume Data.	Rüdiger Westermann	1995	An important challenge in the visualization of three dimensional volume data is the efficient processing and rendering of time-resolved sequences. Only the use of compression techniques, which allow the reconstruction of the original domain from the compressed one locally, makes it possible to evaluate these sequences in their entirety. In the following paper a new approach for the extraction and visualization of so called time-features from within time-resolved volume data will be presented. Based on the asymptotic decay of multiscale representations of spatially localized time evolutions of the data, singular points can be discriminated. Also the corresponding Lipschitz exponents, which describe the signals local regularity, can be determined, and can be taken as a measure of the variation in time. The compression ratio and the comprehension of the underlying signal will be improved, if we restore the extracted regions first, which contain the most important information.
IEEE Visualization	IFS Fractal Interpolation for 2D and 3D Visualization.	Craig M. Wittenbrink	1995	"Reconstruction is used frequently in visualization of one, two, and three--dimensional data. Data uncertainty is typically ignored, and a deficiency of many interpolation schemes is smoothing which may indicate features or characteristics of the data that are not there. In this paper I investigate the use of iterated function systems (IFS's) for interpolation. I show new derivations for fractal interpolation in two and three-dimensional scalar data, and new point and polytope rendering algorithms with tremendous speed advantages over ray tracing. The interpolations may be used to give an indication of the uncertainty of the data, statistically represent the data at a variety of scales, allow tunability from the data, and may allow more accurate data analysis."
IEEE Visualization	Authenticity Analysis of Wavelet Approximations in Visualization.	Pak Chung Wong,R. Daniel Bergeron	1995	Wavelet transforms include data decompositions and reconstructions. This paper is concerned with the authenticity issues of the data decomposition, particularly for data visualization. A total of six datasets are used to clarify the approximation characteristics of compactly supported orthogonal wavelets. We present an error tracking mechanism, which uses the available wavelet resources to measure the quality of the wavelet approximations.
IEEE Visualization	Volume-Based Reasoning and Visualization of Diecastability.	Roni Yagel,Shao-Chiung Lu,Alec B. Rebello,R. A. Miller	1995	Because of the nature of the die casting process, the part geometry severely restricts the die geometry and hence affects the quality of the part. However, as is often the case in other manufacturing processes, diecastings are currently designed purely based on their function. The manufacturability of the diecastings is not considered until the design has been nearly completed and detailed. This is due to the design support limitations of current CAE tools. In this paper, we present a new volume-based approach to support diecastability evaluation, especially in preliminary design. Our approach can be applied to arbitrarily shaped parts without pre-defined feature libraries. The focus is on the identification of geometric characteristics, e.g. heavy mass regions, that could be responsible for thermal-related part defects. A distance transform with city-block metric is used to extract this geometric property. Volume visualization techniques are also adopted to allow users to visualize the results in a clear and precise way.
IEEE Visualization	"IEEE Visualization '95, Proceedings, Atlanta, Georgia, USA, October 29 - November 3, 1995."		1995	"IEEE Visualization '95, Proceedings, Atlanta, Georgia, USA, October 29 - November 3, 1995."
IEEE Trans. Vis. Comput. Graph.	A Predictor-Corrector Technique for Visualizing Unsteady Flow.	David C. Banks,Bart A. Singer	1995	We present a method for visualizing unsteady flow by displaying its vortices. The vortices are identified by using a vorticity-predictor pressure-corrector scheme that follows vortex cores. The cross-sections of a vortex at each point along the core can be represented by a Fourier series. A vortex can be faithfully reconstructed from the series as a simple quadrilateral mesh, or its reconstruction can be enhanced to indicate helical motion. The mesh can reduce the representation of the flow features by a factor of one thousand or more compared with the volumetric dataset. With this amount of reduction it is possible to implement an interactive system on a graphics workstation to permit a viewer to examine, in three dimensions, the evolution of the vortical structures in a complex, unsteady flow.
IEEE Trans. Vis. Comput. Graph.	Visualizing Network Data.	Richard A. Becker,Stephen G. Eick,Allan R. Wilks	1995	Networks are critical to modern society, and a thorough understanding of how they behave is crucial to their efficient operation. Fortunately, data on networks is plentiful; by visualizing this data, it is possible to greatly improve our understanding. Our focus is on visualizing the data associated with a network and not on simply visualizing the structure of the network itself. We begin with three static network displays; two of these use geographical relationships, while the third is a matrix arrangement that gives equal emphasis to all network links. Static displays can be swamped with large amounts of data; hence we introduce direct-manipulation techniques that permit the graphs to continue to reveal relationships in the context of much more data. In effect, the static displays are parameterized so that interesting views may easily be discovered interactively. The software to carry out this network visualization is called SeeNet.
IEEE Trans. Vis. Comput. Graph.	Line Art Rendering via a Coverage of Isoparametric Curves.	Gershon Elber	1995	Line Art Rendering via a Coverage of Isoparametric Curves.
IEEE Trans. Vis. Comput. Graph.	Optimal Sampling for Hemicubes.	Nelson L. Max	1995	The hemicube estimates of form factors are based on a finite set of sample directions. We obtain several optimal arrangements of sample directions, which minimize the variance of these estimates. They are based on changing the size or shape of the pixels or the shape of the hemicube, or using non-uniform pixel grids. The best reduces the variance by 43%.The variance calculation is based on the assumption that the errors in the estimate are caused by the projections of single polygon edges, and that the positions and orientations of these edges are random. This replaces the infinite dimensional space of possible environments by the two dimensional space of great circles on the unit sphere, making the numerical variance minimization possible.
IEEE Trans. Vis. Comput. Graph.	Specification by-Example of Virtual Agents Behavior.	Alberto Del Bimbo,Enrico Vicario	1995	The development of virtual agents running within graphic environments which emulate real-life contexts may largely benefit from the use of visual specification by-example. To support this specification, the development system must be able to interpret the examples and cast their underlying rules into an internal representation language. This language must find a suitable trade-off among a number of contrasting requirements regarding expressiveness, automatic executability, and suitability to the automatic representation of rules deriving from the analysis of examples.A language is presented which attains this trade-off by combining together an operational and a declarative fragment to separately represent the autonomous execution of each individual agent and its interaction with the environment, respectively. While the declarative part permits to capture interaction rules emerging from specification examples, the operational part supports the automatic execution in the operation of the virtual environment. A system is presented which embeds this language within a visual shell to support a behavioral training in which the animation rules of virtual agents are defined through visual examples.
IEEE Trans. Vis. Comput. Graph.	Using Line Integral Convolution for Flow Visualization: Curvilinear Grids, Variable-Speed Animation, and Unsteady Flows.	Lisa K. Forssell,Scott D. Cohen	1995	Line Integral Convolution (LIC), introduced by Cabral and Leedom in SIGGRAPH ¿93, is a powerful technique for imaging and animating vector fields. We extend the LIC technique in three ways:1. The existing algorithm is limited to vector fields over a regular Cartesian grid. We extend the algorithm and the animation techniques possible with it to vector fields over curvilinear surfaces, such as those found in computational fluid dynamics simulations.2. We introduce a technique to visualize vector magnitude as well as vector direction, i.e., variable-speed flow animation.3. We show how to modify LIC to visualize unsteady (time dependent) flows.Our implementation utilizes texture-mapping hardware to run in real time, which allows our algorithms to be included in interactive applications.
IEEE Trans. Vis. Comput. Graph.	Editorials.		1995	Editorials.
IEEE Trans. Vis. Comput. Graph.	Modeling and Visualization of Knitwear.	Eduard Gröller,René T. Rau,Wolfgang Straßer	1995	Visualization and modeling of textile materials has already been investigated in detail in the computer graphics literature. Most of the work, however, concentrates on woven fabrics. In this paper we present a novel approach to the modeling and rendering of knitwear. After the topological specification of a knitting pattern a subdivision into basic elements is done. The yarn microstructure within basic elements is approximated by volume data sets. The repetitive structure of knitted fabrics allows an efficient rendering technique. Resulting images are given that demonstrate the feasibility of our approach.
IEEE Trans. Vis. Comput. Graph.	Visualization of Multidimensional Shape and Texture Features in Laser Range Data Using Complex-Valued Gabor Wavelets.	Markus H. Gross,Rolf M. Koch	1995	This paper describes a new method for visualization and analysis of multivariate laser range data using complex-valued non-orthogonal Gabor wavelets, principal component analysis and a topological mapping network. The initial data set that provides both shape and texture information is encoded in terms of both amplitude and phase of a complex valued 2D image function. A set of carefully designed oriented Gabor filters performs a decomposition of the data and allows for retrieving local shape and texture features. The feature vector obtained from this method is multidimensional and in order to evaluate similar data features, further subspace methods to transform the data onto visualizable attributes, such as R, G, B, have to be determined. For this purpose, a feature-based visualization pipeline is proposed consisting of principal component analysis, normalization and a topological mapping network. This process finally renders a R, G, B subspace representation of the multidimensional feature vector. Our method is primarily applied to the visual analysis of features in human faces_but is not restricted to that.
IEEE Trans. Vis. Comput. Graph.	Exploiting Triangulated Surface Extraction Using Tetrahedral Decomposition.	André Guéziec,Robert A. Hummel	1995	Beginning with digitized volumetric data, we wish to rapidly and efficiently extract and represent surfaces defined as isosurfaces in the interpolated data. The Marching Cubes algorithm is a standard approach to this problem. We instead perform a decomposition of each 8-cell associated with a voxel into five tetrahedra. Following the ideas of Kalvin et al. [18], Thirion and Gourdon [30], and extending the work of Doi and Koide [5], we guarantee the resulting surface representation to be closed and oriented, defined by a valid triangulation of the surface of the body, which in turn is presented as a collection of tetrahedra. The entire surface is ¿wrapped¿ by a collection of triangles, which form a graph structure, and where each triangle is contained within a single tetrahedron. The representation is similar to the homology theory that uses simplices embedded in a manifold to define a closed curve within each tetrahedron.We introduce data structures based upon a new encoding of the tetrahedra that are at least four times more compact than the standard data structures using vertices and triangles. For parallel computing and improved cache performance, the vertex information is stored local to the tetrahedra. We can distribute the vertices in such a way that no tetrahedron ever contains more than one vertex.We give methods to evaluate surface curvatures and principal directions at each vertex, whenever these quantities are defined. Finally, we outline a method for simplifying the surface, that is reducing the vertex count while preserving the geometry. We compare the characteristics of our methods with an 8-cell based method, and show results of surface extractions from CT-scans and MR-scans at full resolution.
IEEE Trans. Vis. Comput. Graph.	Editorial.		1995	Editorial.
IEEE Trans. Vis. Comput. Graph.	A Multiscale Model for Structure-Based Volume Rendering.	Baining Guo	1995	A scalar volume V={(x, f(x)) | x ¿R} is described by a function f(x) defined over some region R of the 3D space. In this paper, we present a simple technique for rendering multiscale interval sets of the form ${\cal I}_{\mbi s}$(a, b) = {(x, fs(x)) |a¿gs(x) ¿b}, where a and b are either real numbers or infinities, and fs(x) is a smoothed version of f(x). At each scale s, the constraint a¿gs (x) ¿b identifies a subvolume in which the most significant variations of V are found. We use dyadic wavelet transform to construct gs(x) from f(x) and derive subvolumes with the following attractive properties: 1) the information contained in the subvolumes are sufficient for reconstructing the entire V, and 2) the shapes of the subvolumes provide a hierarchical description of the geometric structures of V. Numerically, the reconstruction in 1) is only an approximation, but it is visually accurate as errors reside at fine scales where our visual sensitivity is not so acute. We triangulate interval sets as ¿-shapes, which can be efficiently rendered as semi-transparent clouds. Because interval sets are extracted in the object space, their visual display can respond to changes of the view point or transfer function quite fast. The result is a volume rendering technique that provides faster, more effective user interaction with practically no loss of information from the original data. The hierarchical nature of multiscale interval sets also makes it easier to understand the usual complicated structures in scalar volumes.
IEEE Trans. Vis. Comput. Graph.	On Particle Path Generation Based on Quadrilinear Interpolation and Bernstein-Bézier Polynomials.	Bernd Hamann,Donghua Wu,Robert J. Moorhead II	1995	On Particle Path Generation Based on Quadrilinear Interpolation and Bernstein-Bézier Polynomials.
IEEE Trans. Vis. Comput. Graph.	In Memoriam: Dr. Wolfgang Krueger.	Pat Hanrahan	1995	In Memoriam: Dr. Wolfgang Krueger.
IEEE Trans. Vis. Comput. Graph.	Quaternion Frame Approach to Streamline Visualization.	Andrew J. Hanson,Hui Ma	1995	Curves in space are difficult to perceive and analyze, especially when they form dense sets as in typical 3D flow and volume deformation applications. We propose a technique that exposes essential properties of space curves by attaching an appropriate moving coordinate frame to each point, reexpressing that moving frame as a unit quaternion, and supporting interaction with the resulting quaternion field. The original curves in three-space are associated with piecewise continuous four-vector quaternion fields, which map into new curves lying in the unit three-sphere in four-space. Since four-space clusters of curves with similar moving frames occur independently of the curves¿ original proximity in three-space, a powerful analysis tool results. We treat two separate moving-frame formalisms, the Frenet frame and the parallel-transport frame, and compare their properties. We describe several flexible approaches for interacting with and exploiting the properties of the four-dimensional quaternion fields.
IEEE Trans. Vis. Comput. Graph.	Collision Detection for Interactive Graphics Applications.	Philip M. Hubbard	1995	"Solid objects in the real world do not pass through each other when they collide. Enforcing this property of ``solidness'''' is important in many interactive graphics applications; for example, solidness makes virtual reality more believable, and solidness is essential for the correctness of vehicle simulators. These applications use a collision-detection algorithm to enforce the solidness of objects. Unfortunately, previous collision-detection algorithms do not adequately address the needs of interactive applications. To work in these applications, a collision-detection algorithm must run at real-time rates, even when many objects can collide, and it must tolerate objects whose motion is specified ``on the fly'''' by a user. This dissertation describes a new collision-detection algorithm that meets these criteria through approximation and graceful degradation, elements of ``time-critical computing.'''' The algorithm is not only fast but also interruptible, allowing an application to trade accuracy for speed as needed. The algorithm uses two forms of approximate geometry. The first is a four-dimensional structure called a ``space-time bound.'''' This structure provides a conservative estimate of where an object may be in the immediate future based on estimates of the object''s acceleration. The algorithm uses space-time bounds to focus its attention on objects that are likely to collide. The second form of approximate geometry is a ``sphere-tree.'''' This structure contains a hierarchy of unions of spheres, each union approximating the three-dimensional surface of an object at a different level of detail. Sphere-trees allow the algorithm to quickly find approximate contacts between objects, and they allow the application to interrupt the algorithm in order to meet real-time performance goals. Automatically building sphere-trees is an interesting problem in itself, and this dissertation describes several approaches. The simplest approach is based on octrees, and more sophisticated approaches use simulated annealing and approximate medial-axis surfaces. Several of the steps in these algorithms are themselves significant. One step is a simple algorithm for checking whether a union of two-dimensional shapes covers a polygon. Another step builds Voronoi diagrams for three-dimensional data points, and does so more robustly and accurately than previous algorithms. An implementation of the collision-detection algorithm runs significantly faster than a previous algorithm in empirical tests. In particular, this implementation allows real-time performance in a sample application (a simple flight simulator) that is too slow with the previous algorithm; in some cases, the performance improves by more than two orders of magnitude. Experience with this sample application suggests that time-critical computing is not always simple to apply, but it provides enough benefits that it deserves further exploration in other contexts."
IEEE Trans. Vis. Comput. Graph.	Automatic Isosurface Propagation Using an Extrema Graph and Sorted Boundary Cell Lists.	Takayuki Itoh,Koji Koyamada	1995	A high-performance algorithm for generating isosurfaces is presented. In our method, guides to searching for cells intersected by an isosurface are generated as a pre-process. These guides are two kinds of cell lists: an extrema graph, and sorted lists of boundary cells. In an extrema graph, extremum points are connected by arcs, and each arc has a list of cells through which it passes. At the same time, all boundary cells are sorted according to their minimum and maximum values, and two sorted lists are then generated. Isosurfaces are generated by visiting adjacent intersected cells in order. Here, the starting cells for this process are found by searching in an extrema graph and in sorted boundary cell lists. In this process, isosurfaces appear to propagate themselves. Our algorithm is efficient, since it visits only cells that are intersected by an isosurface and cells whose IDs are included in the guides. It is especially efficient when many isosurfaces are interactively generated in a huge volume. Some benchmark tests described in this paper show the efficiency of the algorithm.
IEEE Trans. Vis. Comput. Graph.	Optical Models for Direct Volume Rendering.	Nelson L. Max	1995	This tutorial survey paper reviews several different models for light interaction with volume densities of absorbing, glowing, reflecting, and/or scattering material. They are, in order of increasing realism, absorption only, emission only, emission and absorption combined, single scattering of external illumination without shadows, single scattering with shadows, and multiple scattering. For each model I give the physical assumptions, describe the applications for which it is appropriate, derive the differential or integral equations for light transport, present calculation methods for solving them, and show output images for a data set representing a cloud. Special attention is given to calculation methods for the multiple scattering model.
IEEE Trans. Vis. Comput. Graph.	Multiscale Volume Representation by a DoG Wavelet.	Shigeru Muraki	1995	This article presents a method for decomposing volume data into 3D DoG (difference of Gaussians) functions by using the frame theory [1] of nonorthogonal wavelets. Since we can think of a DoG function as a pair of Gaussian functions, we can consider this method an automatic generation of Blinn¿s blobby objects [2]. We can also use this representation method for data compression by neglecting the insignificant coefficients, since the wavelet coefficients have significant values only where the volume density changes. Further, since the DoG function closely approximates a ¿2G (Laplacian of Gaussian) function, the representation can be considered a hierarchy of the 3D edges on different resolution spaces. Using the spherically symmetric feature of the 3D DoG function, we can easily visualize the 3D edge structure by the density reprojection method [3], [4]. We will apply our representation method to medical CT volume data and show its efficiency in describing the spatial structure of the volume.
IEEE Trans. Vis. Comput. Graph.	Obliq-3D: A High-Level, Fast-Turnaround 3D Animation System.	Marc Najork,Marc H. Brown	1995	This paper describes Obliq-3D, a high-level, fast-turnaround system for building 3D animations. Obliq-3D consists of an interpreted language that is embedded into a 3D animation library. This library is based on a few simple, yet powerful constructs that allow programmers to describe three-dimensional scenes and animations of such scenes. By virtue of its interpretive nature, Obliq-3D provides a fast-turnaround environment. The combination of simplicity and fast turnaround allows programmers to construct nontrivial animations quickly and easily.The paper is divided into three major parts. The first part introduces the basic concepts of Obliq-3D, using a series of graduated examples. The second part shows how the system can be used to implement Cone Trees. The third part develops a complete animation of Dijkstra¿s shortest-path algorithm.
IEEE Trans. Vis. Comput. Graph.	Visualization Takes its Place in the Scientific Community.	Gregory M. Nielson	1995	Visualization Takes its Place in the Scientific Community.
IEEE Trans. Vis. Comput. Graph.	Competent, Compact, Comparative Visualization of a Vortical Flow Field.	Hans-Georg Pagendarm,Birgit Walter	1995	In computational fluid dynamics, visualization is a frequently used tool for data evaluation, understanding of flow characteristics, and qualitative comparison to flow visualizations originating from experiments. Building on an existing visualization software system that allows for a careful selection of state-of-the-art visualization techniques and some extensions, it became possible to present various features of the data in a single image. The visualization shows vortex position and rotation as well as skin-friction lines, experimental oil-flow traces, shock-wave positions, and time surfaces. Animation provides natural perception of flow in combination with abstract representation of phenomena. By adding experimental flow visualization, a comparison between numerical simulation and wind-tunnel flow becomes possible up to a high level of detail. Since some of the underlying algorithms are not yet described in detail in the visualization literature, some experiences gained from the implementation are illustrated. The dedicated techniques which are illustrated here address specific properties of vector quantities in the flow field, such as the velocity vector or the friction vector. Image complexity is reduced by employing complex visualization methods. Thus, the room is created which is necessary to study the interaction of various phenomena.
IEEE Trans. Vis. Comput. Graph.	Real Time Responsive Animation with Personality.	Ken Perlin	1995	Building on principles from our prior work on procedural texture synthesis, we are able to create remarkably lifelike, responsively animated characters in real time. Rhythmic and stochastic noise functions are used to define time varying parameters that drive computer generated puppets. Because we are conveying just the ¿texture¿ of motion, we are able to avoid computation of dynamics and constraint solvers.The subjective impression of dynamics and other subtle influences on motion can be conveyed with great visual realism by properly tuned expressions containing pseudo-random noise functions. For example, we can make a character appear to be dynamically balancing herself, to appear nervous, or to be gesturing in a particular way.Each move has an internal rhythm, and transitions between moves are temporally constrained so that ¿impossible¿ transitions are precluded. For example, if while the character is walking we specify a dance turn, the character will always step into the turn onto the correct weight-bearing foot. An operator can make a character perform a properly connected sequence of actions, while conveying particular moods and attitudes, merely by pushing buttons at a high level.Potential uses of such high level ¿textural¿ approaches to computer graphic simulation include role playing games, simulated conferences, ¿clip animation,¿ graphical front ends for MUDs [15], [6], and synthetic performances.
IEEE Trans. Vis. Comput. Graph.	Dynamic Color Quantization of Video Sequences.	Evgeny Roytman,Craig Gotsman	1995	Dynamic Color Quantization of Video Sequences.
IEEE Trans. Vis. Comput. Graph.	A Unified Hierarchical Algorithm for Global Illumination with Scattering Volumes and Object Clusters.	François X. Sillion	1995	A Unified Hierarchical Algorithm for Global Illumination with Scattering Volumes and Object Clusters.
IEEE Trans. Vis. Comput. Graph.	Vision - An Architecture for Global Illumination Calculations.	Philipp Slusallek,Hans-Peter Seidel	1995	So far, the problem of global illumination calculation has almost exclusively been approached from an algorithmic point of view. In this paper we propose an architectural approach to global illumination. The proposed rendering architecture Vision is derived from a model of the physical rendering process, which is subsequently mapped onto an object-oriented hierarchy of classes. This design is powerful and flexible enough to support and exploit a large body of existing illumination algorithms for the simulation of various aspects of the underlying physical model. Additionally, the Vision architecture offers a platform for developing new algorithms and for combining them to create new rendering solutions.We discuss both abstract design as well as implementation issues. In particular, we give a detailed description of the global Lighting subsystem and show how algorithms for path tracing, bidirectional estimators, irradiance caching, hierarchical radiosity, wavelet radiosity, and wavelet radiance have been implemented within Vision.
IEEE Trans. Vis. Comput. Graph.	Visualization of Geometric Algorithms.	Ayellet Tal,David P. Dobkin	1995	This paper investigates the visualization of geometric algorithms. We discuss how limiting the domain makes it possible to create a system that enables others to use it easily. Knowledge about the domain can be very helpful in building a system which automates large parts of the user¿s task. A system can be designed to isolate the user from any concern about how graphics is done. The application need only specify ¿what¿ happens and need not be concerned with ¿how¿ to make it happen on the screen. We develop a conceptual model and a framework for experimenting with it. We also present a system, GASP, which implements this model. GASP allows quick generation of three-dimensional geometric algorithm visualizations, even for highly complex algorithms. It also provides a visual debugging facility for geometric computing. We show the utility of GASP by presenting a variety of examples.
IEEE Trans. Vis. Comput. Graph.	Calibration Requirements and Procedures for a Monitor-Based Augmented Reality System.	Mihran Tuceryan,Douglas S. Greer,Ross T. Whitaker,David E. Breen,Eric Rose,Klaus H. Ahlers,Chris Crampton	1995	Calibration Requirements and Procedures for a Monitor-Based Augmented Reality System.
IEEE Trans. Vis. Comput. Graph.	Octree-R: An Adaptive Octree for Efficient Ray Tracing.	Kyu-Young Whang,Ju-Won Song,Ji-Woong Chang,Ji-Yun Kim,Wan-Sup Cho,Chong-Mok Park,Il-Yeol Song	1995	Ray tracing requires many ray-obect intersection tests. A way of reducing the number of ray-object intersection tests is to subdivide the space occupied by objects into many non-overlapping subregions, called voxels, and to construct an octree for the subdivided space. In this paper, we propose the Octree-R, an octree-variant data structure for efficient ray tracing. The algorithm for constructing the Octree-R first estimates the number of ray-object intersection tests. Then, it partitions the space along the plane that minimizes the estimated number of ray-object intersection tests. We present the results of experiments for verifying the effectiveness of the Octree-R. In the experiment, the Octree-R provides a 4% to 47% performance gain over the conventional octree. The result shows the more skewed the object distribution (as is typical for real data), the more performance gain the Octree-R achieves.
IEEE Trans. Vis. Comput. Graph.	Grouping Volume Renderers for Enhanced Visualization in Computational Fluid Dynamics.	Roni Yagel,David S. Ebert,James N. Scott,Yair Kurzion	1995	This paper advocates the use of a group of renderers rather than any specific rendering method. We describe a bundle containing four alternative approaches to visualizing volume data. One new approach uses realistic volumetric gas rendering techniques to produce photo-realistic images and animations. The second uses ray casting that is based on a simpler illumination model and is mainly centered around a versatile new tool for the design of transfer functions. The third method employs a simple illumination model and rapid rendering mechanisms to provide efficient preview capabilities. The last one reduces data magnitude by displaying the most visible components and exploits rendering hardware to provide real time browsing capabilities. We show that each rendering tool provides a unique service and demonstrate the combined utility of our group of volume renderers in computational fluid dynamic (CFD) visualization. While one tool allows the explorer to render rapidly for navigation through the data, another tool allows one to emphasize data features (e.g., shock waves), and yet another tool allows one to realistically render the data. We believe that only through the deployment of groups of renderers will the scientist be well served and equipped to form numerous perspectives of the same dataset, each providing different insights into the data.
IEEE Trans. Vis. Comput. Graph.	Hybrid Scan-Conversion of Circles.	Chengfu Yao,Jon G. Rokne	1995	Conventional algorithms for scan-conversion of circles select one pixel in each iteration. Run-length slice circle algorithms have therefore been suggested. These algorithms determine a run of pixels in each iteration. The speed of scan-conversion is therefore increased due to I/O. A hybrid approach to the scan-conversion of circles is presented. The new approach combines the advantages of the two methods into a hybrid algorithm. Speedup is achieved in the hybrid algorithm not only due to the reduction in the number of I/O operations, but also due to a reduction in the number of arithmetic operations.
IEEE Trans. Vis. Comput. Graph.	Volume Rendering of DCT-Based Compressed 3D Scalar Data.	Boon-Lock Yeo,Bede Liu	1995	This paper proposes a scheme to perform volume rendering from compressed scalar data. Instead of decompressing the entire data set before rendering, blocks of data are decompressed as needed. Discrete cosine transform based compression technique is used to illustrate the method. The data is partitioned into overlapping blocks to permit local rendering and allow easy parallelization. Compression by factor of 20 to 30 produces rendering virtually indistinguishable from rendering using the original uncompressed data. Speedup is obtained by making use of spatial homogeneity detected in the transform domain. Rendering time using the proposed approach is less than that of direct rendering from the entire uncompressed data. The proposed method thus offers an attractive option to reduce storage, computation, and transmission overhead of otherwise huge data sets.
CVPR	Pattern Rejection.	Simon Baker,Shree K. Nayar	1996	"The efficiency of pattern recognition is particularly crucial in two scenarios; whenever there are a large number of classes to discriminate, and, whenever recognition must be performed a large number of times. We propose a single technique, namely, pattern rejection, that greatly enhances efficiency in both cases. A rejector is a generalization of a classifier, that quickly eliminates a large fraction of the candidate classes or inputs. This allows a recognition algorithm to dedicate its efforts to a much smaller number of possibilities. Importantly, a collection of rejectors may be combined to form a composite rejector, which is shown to be far more effective than any of its individual components. A simple algorithm is proposed for the construction of each of the component rejectors. Its generality is established through close relationships with the Karhunen-Loéve expansion and Fisher's discriminant analysis. Composite rejectors were constructed for two representative applications, namely, appearance matching based object recognition and local feature detection. The results demonstrate substantial efficiency improvements over existing approaches, most notably Fisher's discriminant analysis."
CVPR	Extracting Salient Curves from Images: An Analysis of the Saliency Network.	T. D. Alter,Ronen Basri	1996	The Saliency Network proposed by Shashua and Ullman is a well-known approach to the problem of extracting salient curves from images while performing gap completion. This paper analyzes the Saliency Network. The Saliency Network is attractive for several reasons. First, the network generally prefers long and smooth curves over short or wiggly ones. While computing saliencies, the network also fills in gaps with smooth completions and tolerates noise. Finally, the network is locally connected, and its size is proportional to the size of the image. Nevertheless, our analysis reveals certain weaknesses with the method. In particular, we show cases in which the most salient element does not lie on the perceptually most salient curve. Furthermore, in some cases the saliency measure changes its preferences when curves are scaled uniformly. Also, we show that for certain fragmented curves the measure prefers large gaps over a few small gaps of the same total size. In addition, we analyze the time complexity required by the method. We show that the number of steps required for convergence in serial implementations is quadratic in the size of the network, and in parallel implementations is linear in the size of the network. We discuss problems due to coarse sampling of the range of possible orientations. We show that with proper sampling the complexity of the network becomes at least cubic in the size of the network. Finally, we consider the possibility of using the Saliency Network for grouping. We show that the Saliency Network recovers the most salient curve efficiently, but it has problems with identifying any salient curve other than the most salient one.
CVPR	Extraction of maximal inscribed disks from discrete Euclidean distance maps.	Yaorong Ge,J. Michael Fitzpatrick	1996	Extraction of maximal inscribed disks from discrete Euclidean distance maps.
CVPR	Real-time Extraction of connected Components in 3-D Sonar Range Images.	Per Gunnar Auran,Kjell E. Malvig	1996	This paper describes an efficient algorithm for the segmentation of echo clusters within a dynamic 3-D sonar image. The sensor centered image is an echo management framework grouping sonar returns into spherical cells, and allowing real-time organisation of 3-D range data using inexpensive equipment. Each cell acts as a spatial key to the features related to this location. The spherical representation is effectively exploited for segmentation using an approach motivated from connected components analysis in binary video images. A fast algorithm (linear in time complexity) based on cell connectivity between sonar beams is presented, including methods for coping with sparse data.
CVPR	Finding Corresponding Points Based on Bayesian Triangulation.	Anand S. Bedekar,Robert M. Haralick	1996	In this paper, we consider the problems of finding corresponding points from multiple perspective projection images (the correspondence problem), and estimating the 3-D point from which these points have arisen (the triangulation problem). We pose the triangulation problem as that of finding the Bayesian maximum a posteriori estimate of the 3-D point, given its projections in N images, assuming a Gaussian error model for the image point co-ordinates and the camera parameters. We solve this by an iterative steepest descent method. We then consider the correspondence problem as a statistical hypothesis verification problem. Given a set of 2-D points, under the hypothesis that the points are in correspondence, the MAP estimate of the 3-D point is computed. Based on the MAP estimate, we derive a statistical test for verifying this hypothesis. To find sets of corresponding points when multiple points in each of N images are given, we propose a method that does the Bayesian triangulation and hypothesis verification on each N-tuple of points, selecting those that pass the hypothesis test. We characterize the performance of the Bayesian triangulation in terms of the average distance of the triangulated 3-D point from the true 3-D point, and of the point correspondence method in terms of its misdetection and false alarm rates.
CVPR	What is the set of images of an object under all possible lighting conditions?	Peter N. Belhumeur,David J. Kriegman	1996	The appearance of a particular object depends on both the viewpoint from which it is observed and the light sources by which it is illuminated. If the appearance of two objects is never identical for any pose or lighting conditions, then-in theory - the objects can always be distinguished or recognized. The question arises: What is the set of images of an object under all lighting conditions and pose? In this paper, we consider only the set of images of an object under variable illumination (including multiple, extended light sources and attached shadows). We prove that the set of n-pixel images of a convex object with a Lambertian reflectance function, illuminated by an arbitrary number of point light sources at infinity, forms a convex polyhedral cone in IR/sup n/ and that the dimension of this illumination cone equals the number of distinct surface normals. Furthermore, we show that the cone for a particular object can be constructed from three properly chosen images. Finally, we prove that the set of n-pixel images of an object of any shape and with an arbitrary reflectance function, seen under all possible illumination conditions, still forms a convex cone in IR/sup n/. These results immediately suggest certain approaches to object recognition. Throughout this paper, we offer results demonstrating the empirical validity of the illumination cone representation.
CVPR	Interpreting and representing tabular documents.	Juan F. Arias,Atul K. Chhabra,Vishal Misra	1996	This paper describes a methodology to interpret the information from telephone company DSX assignment table drawings. Horizontal lines are found using an efficient algorithm that works over the run-length encoded representation of the image. For vertical lines, the image is transposed using an efficient method we developed, and the algorithm for horizontal lines is applied again. Using the information about the lines, the tabular structures are extracted by finding biconnected components on the graph formed by the lines and their intersections. A methodology has also been developed for the representation of end access to the entries inside the tables.
CVPR	Multilinear Constraints in the Infinitesimal-time Cas.	Kalle Åström,Anders Heyden	1996	In this paper we study the infinitesimal-time case of the so called multilinear constraints that exist for each subsequence in a sequence of images. These constraints link the infinitesimal motion of the image points with the infinitesimal viewer motion. The analysis is done both for calibrated and uncalibrated cameras. Two simplifications are also presented for the uncalibrated camera case. One simplification is made using affine reduction and kinetic depth. The second simplification is based upon a projective reduction with respect to the image of a planar patch.
CVPR	Shadows and shading flow fields.	Pierre Breton,Steven W. Zucker	1996	Many presume that parsing the shadows out of an image is a high-level task, because of the global nature of the shadow formation process. But shape-from-shading algorithms are low-level, in the sense that they seek solutions (surface normals or depth values) directly from image intensities. A dilemma arises: since shape-from-shading involves an illumination term, shadows must first be identified. We show that a structure intermediate between intensities and surfaces-the shading flow field-provides a solution to this dilemma. Our analysis is based on the observation that the geometric information that can be derived from images supports different inferences than the photometric information, and our specific goal will be to articulate this geometric structure and to show how shading flow fields can be reliably computed.
CVPR	Calibration of a Foveated Wide-Angle Lens on an Active Vision Head.	Luc Berthouze,Sebastien Rougeaux,Florent Chavand,Yasuo Kuniyoshi	1996	Inspired by the properties of the humam visual system, a new active vision system called ESCHeR (Etl Stereo Compact Head For Robot Vision) has been recently implemented with foveated wide-angle lenses. The lenses exhibit a wide field of view along with a space-varying resolution for facilitating both detection and close observation. However, to handle such optical properties and achieve basic eye movement functions, new calibration methods are needed. Therefore, two novel and online techniques are presented that in one case perform a global identification of the optical process through artificial neural techniques and in the other case compute the physical parameters by using environmental feature-tracking and controlled rotations of the cameras. Self-alignment of the cameras is also achieved using a similar technique.
CVPR	Feature Correspondence by Interleaving Shape and Texture Computations.	David Beymer	1996	"The correspondence problem in computer vision is basically a matching task between two or more sets of features. In this paper, we introduce a vectorized image representation, which is a feature-based representation where correspondence has been established with respect to a reference image. The representation consists of two image measurements made at the feature points: shape and texture. Feature geometry, or shape, is represented using the (x, y) locations of features relative to the some standard reference shape. Image grey levels, or texture, are represented by mapping image grey levels onto the standard reference shape. Computing this representation is essentially a correspondence task, and in this paper we explore an automatic technique for ""vectorizing"" face images. Our face vectorizer alternates back and forth between computation steps for shape and texture, and a key idea is to structure the two computations so that each one uses the output of the other. In addition to describing the vectorizer, an application to the problem of facial feature detection will be presented."
CVPR	Ordinal Measures for Visual Correspondence.	Dinkar N. Bhat,Shree K. Nayar	1996	We present ordinal measures for establishing image correspondence. Linear correspondence measures like correlation and the sum of squared differences are known to be fragile. Ordinal measures, which are based on relative ordering of intensity values in windows, have demonstrable robustness to depth discontinuities, occlusion and noise. The relative ordering of intensity values in each window is represented by a rank permutation which is obtained by sorting the corresponding intensity data. By using a novel distance metric between the rank permutations, we arrive at ordinal correlation coefficients. These coefficients are independent of absolute intensity scale, i.e they are normalized measures. Further, since rank permutations are invariant to monotone transformations of the intensity values, the coefficients are unaffected by nonlinear effects like gamma variation between images. We have developed a simple algorithm for their efficient implementation. Experiments suggest the superiority of ordinal measures over existing techniques under non-ideal conditions. Though we present ordinal measures in the context of stereo, they serve as a general tool for image matching that is applicable to other vision problems such as motion estimation and image registration.
CVPR	Recognition of Planar Object Classes.	Michael C. Burl,Pietro Perona	1996	We present a new framework for recognizing planar object classes, which is based on local feature detectors and a probabilistic model of the spatial arrangement of the features. The allowed object deformations are represented through shape statistics, which are learned from examples. Instances of an object in an image are detected by finding the appropriate features in the correct spatial configuration. The algorithm is robust with respect to partial occlusion, detector false alarms, and missed features. A 94\% success rate was achieved for the problem of locating quasi-frontal views of faces in cluttered scenes.
CVPR	On a spectral attentional mechanism.	Philippe Burlina,Bruce Lin,Rama Chellappa	1996	"This paper describes an attentional mechanism based on the interpretation of spectral signatures for detecting regular object configurations in areas of an image delineated using context information. The proposed global operator relies on the spectral analyse's of edge structure and exploits spatial as well as frequency domain constraints derived from known geometrical models of monitored objects. A decision theoretic method for learning decision regions is presented. Applications of this mechanism are demonstrated for several aerial image interpretation tasks. Specific examples are described for detecting vehicle formations (such as convoys), qualifying the geometry of detected formations, or monitoring the occupancy of regions of interest (such as parking areas, roads, or open areas). Experiments and sensitivity analysis results are reported."
CVPR	Recognition via consensus of local moments of brightness and orientation.	J. Brian Burns	1996	This study combines two useful methods in recognition: consensus or voting-based approaches and moment-based representations. Matches between image patches are generated using a Gaussian-weighted moment encoding of the patches and a feature indexing process. Each match implies an object 3D position and orientation (pose) and generates a vote for this pose. Recognition is accomplished by detecting significant clusters of votes in pose space. This combined method is an improvement over voting and moment methods in isolation. Using image brightness moments, the idea is successfully demonstrated on examples of human faces undergoing full 3D pose change, as well as changes in features such as talking and blinking. The idea is then extended to moments of local texture orientation and successfully demonstrated under large variations in lighting nature and geometry.
CVPR	Autonomous recognition: driven by ambiguit.	Franco Callari,Frank P. Ferrie	1996	Recognition ambiguity, due to noisy measurements and uncertain object models, can be quantified and actively used by an autonomous agent to efficiently gather new data and improve its information about the environment. In this work an information-based utility measure is used to derive from a learned classification of shape models an efficient data collection strategy, specifically aimed at increasing classification confidence when recognizing uncertain shapes. Promising simulation results are presented and discussed.
CVPR	Lie groups, Space-Variant Fourier Analysis and the Exponential Chirp Transform.	Giorgio Bonmassar,Eric L. Schwartz	1996	"The use of visual representations in which retinal neurons receptive fields are not constant over the visual field is universal in the visual systems of higher vertebrates, and is coming to play an important role in active vision applications. The breaking of translation symmetry that is unavoidably associated with non-uniform sampling presents a major algorithmic complication for image processing. In this paper we use a Lie group approach to derive a kernel which provides a quasi-shift (i.e. approximate shift) invariant template matching capability, under normal convolution in the distorted (range) coor- dinates of the non-uniform mapping. We work out the special case of the log- polar mapping, which is of great interest in vision; in this case, we call the associated linear integral transform the ``exponential chirp transform'' (ECT). The method is, however, general for other forms of mapping, or warp, function."
CVPR	A hierarchical approach to high resolution edge contour reconstruction.	Stefano Casadei,Sanjoy K. Mitter	1996	"Efficient edge detection algorithms such as Canny's (1986) fail near curve singularities. Moreover, the standard linking algorithms used on top of these detectors often fail because of instabilities in the tracking process (due to multiple responses to the same edge and interference of nearby edges). We propose a hierarchical approach to edge detection based on a graph stabilization method that allows bifurcation resolution in stages. Curve singularities are recovered at the last stage by using ""top-down"" feedback to select the best curve connections."
CVPR	Multi-Stage Target Recognition Using Modular Vector Quantizers and Multilayer Perceptrons.	LipChen Alex Chan,Nasser M. Nasrabadi,Vincent Mirelli	1996	Multi-Stage Target Recognition Using Modular Vector Quantizers and Multilayer Perceptrons.
CVPR	Indexing to 3D Model Aspects using 2D Contour Features.	Jin-Long Chen,George C. Stockman	1996	We present a shape-based method of indexing to model aspects from a single intensity image. Objects are assumed to be rigid. A model aspect is represented by a 2 1/2D edgemap and the parts of the object silhouette. Part decomposition is derived from a codon representation of the object silhouette. Invariant features extracted from each part are then used to index into a hash table to generate model-aspect hypotheses. Knowledge about parts is incorporated in voting schemes to order hypotheses for efficient verification of candidate models. Verification of model-aspect hypotheses is carried out by an alignment algorithm that is robust to partial occlusion. Results of tests using 658 model aspects from 100 objects demonstrate that accurate recognition can be achieved with very few verification attempts.
CVPR	Competitive Mixture of Deformable Models for Pattern Classification.	Kwok-Wai Cheung,Dit-Yan Yeung,Roland T. Chin	1996	Following the success of applying deformable models to feature extraction, a natural next step is to apply such models to pattern classification. Recently, we have cast a deformable model under a Bayesian framework for classification, giving promising results. However, deformable model methods are computationally expensive due to the required iterative optimization process. The problem is even more severe when there are a large number of models (e.g., for character recognition), because each of them has to deform and match with the input data before a final classification can be derived. In this paper, we propose to combine the deformable models into a mixture, in which the individual models compete with each other to survive the matching process during classification. Models that do not compete well are eliminated early, thus allowing substantial savings in computation. This process of competition- elimination has been applied to handwritten digit recognition in which significant speedup can be achieved without sacrificing recognition accuracy.
CVPR	Global Minimum for Active Contour Models: A Minimal Path Approach.	Laurent D. Cohen,Ron Kimmel	1996	A new boundary detection approach for shape modeling is presented. It detects the global minimum of an active contour model&lsquo;s energy between two end points. Initialization is made easier and the curve is not trapped at a local minimum by spurious edges. We modify the &ldquo;snake&rdquo; energy by including the internal regularization term in the external potential term. Our method is based on finding a path of minimal length in a Riemannian metric. We then make use of a new efficient numerical method to find this shortest path.It is shown that the proposed energy, though based only on a potential integrated along the curve, imposes a regularization effect like snakes. We explore the relation between the maximum curvature along the resulting contour and the potential generated from the image.The method is capable to close contours, given only one point on the objects&lsquo; boundary by using a topology-based saddle search routine.We show examples of our method applied to real aerial and medical images.
CVPR	A Space-Sweep Approach to True Multi-Image Matching.	Robert T. Collins	1996	"The problem of determining feature correspondences across multiple views is considered. The term ""true multi-image"" matching is introduced to describe techniques that make full and efficient use of the geometric relationships between multiple images and the scene. A true multi-image technique must generalize to any number of images, be of linear algorithmic complexity in the number of images, and use all the images in an equal manner. A new space-sweep approach to true multi-image matching is presented that simultaneously determines 2D feature correspondences and the 3D positions of feature points in the scene. The method is based on the premise that areas of space where several viewing rays intersect are the likely locations of observed 3D scene features. It is shown that the intersections of viewing rays with a plane sweeping through space can be determined very efficiently, and a statistical model is developed to tell how likely it is that a given number of viewing rays will pass through an area of the plane by chance. The method is illustrated on a seven-image matching example from the aerial image domain. The ACM Portal is published by the Association for Computing Machinery. Copyright © 2010 ACM, Inc. Terms of Usage Privacy Policy Code of Ethics Contact Us Useful downloads: Adobe Acrobat QuickTime Windows Media Player Real Player"
CVPR	Feature-Based Face Recognition Using Mixture-Distance.	Ingemar J. Cox,Joumana Ghosn,Peter N. Yianilos	1996	Feature-Based Face Recognition Using Mixture-Distance.
CVPR	Hand segmentation using learning-based prediction and verification for hand sign recognition.	Yuntao Cui,John J. Weng	1996	This paper presents a prediction-and-verification segmentation scheme wing attention images from multiple fixations. A major advantage of this scheme is that it can handle a large number of different deformable objects presented in complex backgrounds. The scheme is also relatively efficient since the segmentation is guided by the past knowledge through a prediction-and-verification scheme. The system has been tested to segment hands in the sequences of intensity images, where each sequence represents a hand sign. The experimental result showed a 95% correct segmentation rate with a 3% false rejection rate.
CVPR	Active Intrinsic Calibration Using Vanishing Points.	Konstantinos Daniilidis,Jörg Ernst	1996	We propose a new method for the estimation of the intrinsic parameters of an active camera. During a fixed axis camera rotation every point is moving on a conic section. If the point used is a vanishing point the conic section is invariant to possible translations of the observer. Given the rotation axis and the inter-frame correspondence of a set of parallel lines we are able to compute the intrinsic parameters without knowledge of the rotation angles. We propagate the error covariances and we remove the bias in the computation of the conic. We experimentally study the sensitivity of calibration to the amount of rotation and we compare our performance to the performance of a recent active calibration technique.
CVPR	Active Face Tracking and Pose Estimation in an Interactive Room.	Trevor Darrell,Baback Moghaddam,Alex Pentland	1996	"We demonstrate real-time face tracking and pose estimation in an unconstrained office environment with an active foveated camera. Using vision routines previously implemented for an interactive environment, we determine the spatial location of a user's head and guide an active camera to obtain foveated images of the face. Faces are analyzed using a set of eigenspaces indexed over both pose and world location. Closed loop feedback from the estimated facial location is used to guide the camera when a face is present in the foveated view. Our system can detect the head pose of an unconstrained user in real-time as he or she moves about an open room."
CVPR	Convexity Analysis of Active Contour Problems.	Christos Davatzikos,Jerry L. Prince	1996	Convexity Analysis of Active Contour Problems.
CVPR	The Integration of Optical Flow and Deformable Models with Applications to Human Face Shape and Motion Estimation.	Douglas DeCarlo,Dimitris N. Metaxas	1996	We present a formal methodology for the integration of optical flow and deformable models. The optical flow constraint equation provides a non-holonomic constraint on the motion of the deformable model. In this augmented system, forces computed from edges and optical flow are used simultaneously. When this dynamic system is solved, a model-based least-squares solution for the optical flow is obtained and improved estimation results are achieved. The use of a 3-D model reduces or eliminates problems associated with optical flow computation. This approach instantiates a general methodology for treating visual cues as constraints on deformable models. We apply this framework to human face shape and motion estimation. Our 3-D deformable face model uses a small number of parameters to describe a rich variety of face shapes and facial expressions. We present experiments in extracting the shape and motion of a face from image sequences.
CVPR	The Integration of Optical Flow and Deformable Models with Applications to Human Face Shape and Motion Estimation.	Douglas DeCarlo,Dimitris N. Metaxas	1996	We present a formal methodology for the integration of optical flow and deformable models. The optical flow constraint equation provides a non-holonomic constraint on the motion of the deformable model. In this augmented system, forces computed from edges and optical flow are used simultaneously. When this dynamic system is solved, a model-based least-squares solution for the optical flow is obtained and improved estimation results are achieved. The use of a 3-D model reduces or eliminates problems associated with optical flow computation. This approach instantiates a general methodology for treating visual cues as constraints on deformable models. We apply this framework to human face shape and motion estimation. Our 3-D deformable face model uses a small number of parameters to describe a rich variety of face shapes and facial expressions. We present experiments in extracting the shape and motion of a face from image sequences.
CVPR	From projective to Euclidean reconstruction.	Frederic Devernay,Olivier D. Faugeras	1996	To make a Euclidean reconstruction of the world seen through a stereo rig, we can either use a calibration grid, and the results will rely on the precision Of the grid and the extracted points of interest, or use self-calibration. Past work on self-calibration is focussed on the use of only one camera, and gives sometimes very unstable results. In this paper, we use a stereo rig which is supposed to be weakly calibrated using a method such as the one described in Deriche et al. (1994). Then, by matching two sets of points of the same scene reconstructed from different points of view, we try to find both the homography that maps the projective reconstruction to the Euclidean space and the displacement from the first set of points to the second set of points. We present results of the Euclidean reconstruction of a whole object from uncalibrated cameras using the method proposed here.
CVPR	Sparse Representations for Image Decomposition with Occlusions.	Michael J. Donahue,Davi Geiger,Tyng-Luh Liu,Robert A. Hummel	1996	"We study the problem of how to detect ""interesting objects'' appeared in a given image, I. Our approach is to treat it as a function approximation problem based on an over-redundant basis, and also account for occlusions, where the basis superposition principle is no longer valid. Since the basis (a library of image templates) is over-redundant, there are infinitely many ways to decompose I. We are motivated to select a sparse/compact representation of I, and to account for occlusions and noise. We then study a greedy and iterative ""weighted Lp Matching Pursuit"" strategy, with 0"
CVPR	An MIMD Computing Platform for a Hierarchical Foveal Machine Vision System.	Fenglei Du,Andrew Izatt,Cesar Bandera	1996	A Multiple Instruction Multiple Data (MIMD) parallel computing platform built upon a network of TMS320C40/44s (C40/C44) for real-time image processing of a hierarchical foveal machine vision (HFMV) system is described in this paper. The architecture of the system, the parallel algorithm development environment, and strategies to map tasks into the computing platform are described. The platform supports both static and dynamic computing resource allocation. The performance of the computing platform is illustrated by examples.
CVPR	Space Scale Localization, Blur, and Contour-Based Image Coding.	James H. Elder,Steven W. Zucker	1996	We have recently proposed a scale-adaptive algorithm for reliable edge detection and blur estimation. The algorithm produces a contour code which consists of estimates of position, brightness, contrast and blur for each edge point in the image. Here we address two questions: 1. Can scale adaptation be used to achieve precise localization of blurred edges? 2. How much of the perceptual content of an image is carried by the 1-D contour code? We report an efficient algorithm for subpixel localization, and show that local scale control allows excellent precision even for highly blurred edges. We further show how local scale control can quantitatively account for human visual acuity of blurred edge stimuli. To address the question of perceptual content, we report an algorithm for inverting the contour code to reconstruct an estimate of the original image. While reconstruction based on edge brightness and contrast alone introduces significant artifact, restitution of the local blur signal is shown to produce perceptually accurate reconstructions.
CVPR	A Robust Clustering Algorithm Based on Competitive Agglomeration and Soft Rejection of Outliers.	Hichem Frigui,Raghu Krishnapuram	1996	We present a new clustering algorithm that addresses two major issues associated with conventional partitional clustering: the difficulty in determining the number of clusters, and the sensitivity to noise and outliers. The proposed algorithm determines the number of clusters by a process of competitive agglomeration. Noise immunity is achieved by integrating concepts from robust statistics into the algorithm. The proposed approach can incorporate different distance measures in the objective function to find an unknown number of clusters of various types including lines, planes and surfaces.
CVPR	The use of hybrid models to recover cardiac wall motion in tagged MR images.	Gareth Funka-Lea,Alok Gupta	1996	We present a new algorithm for the automatic recovery of tag grid-line intersections in Tagged MR images of the left ventricle of the heart. Our method uses an active spring mesh to capture local properties of the motion and a global motion model to capture the global coherence of the motion. We recover the global component of the motion using robust estimation. Different motion models have been developed for short and long axis views of the heart. The algorithm has been tested on healthy and pathological data.
CVPR	3-D model-based tracking of humans in action: a multi-view approach.	Dariu Gavrila,Larry S. Davis	1996	We present a vision system for the 3-D model-based tracking of unconstrained human movement. Using image sequences acquired simultaneously from multiple views, we recover the 3-D body pose at each time instant without the use of markers. The pose-recovery problem is formulated as a search problem and entails finding the pose parameters of a graphical human model whose synthesized appearance is most similar to the actual appearance of the real human in the multi-view images. The models used for this purpose are acquired from the images. We use a decomposition approach and a best-first technique to search through the high dimensional pose parameter space. A robust variant of chamfer matching is used as a fast similarity measure between synthesized and real edge images. We present initial tracking results from a large new Humans-In-Action (HIA) database containing more than 2500 frames in each of four orthogonal views. They contain subjects involved in a variety of activities, of various degrees of complexity, ranging from the more simple one-person hand waving to the challenging two-person close interaction in the Argentine Tango.
CVPR	Visual Organization for Figure/Ground Separation.	Davi Geiger,Krishnan Kumaran,Laxmi Parida	1996	A common factor in all illusory contour figures is the perception of a surface occluding part of a background. In our previous work we have shown that by detecting junctions and assigning a proper set of hypothesis at each junction, we could diffuse this information and obtain surface reconstructions where the surface boundaries represented illusory contours. Amodal completions emerge at the overlapping surfaces. We here address the problem of selecting the best image organization (set of hypothesis). We propose two optimization criteria, one based on a coherence measure between pairs of junctions (correlation between the diffusion of each pair) and another one based on an entropy measure (sharpness of the reconstruction). We show their similarity and a statistical physics approach to select the best organization. The experiments suggest that despite the large number of possible organizations our approach may take a few steps to select the best organization (starting from random organizations).
CVPR	Target Detection in Foveal ATR Systems.	Sugata Ghosal,Douglas C. McKee	1996	Automatic target recognition (ATR) applications require simultaneously a wide field of view (FOV) for better detection and situation awareness, high resolution for target recognition and threat assessment, and high frame rate for detecting brief events and disambiguating frame-to-frame correlation. Uniformly sampling the entire FOV at recognition resolution is simply wasteful in ATR scenarios with localized regions of interest (ROIs). Foveal data acquisition with space-variant sampling and context- sensitive sensor articulation is highly optimized for active ATR applications. We propose a multiscale local Zernike filter-based front end target detection technique for a commercially feasible foveal sensor topology with piecewise constant resolution profile. Anisotropic heat diffusion is employed for preprocessing of the foveal data. Expansion template matching is used to derive a detection filter that optimizes the discriminant signal-to-noise ratio (SNR). Results are presented with simulated foveal imagery derived from real uniform acuity FLIR data.
CVPR	Real-time tracking of image regions with changes in geometry and illumination.	Gregory D. Hager,Peter N. Belhumeur	1996	Historically, SSD or correlation-based visual tracking algorithms have been sensitive to changes in illumination and shading across the target region. This paper describes methods for implementing SSD tracking that is both insensitive to illumination variations and computationally efficient. We first describe a vector-space formulation of the tracking problem, showing how to recover geometric deformations. We then show that the same vector space formulation can be used to account for changes in illumination. We combine geometry and illumination into an algorithm that tracks large image regions on live video sequences using no more computation than would be required to trade with no accommodation for illumination changes. We present experimental results which compare the performance of SSD tracking with and without illumination compensation.
CVPR	/spl lambda//spl tau/-space representation of images and generalized edge detector.	Muhittin Gökmen,Anil K. Jain	1996	An image clad surface representation based on regularization theory is introduced in this paper. This representation is based on a hybrid model derived from the physical membrane and plate models. The representation, called the /spl lambda//spl tau/-representation, has two dimensions; one dimension represents smoothness or scale while the other represents the continuity of the image or surface. It contains images/surfaces sampled both in scale space and the weighted Sobolev space of continuous functions. Thus, this new representation can be viewed as an extension of the well-known scale space representation. We have experimentally shown that the proposed hybrid model results in improved results compared to the two extreme constituent model, i.e., the membrane and the plate models. Based on this hybrid model, a generalized edge detector (GED) which encompasses most of the well-known edge detectors under a common framework is developed. The existing edge detectors can be obtained from the generalized edge detector by simply specifying the valves of two parameters, one of which controls the shape of the filter (/spl tau/) and the other controls the scale of the filter (/spl lambda/). By sweeping the valves of these two parameters continuously, one can generate an edge representation in the /spl lambda//spl tau/ space, which is very useful for developing a goal-directed edge detection scheme for a specific task. The proposed representation and the edge detector have been evaluated qualitatively and quantitatively on several different types of image data such as intensity, range and stereo images.
CVPR	Graph matching by graduated assignment.	Steven Gold,Anand Rangarajan	1996	"A new algorithm for graph matching, which uses graduated assignment is presented, along with experimental results demonstrating large improvements in speed and accuracy over previous techniques. The softassign, a novel constraint satisfaction technique, is applied to a new graph matching energy function that uses a robust, sparse distance measure between the links of the two graphs. The softassign, which has emerged out of the neural network/statistical physics framework enforces two-way (assignment) constraints without the use of penalty terms. The algorithm's low order computational complexity [0(lm), where l and m are the number of links in the two graphs] compares favorably with most competing approaches. The method, not restricted to any special class of graph, is applied to subgraph isomorphism, weighted graph matching, and attributed relational graph matching. Experiments on graphs generated from images and on randomly generated graphs, including benchmarks against a relation labeling algorithm and an algorithm employing Potts glass dynamics are reported. Over twenty-five thousand experiments were conducted. No comparable results have been reported by any other graph matching algorithm before in the research literature."
CVPR	Mirror and Point Symmetry under Perspective Skewing.	Luc J. Van Gool,Theo Moons,Marc Proesmans	1996	Over recent years, symmetry research has shifted from the detection of affinely to perspectively skewed mirror symmetry. Also, links between invariance research and symmetry-specific geometric constraints have been established. The paper aims to contribute to both strands. Several sets of symmetry specific invariants are derived, that can be used in different situations, depending on the a priori assumptions made. It is also argued that all the results directly apply to the case of perspectively skewed point symmetry.
CVPR	Inference of segmented, volumetric shape from three intensity images.	Parag Havaldar,Gérard G. Medioni	1996	"We present a method to infer segmented and full volumetric descriptions of objects from intensity images. We use three weakly calibrated images from closely spaced viewpoints as input. Deriving full volumetric descriptions requires the development of robust inference rules. The inference rules are based on local properties of generalized cylinders (GCs). We first detect groups in each image based on proximity, parallelism and symmetry. The groups in the three images are matched and their contours are labelled as ""true"" and ""limb"" edges. We use the information about groups and the label associated with their contours to recover visible surfaces and their surface axes. To extract the complete volume in terms of a GC, we need to infer the GC axis, its cross section and the scaling function. The properties of straight and curved axis generalized cylinders are used locally on the visible surfaces to obtain the GC axis. The cross section is recovered if seen in the images, else it is inferred using the visible surfaces and GC properties. We consider groups with true edges, limb edges or a combination of both. The final descriptions are volumetric and in terms of parts. Sometimes, when not enough information is present to make volumetric inferences, the descriptions remain at the surface level. We demonstrate results on real images of moderately complex objects with texture and shadows."
CVPR	Using physics-based invariant representations for the recognition of regions in multispectral satellite images.	Glenn Healey,Amit Jain	1996	We present a set of algorithms and a search strategy for the robust content- based retrieval of multispectral satellite images. Since the property of interest in these images is usually the physical characteristics of ground cover, we use representations and methods that are invariant to illumination and atmospheric conditions. The representations and algorithms are derived for this application from a physical model for the formation of multispectral satellite images. The use of several representations and algorithms is necessary to interpret the diversity of physical and geometric structure in these images. Algorithms are used that exploit multispectral distributions, multispectral spatial structure, and labeled classes. The performance of the system is demonstrated on a large set of multispectral satellite images taken over different areas of the United States under different illumination and atmospheric conditions.
CVPR	Comparison of Edge Detectors: A Methodology and Initial Study.	Michael D. Heath,Sudeep Sarkar,Thomas A. Sanocki,Kevin W. Bowyer	1996	Because of the difficulty of obtaining ground truth for real images, the traditional technique for comparing low-level vision algorithms is to present image results, side by side, and to let the reader subjectively judge the quality. This is not a scientifically satisfactory strategy. However, human rating experiments can be done in a more rigorous manner, to provide useful quantitative conclusions. We present a paradigm based on experimental psychology and statistics, in which humans rate the output of low level vision algorithms. We demonstrate the proposed experimental strategy by comparing four well known edge detectors: Canny, Nalwa-Binford, Sarkar-Boyer, and Sobel. We answer the following questions: Is there a statistically significant difference in edge detector outputs as perceived by humans? Do the edge detection results of an operator vary significantly with the choice of its parameters? For each detector, is it possible to choose a single set of optimal parameters for all the images without significantly affecting the edge output quality? Does an edge detector produce edges of the same quality for all images, or does the edge quality vary with the image?
CVPR	Canonical Decomposition of Steerable Functions.	Yacov Hel-Or,Patrick C. Teo	1996	This paper describes a general mathematical formulation for the problem of constructing steerable functions. The formulation is based on Lie group theory and is thus applicable to transformations which are Lie groups, such as, rotation, translation, scaling, and affine transformation. For one-parameter and Abelian multi-parameter Lie transformation groups, a canonical decomposition of all possible steerable functions, derived using the Jordan decomposition of matrices, is developed. It is shown that any steerable function under Lie transformation groups can be described using this decomposition. Finally, a catalog of steerable functions for several common multi-parameter image transformation groups is also provided.
CVPR	Invariant histograms and deformable template matching for SAR target recognition.	Katsushi Ikeuchi,Takeshi Shakunaga,Mark D. Wheeler,Taku Yamazaki	1996	Recognizing a target in synthetic-aperture radar (SAR) images is an important, yet challenging, application of the model-based vision technique. This paper describes a model-based SAR recognition system based on invariant histograms and deformable template matching techniques. An invariant histogram is a histogram of invariant values defined by geometric features such as points and lines in SAR images. Although a few invariants are sufficient to recognize a target, we use a histogram of all invariant values given by all possible target feature pairs. This redundant histogram enables robust recognition under severe occlusions typical in SAR recognition scenarios. Multi-step deformable template matching examines the existence of an object by superimposing templates over potential energy field generated from images or primitive features. It determines the template configuration which has the minimum deformation and the best alignment of the template with features. The deformability of the template absorbs the instability of SAR features. We have implemented the system and evaluated the system performance using hybrid SAR images, generated from synthesized model signatures and real SAR background signatures.
CVPR	Recognizing 3D Objects by Generating Random Actions.	Stéphane Herbin	1996	This paper presents a formal model of an active recognition system that can be programmed by learning. At each time step the system decides between producing an action to generate new data and stopping to issue the name of the object observed. The actions can be directed either towards the external environment or towards the internal perceptual system of the agent. The decision strategy is based on a quantitative evaluation of the system learning experience. The problem studied is the recognition of chess pieces using a moving camera and a multiscale feature detector. The recognition is difficult because the objects are complex -- neither polyhedral nor smooth -- and rather similar between classes, especially in certain view configurations. The system uses the information obtained by observing internal state transitions when the camera is moved or when the feature detector scale is changed. A simulation of the agent and the environment is used for experimental measures of the model performances.
CVPR	SiteCity: A Semi-Automated Site Modelling System.	Yuan C. Hsieh	1996	This paper presents SiteCity, a semi-automated building extraction system integrating photogrammetry, geometric constraints and image understanding algorithms. Existing automated building extraction systems produce mixed results and it is clear that human intervention is required to correct mistakes from fully automated systems. SiteCity gives human operators the ability to construct and manipulate three dimensional building objects using multiple images. Image understanding algorithms are integrated into SiteCity to assist users. The automated processes in SiteCity use user-delineated roof boundaries as cues, and attempt to locate the floor of a building and match the building object in other images. In addition, photogrammetric cues are used to assist automated processes. These automated processes are described and their performance is evaluated, illustrating that automated processes in SiteCity produce comparable performance to that of human subjects.
CVPR	Recognizing Three-Dimensional Objects by Comparing Two-Dimensional Images.	Daniel P. Huttenlocher,Liana M. Lorigo	1996	"In this paper we address the problem of recognizing an object from a novel viewpoint, given a single ``model'' view of that object. As is common in model-based recognition, objects and images are represented as sets of feature points. We present an efficient algorithm for determining whether two sets of image points (in the plane) could be projections of a common object (a three-dimensional point set). The method relies on the fact that two sets of points in the plane are orthographic projections of the same three-dimensional point set exactly when they have a common projection onto a line. This is a form of the well-known epipolar constraint used in stereopsis. Our algorithm can be used to recognize an object by comparing a stored two-dimensional view of the object against an unknown view, without requiring the correspondence between points in the views to be known a priori. We provide some examples illustrating the approach."
CVPR	Skin and Bones: Multi-layer, Locally Affine, Optical Flow and Regularization with Transparency.	Shanon X. Ju,Michael J. Black,Allan D. Jepson	1996	"This paper describes a new method for estimating optical flow that strikes a balance between the flexibility of local dense computations and the robustness and accuracy of global parameterized flow models. An affine model of image motion is used within local image patches while a spatial smoothness constraint on the affine flow parameters of neighboring patches enforces continuity of the motion. We refer to this as a ``Skin and Bones'' model in which the affine patches can be thought of as rigid ``bones'' connected by a flexible ``skin''. Since local image patches may contain multiple motions we use a layered representation for the affine bones. To regularize this layered motion representation we develop a new framework for ""regularization with transparency""."
CVPR	Gesture recognition using the Perseus architecture.	Roger E. Kahn,Michael J. Swain,Peter N. Prokopowicz,R. James Firby	1996	Interpersonal communication involves more than simply spoken information. Gestures are commonly used to more efficiently and precisely communicate. An important gesture because of its descriptive power and frequency of use is pointing. To produce a more natural and powerful human-robot interface, we have developed a purposive visual architecture called Perseus and have used it to locate objects a person is pointing to. With Perseus, in real-time, we are able to determine when a person enters the scene, track the relevant parts of the person including the hands and head, and recognize when she is pointing. Once the person points, the object pointed to is located. The Perseus architecture allows knowledge about the task and context to be used at all levels of visual analysis for improved performance. This knowledge is explicitly represented in the Perseus system to facilitate the extension of Perseus to other tasks and environments. In this paper we describe Perseus and how it is used to solve this task. Previous work on Perseus is extended by describing a more sophisticated representation of the person and superior segmenting and tracking methods. Experiments showing the success of the Perseus system with numerous naive users in varied environments is presented.
CVPR	Model-based estimation of 3D human motion with occlusion based on active multi-viewpoint selection.	Ioannis A. Kakadiaris,Dimitris N. Metaxas	1996	"Dept. of Comput. & Inf. Sci., Pennsylvania Univ., Philadelphia, PA, USA Abstract: We present a new method for the 3D model-based tracking of human body parts. To mitigate the difficulties arising due to occlusion among body parts, we employ multiple calibrated cameras in a mutually orthogonal configuration. In addition, we develop criteria for a time varying active selection of a set of cameras to track the motion of a particular human part. In particular, at every frame, each camera tracks a number of parts depending on the visibility of these parts and the observability of their predicted motion from the specific camera. To relate points on the occluding contours of the parts to points on their models we apply concepts from projective geometry. Then, within the physics-based framework we compute the generalized forces applied from the parts' occluding contours to model points of the body parts. These forces update the translational and rotational degrees of freedom of the model, such as to minimize the discrepancy between the sensory data and the estimated model state. We present initial tracking results from a series of experiments involving the recovery of complex 3D motions in the presence of significant occlusion."
CVPR	A Stereo Machine for Video-Rate Dense Depth Mapping and Its New Applications.	Takeo Kanade,Atsushi Yoshida,Kazuo Oda,Hiroshi Kano,Masaya Tanaka	1996	We have developed a video-rate stereo machine that has the capability of generating a dense depth map at the video rate. The performance bench marks of the CMU video-rate stereo machine are: 1) multi image input of up to 6 cameras; 2) throughput of 30 million point-disparity measurements per second; 3) frame rate of 30 frame/sec; 4) a dense depth map of up to 256 x 240 pixels; 5) disparity search range of up to 60 pixels; 6) high precision of depth output up to 8 bi ts (with interpolation). The capability of passively producing such a dense depth map (3D representation) of a scene at the video rate can open up a new class of applications of 3D vision:merging real and virtual worlds in real time.
CVPR	3-D Scene Data Recovery using Omnidirectional Multibaseline Stereo.	Sing Bing Kang,Richard Szeliski	1996	A traditional approach to extracting geometric information from a large scene is to compute multiple 3-D depth maps from stereo pairs or direct range finders, and then to merge the 3-D data. However, the resulting merged depth maps may be subject to merging errors if the relative poses between depth maps are not known exactly. In addition, the 3-D data may also have to be resampled before merging, which adds additional complexity and potential sources of errors.This paper provides a means of directly extracting 3-D data covering a very wide field of view, thus by-passing the need for numerous depth map merging. In our work, cylindrical images are first composited from sequences of images taken while the camera is rotated 360&deg; about a vertical axis. By taking such image panoramas at different camera locations, we can recover 3-D data of the scene using a set of simple techniques: feature tracking, an 8-point structure from motion algorithm, and multibaseline stereo. We also investigate the effect of median filtering on the recovered 3-D point distributions, and show the results of our approach applied to both synthetic and real scenes.
CVPR	Recognition of Handwritten Phrases as Applied to Street Name Images.	Gyeonghwan Kim,Venu Govindaraju	1996	A method for recognition of street name phrases collected from mail pieces is presented in this paper. Some of the challenges posed by the problem are: (i) patron errors, (ii) non-standardized way of abbreviating names, and (iii) variable number of words in a street name image. A neural network has been designed to segment words in a phrase, a street name in this case, using distances between components and style of writing. The network learns the type of spacing (including size) that one should expect between different pairs of characters in handwritten text. Experiments show perfect word segmentation performance at about 85% of cases. Unlike conventional methods, where lexicon entries are expanded to take care of all variations of prefixes and suffixes, substring matching is attempted only between the main body of a lexicon entry and the word segments of an image. Efforts to reduce computational complexity are successfully made by the sharing of character segmentation results between the segmentation and recognition phases. 83% phrase recognition accuracy is achieved on a test set.
CVPR	Panoramic Image Acquisition.	Arun Krishnan,Narendra Ahuja	1996	This paper is concerned with acquiring panoramic focused images using a small field of view video camera. When scene points are distributed over a range of distances from the sensor, obtaining a focused composite image involves focus computations and mechanically changing some sensor parameters (translation of sensor plane, panning of camera etc.) which can be time intensive. In this paper we present methods to optimize the image acquisition strategy in order to reduce redundancy. We show that panning a camera about a point f (focal length) in front of the camera eliminates redundancy. The Non-frontal imaging camera (NICAM) with tilted sensor plane has been previously introduced as a sensor that can acquire focused panoramic images. In this paper we also describe strategies for optimal selection of panning angle increments and sensor plane tilt for NICAM. Experimental results are presented for panoramic image acquisition using a regular camera as well as using NICAM.
CVPR	Eigenfeatures for planar pose measurement of partially occluded objects.	John Krumm	1996	"Planar pose measurement from images is an important problem for automated assembly and inspection. In addition to accuracy and robustness, ease of use is very important for real world applications. Recently, Murase and Nayar have presented the ""parametric eigenspace "" for object recognition and pose measurement based on training images. Although their system is easy to use, it has potential problems with background clutter and partial occlusions. We present an algorithm that is robust in these terms. It uses several small features on the object rather than a monolithic template. These ""eigenfeatures"" are matched using a median statistic, giving the system robustness in the face of background clutter and partial occlusions. We demonstrate our algorithm's pose measurement accuracy with a controlled test, and we demonstrate its detection robustness on cluttered images with the objects of interest partially occluded."
CVPR	Recovery of Global Nonrigid Motion - Based Approach Without Point Correspondences.	Senthil Kumar,Dmitry B. Goldgof	1996	Recovery of Global Nonrigid Motion - Based Approach Without Point Correspondences.
CVPR	Novel active-vision-based visual-threat-cue for autonomous navigation tasks.	Sridhar R. Kundur,Daniel Raviv	1996	This paper presents a new visual motion cue, we call the Visual Threat Cue (VTC) that provides some measure for a relative change in range as well as clearance between a 3D surface and a fixing observer in motion. The VTC corresponds to visual fields surrounding a moving observer. The fields are time-based imaginary 3-D surfaces that move with the observer. They are analogous to equi-potential fields of an electric dipole. A practical method to extract the VTC is presented. The approach is independent of the 3D surface texture and needs no optical flow information, 3D reconstruction, segmentation, feature tracking or pre-processing. This algorithm to extract the VTC was applied to several indoor as well as outdoor real images of textures, where we observed a similar behavior for most of the textures employed.
CVPR	Recovering the viewing parameters of random, translated and noisy projections of asymmetric objects.	Peter D. Lauren,N. Nandhakumar	1996	A method is described for the determination of the viewing parameters of randomly acquired projections of asymmetric objects. It extends upon the common lines algorithm by determining the relative orientation of projections from the location of lines of intersection among the Fourier transforms of the projections in three-dimensional Fourier space. A new technique for finding the lines of intersection in the presence of translational displacement, and for subsequently finding the translational displacement, is presented. A new technique for dealing with noise is also presented. The complete algorithm is described and its efficacy is demonstrated using real data. This technique may be applied to the three-dimensional reconstruction of viruses, molecules, and cells from in vivo images. It also has many other applications including the reconstruction of underwater scenes, radioastronomy, geoseismic analysis, and portable radiography for medical diagnosis and industrial inspection.
CVPR	Convolutional Neural Networks for Face Recognition.	Steve Lawrence,C. Lee Giles,Ah Chung Tsoi	1996	Convolutional Neural Networks for Face Recognition.
CVPR	New, faster, more controlled fitting of implicit polynomial 2D curves and 3D surfaces to data .	Zhibin Lei,David B. Cooper	1996	Denote a point in the plane by z=(z,y) and a polynomial of nth degree in z by f(z) /spl Sigma//sub i,j//spl ges/o/sub 1/i+j/spl les/n(a/sub ij/x/sup i/y/sup j/). Denote by Z(f) the set of points for which f(z)=0. Z(f) is the 2D curve represented by f(z). In this paper, we present a new approach to fitting 2D curves to data in the plane (or 3D surfaces to range data) which has significant advantages over presently known methods. It requires considerably less computation and the resulting curve can be forced to lie close to the data set at prescribed points provided that there is an nth degree polynomial that can reasonably approximate the data. Linear programming is used to do the fitting. The approach can incorporate a variety of distance measures and global geometric constraints.
CVPR	Dealing with occlusions in the eigenspace approach.	Ales Leonardis,Horst Bischof	1996	The basic limitations of the current appearance-based matching methods using eigenimages are non-robust estimation of coefficients and inability to cope with problems related to occlusions and segmentation. In this paper we present a new approach which successfully solves these problems. The major novelty of our approach lies in the way how the coefficients of the eigenimages are determined. Instead of computing the coefficients by a projection of the data onto the eigenimages, we extract them by a hypothesize-and-test paradigm using subsets of image points. Competing hypotheses are then subject to a selection procedure based on the Minimum Description Length principle. The approach enables us not only to reject outliers and to deal with occlusions but also to simultaneously use multiple classes of eigenimages.
CVPR	Bayesian image restoration and segmentation by constrained optimizatio.	Stan Z. Li,Kap Luk Chan,Han Wang	1996	Abstract: A constrained optimization method, called the Lagrange-Hopfield (LH) method, is presented for solving Markov random field (MRF) based Bayesian image estimation problems for restoration and segmentation. The method combines the augmented Lagrangian multiplier technique with the Hopfield network to solve a constrained optimization problem into which the original Bayesian estimation problem is reformulated. The LH method effectively overcomes instabilities that are inherent in the penalty method (e.g. Hopfield network) or the Lagrange multiplier method in constrained optimization. An additional advantage of the LH method is its suitability for neural-like analog implementation. Experimental results are presented which show that LH yields good quality solutions at reasonable computational costs.
CVPR	Geometric and Photometric Constraints for Surface Recovery.	Jiping Lu,James J. Little	1996	In this paper we present a novel approach to surface recovery from an image sequence of a rotating object. In this approach, the object is illuminated under a collinear light source (where the light source lies on or near the optical axis) and rotated on a controlled turntable. A wire-frame of 3D curves on the object surface is extracted by using shading and occluding contours in the image sequence. Then the whole object surface is recovered by interpolating the surface between curves on the wire-frame. The interpolation can be done by using geometric or photometric methods. The photometric method uses shading information and is more powerful than geometric methods. The experimental results on real image sequence of matte and specular surfaces show that the technique is feasible and promising.
CVPR	Texture Features and Learning Similarity.	Wei-Ying Ma,B. S. Manjunath	1996	This paper addresses two important issues related to texture pattern retrieval: feature extraction and similarity search. A Gabor feature representation for textured images is proposed, and its performance in pattern retrieval is evaluated on a large texture image database. These features compare favorably with other existing texture representations. A simple hybrid neural network algorithm is used to learn the similarity by simple clustering in the texture feature space. With learning similarity, the performance of similar pattern retrieval improves significantly. An important aspect of this work is its application to real image data. Texture feature extraction with similarity learning is used to search through large aerial photographs. Feature clustering enables efficient search of the database as our experimental results indicate.
CVPR	Edge Detection and Ridge Detection with Automatic Scale Selection.	Tony Lindeberg	1996	When computing descriptors of image data, the type of information that can be extracted may be strongly dependent on the scales at which the image operators are applied. This article presents a systematic methodology for addressing this problem. A mechanism is presented for automatic selection of scale levels when detecting one-dimensional image features, such as edges and ridges.A novel concept of a scale-space edge is introduced, defined as a connected set of points in scale-space at which: (i) the gradient magnitude assumes a local maximum in the gradient direction, and (ii) a normalized measure of the strength of the edge response is locally maximal over scales. An important consequence of this definition is that it allows the scale levels to vary along the edge. Two specific measures of edge strength are analyzed in detail, the gradient magnitude and a differential expression derived from the third-order derivative in the gradient direction. For a certain way of normalizing these differential descriptors, by expressing them in terms of so-called &gamma;-normalized derivatives, an immediate consequence of this definition is that the edge detector will adapt its scale levels to the local image structure. Specifically, sharp edges will be detected at fine scales so as to reduce the shape distortions due to scale-space smoothing, whereas sufficiently coarse scales will be selected at diffuse edges, such that an edge model is a valid abstraction of the intensity profile across the edge.Since the scale-space edge is defined from the intersection of two zero-crossing surfaces in scale-space, the edges will by definition form closed curves. This simplifies selection of salient edges, and a novel significance measure is proposed, by integrating the edge strength along the edge. Moreover, the scale information associated with each edge provides useful clues to the physical nature of the edge.With just slight modifications, similar ideas can be used for formulating ridge detectors with automatic selection, having the characteristic property that the selected scales on a scale-space ridge instead reflect the width of the ridge.It is shown how the methodology can be implemented in terms of straightforward visual front-end operations, and the validity of the approach is supported by theoretical analysis as well as experiments on real-world and synthetic data.
CVPR	Viewpoint Variation in the Noise Sensitivity of Pose Estimation.	Claus B. Madsen	1996	The paper presents an analysis of the stability of pose estimation. The investigated pose estimation technique is based on orientations of three edge segments and provides the rotation part of object pose.The specific emphasis of the analysis is on determining how the stability varies with view point relative to an object. The stability investigation propagates the uncertainty in edge segment orientations to the resulting effect on the pose parameters. It is shown that there is a very strong variation in noise sensitivity over the range of viewpoints and that exactly what viewpoints offer highest robustness towards noise can be determined in advance. Experiments on real images verify the theoretical results and show that, dependent on viewpoint, pose parameter variance varies from 0.05 to 20 (degrees squared).
CVPR	Word Spotting: A New Approach to Indexing Handwriting.	R. Manmatha,Chengfeng Han,Edward M. Riseman	1996	"There are many historical manuscripts written in a single hand which it would be useful to index. Examples include the early Presidential papers at the Library of Congress and the collected works of W. B. DuBois at the library of the University of Massachusetts. The standard technique for indexing documents is to scan them in, convert them to machine readable form (ASCII) using Optical Character Recognition (OCR) and then index them using a text retrieval engine. However, OCR does not work well on handwriting. Here an alternative scheme is proposed for indexing such texts. Each page of the document is segmented into words. The images of the words are then matched against each other to create equivalence classes (each equivalence classes contains multiple instances of the same word). The user then provides ASCII equivalents for say the top 2000 equivalence classes. The current paper deals with the matching aspects of this process. Due to variations in even a single person''s handwriting, it is expected that the matching will be the most difficult step in the whole process. Two different techniques for matching words are discussed. The first method, based on Euclidean distance mapping, matches words assuming that the transformation between the words may be modelled by a translation (shift). The second method, based on an algorithm developed by Scott and Longuet Higgins, matches words assuming that the transformation between the words may be modelled by an affine transform. Experiments are shown demonstrating the feasibility of the approach for indexing handwriting."
CVPR	Controlled camera motions for scene reconstruction and exploration.	Éric Marchand,François Chaumette	1996	This paper deals with the 3D structure estimation and exploration of a scene using active vision. Our method is based on the structure from controlled motion approach which consists in constraining the camera motion in order to obtain a precise and robust estimation of the 3D structure of a geometrical primitive. Since this approach involves to gaze on the considered primitive, we present a method for connecting up many estimations in order to recover the complete spatial structure of scenes composed of cylinders and segments. We have developed perceptual strategies able to perform a succession of robust estimations without any assumption on the number and on the localization of the different objects. Furthermore, the proposed strategy ensures the completeness of the reconstruction. An exploration process centered on current visual features and on the structure of the previously studied primitives is presented. This leads to a gaze planning strategy that mainly uses a representation of known and unknown areas as a basis for selecting viewpoints. Finally, experiments carried out on a robotic cell have proved the validity of our approach.
CVPR	Physics-Based Segmentation: Moving Beyond Color.	Bruce A. Maxwell,Steven A. Shafer	1996	We previously presented a framework for segmentation of complex scenes using multiple physical hypotheses for simple image regions. A consequence of that framework was a proposal for a new approach to the segmentation of complex scenes into regions corresponding to coherent surfaces rather than merely regions of similar color. Herein we present an implementation of this new approach and show example segmentations for scenes containing multi-colored piece-wise uniform objects. Using our approach we are able to intelligently segment scenes with objects of greater complexity than previous physics-based segmentation algorithms. The results show that by using general physical models we obtain segmentations that correspond more closely to coherent surfaces in the scene than segmentations found using only color.
CVPR	Isotropic Gradient Estimatio.	Jason Merron,Michael Brady	1996	The vast majority of corner and edge detectors measure image intensity gradients in order to estimate the positions and strengths of features. However, many of the most popular intensity gradient estimators are inherently and significantly anisotropic. In spite of this, few algorithms take the anisotropy into account, and so the set of features uncovered is typically sensitive to rotations of the image, compromising recognition, matching (e.g. stereo), and tracking. We introduce an effective technique for removing unwanted anisotropies from analytical gradient estimates, by measuring local intensity gradients in four directions rather than the more traditional two. In experiments using real image data, our algorithm reduces the gradient anisotropy associated with conventional analytical gradient estimates by up to 85%, yeilding more consistent feature topologies.
CVPR	Dense Nonrigid Motion Tracking from a Sequence of Velocity Fields.	François G. Meyer,R. Todd Constable,Albert J. Sinusas,James S. Duncan	1996	We have addressed the problem of tracking the nonrigid motion of the heart using a sequence of velocity fields and a sequence of contours. The information from both the contours and the dense velocity fields is integrated into a deforming mesh that is placed over the myocardium at one time frame and then tracked over the entire cardiac cycle. The deformation is guided by a smoothing filter that provides a compromise between (i) believing the dense field velocity and the contour data when it is crisp and coherent in a local spatial and temporal sense and (ii) employing a temporally smooth cyclic model of cardiac motion when contour and velocity data are not trustworthy. The method has been carefully evaluated with simulated data and phantom data. Experiments with in vivo data have also been conducted.
CVPR	MUSE: Robust Surface Fitting using Unbiased Scale Estimates.	James V. Miller,Charles V. Stewart	1996	"Despite many successful applications of robust statistics, they have yet to be completely adapted to many computer vision problems. Range reconstruction, particularly in unstructured environments, requires a robust estimator that not only tolerates a large outlier percentage but also tolerates several discontinuities, extracting multiple surfaces in an image region. Observing that random outliers and/or points from across discontinuities increase a hypothesized fit's scale estimate (standard deviation of the noise), our new operator, called MUSE (Minimum Unbiased Scale Estimator), evaluates a hypothesized fit over potential inlier sets via an objective function of unbiased scale estimates. MUSE extracts the single best fit from the data by minimizing its objective function over a set of hypothesized fits and can sequentially extract multiple surfaces from an image region. We show MUSE to be effective on synthetic data modelling small scale discontinuities and in preliminary experiments on complicated range data."
CVPR	"Interactive Learning with a ""Society of Models"" ."	Thomas P. Minka,Rosalind W. Picard	1996	"Digital library access is driven by features, but the relevance of a feature for a query is not always obvious. This paper describes an approach for integrating a large number of context-dependent features into a semi-automated tool. Instead of requiring universal similarity measures or manual selection of relevant features, the approach provides a learning algorithm for selecting and combining groupings of the data, where groupings can be induced by highly specialized features. The selection process is guided by positive and negative examples from the user. The inherent combinatorics of using multiple features is reduced by a multistage grouping generation, weighting, and collection process. The stages closest to the user are trained fastest and slowly propagate their adaptations back to earlier stages. The weighting stage adapts the collection stage's search space across uses, so that, in later interactions, good groupings are found given few examples from the user."
CVPR	Bayesian face recognition using deformable intensity surfaces.	Baback Moghaddam,Chahab Nastar,Alex Pentland	1996	We describe a novel technique for face recognition based on deformable intensity surfaces which incorporates both the shape and texture components of the 2D image. The intensity surface of the facial image is modeled as a deformable 3D mesh in (z, y, I(x, y)) space. Using an efficient technique for matching two surfaces (in terms of the analytic modes of vibration), we obtain a dense correspondence field (or 3D warp) between two images. The probability distributions of two classes of warps are then estimated from training data: interpersonal and extrapersonal variations. These densities are then used in a Bayesian framework for image matching and recognition. Experimental results with facial data from the US Army FERET database demonstrate an increased recognition rate over the previous best methods.
CVPR	Closest Point Search in High Dimensions.	Sameer A. Nene,Shree K. Nayar	1996	The problem of finding the closest point in high-dimensional spaces is common in computational vision. Unfortunately, the complexity of most existing search algorithms, such as k-d tree and R-tree, grows exponentially with dimension, making them impractical for dimensionality above 15. In nearly all applications, the closest point is of interest only if it lies within a user specified distance \epsilon. We present a simple and practical algorithm to efficiently search for the nearest neighbor within Euclidean distance \epsilon. Our algorithm uses a projection search technique along with a novel data structure to dramatically improve performance in high dimensions. A complexity analysis is presented which can help determine \epsilon in structured problems. Benchmarks clearly show the superiority of the proposed algorithm for high dimensional search problems frequently encountered in machine vision, such as real-time object recognition.
CVPR	Parametric Feature Detection.	Shree K. Nayar,Simon Baker,Hiroshi Murase	1996	We propose an algorithm to automatically construct feature detectors for arbitrary parametric features. To obtain a high level of robustness we advocate the use of realistic multi-parameter feature models and incorporate optical and sensing effects. Each feature is represented as a densely sampled parametric manifold in a low dimensional subspace of a Hilbert space. During detection, the brightness distribution around each image pixel is projected into the subspace. If the projection lies sufficiently close to the feature manifold, the feature is detected and the location of the closest manifold point yields the feature parameters. The concepts of parameter reduction by normalization, dimension reduction, pattern rejection, and heuristic search are all employed to achieve the required efficiency. By applying the algorithm to appropriate parametric feature models, detectors have been constructed for five features, namely, step edge, roof edge, line, corner, and circular disc. Detailed experiments are reported on the robustness of detection and the accuracy of parameter estimation.
CVPR	3-D Object Pose Estimation by Shading and Edge Data Fusion ----Simulating virtual manipulation on mental images.	Yoshihiko Nomura,Dili Zhang,Yuko Sakaida,Seizo Fujii	1996	Human beings seem to recognize objects based on a kind of model-matching, i.e., a virtual manipulation on mental images. This paper presents a 3-D object pose estimation method simulating the human recognition scheme. Computer synthesizes not only an edge image but also a shading image from an object model. Then, it matches the two kinds of synthesized images with the inputted images individually by using a non-linear least-squares method, and estimates the pose parameter values. Finally, it chooses the better of the individually estimated poses. Thus, the fusion of the shading and the edge information is achieved. Since the two pieces of information complement each other, this method has the advantage of much higher robustness and accuracy of pose estimation than ordinary model-matching techniques which rely only on geometrical features such as vertices or edges.
CVPR	Recovering Affine Motion and Defocus Blur Simultaneously.	Zarina Myles,Niels da Vitoria Lobo	1996	Motion in depth and/or zooming cause defocus blur. We show how the defocus blur in an image can be recovered simultaneously with affine motion. We introduce the theory, develop a solution method and demonstrate the validity of the theory and the solution by conducting experiments with real scenery.
CVPR	Global Models with Parametric Offsets as Applied to Cardiac Motion Recovery.	"Thomas O'Donnell,Terrance E. Boult,Alok Gupta"	1996	"We introduce a new solid shape model formulation that includes built-in offsets from a base global component (e.g. an ellipsoid) which are functions of the global component's parameters. The offsets provide two features. First, they help to form an expected model shape which facilitates appropriate model data correspondences. Second, they scale with the base global model to maintain the expected shape even in the presence of large global deformations. We apply this model formulation to the recovery of 3-D cardiac motion from a volunteer dataset of tagged-MR images. The model instance is a variation of the Hybrid Volumetric Ventriculoid (HVV), a deformable thick-walled ellipsoid model resembling the left ventricle (LV) of the heart. A unique aspect of our implementation is the employment of constant volume constraints when recovering the cardiac motion. In addition, we present a novel geodesic-like prismoidal tessellation of the model which provides for more stable fits."
CVPR	Occlusion Detectable Stereo -- Occlusion Patterns in Camera Matrix.	Yuichi Nakamura,Tomohiko Matsuura,Kiyohide Satoh,Yuichi Ohta	1996	In stereo algorithms with more than two cameras, the improvement of accuracy is often reported since they are robust against noise. However, another important aspect of the polynocular stereo, that is the ability of occlusion detection, has been paid less attention. We intensively analyzed the occlusion in the camera matrix stereo (SEA) and developed a simple but effective method to detect the presence of occlusion and to eliminate its effect in the correspondence search. By considering several statistics on the occlusion and the accuracy in the SEA, we derived a few base masks which represent occlusion patterns and are effective for the detection of occlusion. Several experiments using typical indoor scenes showed quite good performance to obtain dense and accurate depth maps even at the occluding boundaries of objects.
CVPR	Structure from Linear or Planar Motions.	John Oliensis	1996	We recently demonstrated a new approach to multiframe structure from motion from point features which, in the appropriate domain, provably reconstructs structure and motion correctly. The algorithm works for general motion and large perspective effects. In this paper, we describe how to adapt our approach to translational motion lying along a line or in a plane, with arbitrary rotations. An analysis of the bas--relief effect for multiple--motion sequences is also presented.
CVPR	Connectionist networks for feature indexing and object recognition.	Clark F. Olson	1996	Feature indexing techniques are promising for object recognition because of their ability to eliminate many feature set matches from consideration without much computation. This work exploits another property of such techniques. They have inherently parallel structure and connectionist network formulations are easy to develop. Once indexing has been performed, a voting scheme such as geometric hashing [Lamdan et al. 1990] can be used to generate object hypotheses in parallel. We give a framework for the connectionist implementation for such indexing and recognition techniques. With sufficient processing elements, recognition can be performed in a small number of time steps. The number of processing elements necessary to achieve peak performance and the fan-in/fan-out required for the processing elements is determined. These techniques have been simulated on a conventional architecture with good results.
CVPR	Affine Invariant Detection: Edges, Active Contours, and Segments.	Peter J. Olver,Guillermo Sapiro,Allen Tannenbaum	1996	In this paper we undertake a systematic investigation of affine invariant object detection. Edge detection is first presented from the point of view of the affine invariant scale-space obtained by curvature based motion of the image level-sets. In this case, affine invariant edges are obtained as a weighted difference of images at different scales. We then introduce the affine gradient as the simplest possible affine invariant differential function which has the same qualitative behavior as the Euclidean gradient magnitude. These edge detectors are the basis both to extend the affine invariant scale-space to a complete affine flow for image denoising and simplification, and to define affine invariant active contours for object detection and edge integration. The active contours are obtained as a gradient flow in a conformally Euclidean space defined by the image on which the object is to be detected. That is, we show that objects can be segmented in an affine invariant manner by computing a path of minimal weighted affine distance,the weight being given by functions of the affine edge detectors. The geodesic path is computed via an algorithm which allows to simultaneously detect any number of objects independently of the initial curve topology.
CVPR	A fast and flexible statistical method for text extraction in document pages.	Pietro Parodi,Giulia Piccioli	1996	This paper describes a fast and flexible method for extracting text regions from a document page containing text, graphics, and pictures. Such regions can be given as an input to an OCR system. The user fixes two parameters, the minimum width w of the text to be detected, and the precision \epsilon needed (both expressed as a percentage of the image width), according to the implementation needs. The method works by subdividing the page into overlapping columns whose width and inter-shift depend on w and \epsilon, and by performing text lines extraction on each column separately. Successively, a statistical analysis of the text line elements found in each column is performed, and they are connected to form complete text lines. Finally, related pieces of text are merged into blocks so that a sensible reading order is provided for the OCR system. The algorithm is very fast, is able to work on low-resolution document pages and is robust against skew. The algorithm is also very flexible: no assumptions are made on the layout of the document, the shape of the text regions, and the font size and style; the main assumption is that the background is uniform and the text approximately horizontal. Despite the statistical nature of the method, a single line of text of a certain font size is generally sufficient to warrant detection. Experimental results are shown which demonstrate the effectiveness of the method on several different kinds of documents.
CVPR	A Factorization Method for Affine Structure from Line Correspondences.	Long Quan,Takeo Kanade	1996	A family of structure from motion algorithms called the factorization method has been recently developed from the orthographic projection model to the affine camera model. All these algorithms are limited to handling only point features of the image stream. We propose in this paper an algorithm for the recovery of shape and motion from line correspondences by the factorization method with the affine camera. Instead of one step factorization for points, a multi-step factorization method is developed for lines based on the decomposition of the whole shape and motion into three separate substructures. Each of these substructures can then be linearly solved by factorizing the appropriate measurement matrices. It is also established that affine shape and motion with uncalibrated affine cameras can be achieved with at least seven lines over three views, which extends the previous results of Koenderink and Van Doorn for points to lines.
CVPR	Closed-Loop Object Recognition Using Reinforcement Learning.	Jing Peng,Bir Bhanu	1996	Current computer vision systems whose basic methodology is open-loop or filter type typically use image segmentation followed by object recognition algorithms. These systems are not robust for most real-world applications. In contrast, the system presented here achieves robust performance by using reinforcement learning to induce a mapping from input images to corresponding segmentation parameters. This is accomplished by using the confidence level of model matching as a reinforcement signal for a team of learning automata to search for segmentation parameters during training. The use of the recognition algorithm as part of the evaluation function for image segmentation gives rise to significant improvement of the system performance by automatic generation of recognition strategies. The system is verified through experiments on sequences of indoor and outdoor color images with varying external conditions.
CVPR	Randomness and Geometric Features in Computer Vision.	Xavier Pennec,Nicholas Ayache	1996	"It is often necessary to handle randomness and geometry in computer vision, for instance to match and fuse together noisy geometric features such as points, lines or 3D frames, or to estimate a geometric transformation from a set of matched features. However, the proper handling of these geometric features is far more difficult than for points, and a number of paradoxes can arise. We analyse in this article three basic problems: (1) what is a uniform random distribution of features, (2) how to define a distance between features, and (3) what is the ``mean feature'' of a number of feature measurements, and we propose generic methods to solve them."
CVPR	Epipolar Geometry and Linear Subspace Methods: A New Approach to Weak Calibration.	Jean Ponce,Yakup Genc	1996	This paper addresses the problem of estimating the epipolar geometry from point correspondences between two images taken by uncalibrated perspective cameras. It is shown that Jepson&lsquo;s and Heeger&lsquo;s linear subspace technique for infinitesimal motion estimation can be generalized to the finite motion case by choosing an appropriate basis for projective space. This yields a linear method for weak calibration. The proposed algorithm has been implemented and tested on both real and synthetic images, and it is compared to other linear and non-linear approaches to weak calibration.
CVPR	Constrained Phase Congruency: Simultaneous Detection of Interest Points and of their Scales.	Daniel Reisfeld	1996	"A novel feature detector -- the Constrained Phase Congruency Transform (CPCT) is introduced. It simultaneously detects interest points as well as their scale in various orientations. The CPCT is especially important in registration applications: the local transformation between interest points can be determined based on their orientational scales. The CPCT detects the features in Mach bands and in sinusoidal waves. This cannot be done simply by looking for local maxima in intensity gradient nor by looking for local energy maxima. I conjecture that constraining the general phase congruency is sufficient for feature detection. The correct detection of features' location and of their scale is demonstrated. The robustness of the CPCT is achieved by constraining the local phase. Only four easy-to-detect phases are used in the computations. They correspond to symmetry and anti-symmetry in their neighborhood. The scale at any location and orientation is determined by the scale of the channel that conforms to the constraints and maximizes energy."
CVPR	Modeling Clutter and Context for Target Detection in Infrared Images.	Songnian Rong,Bir Bhanu	1996	In order to reduce false alarms and to improve the target detection performance of an automatic target detection and recognition system operating in a cluttered environment, it is important to develop the models not only for man-made targets but also of natural background clutters. Because of the high complexity of natural clutters, this clutter model can only be reliably built through learning from real examples. If available, contextual information that characterizes each training example can be used to further improve the learned clutter model. In this paper, we present such a clutter model aided target detection system. Emphases are placed on two topics: (1) learning the background clutter model from sensory data through a self-organizing process, (2) reinforcing the learned clutter model using contextual information.
CVPR	Affine Structure and Photometry.	Ruth Rosenholtz,Jan J. Koenderink	1996	Motion of an observer relative to objects in a scene provides information about the structure of the scene. Changing patterns of shading due to motion relative to the light source provide information about surface structure, albedos, and light sources. One can stratify this photometric information into affine, unitary, and metric structure, much like the stratification of structure from motion. For Lambertian surfaces, if either motion or photometry give us more than affine structure, the two cues can be combined to yield full metric information. Edge constraints plus unitary photometry also give us full metric photometry. Affine structure alone contains much of the quantitative structure information, allowing us to judge such things as the ordinal relationships between the albedos.
CVPR	Optimal servoing for active foveated vision.	Héctor Rotstein,Ehud Rivlin	1996	"Foveated vision and two-mode tracking, as inspired by the human oculomotor system, are often used in active vision system. The purpose of this paper is to provide answers to the following basic questions which arise from implementations. First, is it beneficial to have foveated vision and what is the optimal size of the foveal window? Second, is there a need for two control mechanisms (smooth pursuit and saccade) for improved performance and how can one efficiently switch between them? In order to do so, a setup is proposed in which these strategies can be evaluated in a systematic manner. It is shown that the fovea appears as a compromise between the tightness of the tracking specifications and computational constraints. Introducing a model for the later and postulating some a priori knowledge of the target behavior, it is possible to compute the size of the fovea in an optimal way. As a by-product, ""smooth-pursuit"" can be defined in a natural way, and the use of a two-mode tracking scheme is justified. The second mode, i.e. ""saccadic control"", aims at re-centering the target on the fovea so that the smooth pursuit controller can continue to operate. It is shown that a control strategy can indeed be defined so that this objective can be met under appropriate operating conditions."
CVPR	Robust Recovery of Camera Rotation from Three Frames.	Benny Rousso,Shai Avidan,Amnon Shashua,Shmuel Peleg	1996	Computing camera rotation from image sequences can be used for image stabilization, and when the camera rotation is known the computation of translation and scene structure are much simplified as well. A robust approach for recovering camera rotation is presented, which does not assume any specific scene structure (e.g. no planar surface is required), and which avoids prior computation of the epipole. Given two images taken from two different viewing positions, the rotation matrix between the images can be computed from any three homgraphy matrices. The homographies are computed using the trilinear tensor which describes the relations between the projections of a 3D point into three images. The entire computation is linear for small angles, and is therefore fast and stable. Iterating the linear computation can then be used to recover larger rotations as well.
CVPR	Neural Network-Based Face Detection.	Henry A. Rowley,Shumeet Baluja,Takeo Kanade	1996	We present a neural network-based upright frontal face detection system. A retinally connected neural network examines small windows of an image and decides whether each window contains a face. The system arbitrates between multiple networks to improve performance over a single network. We present a straightforward procedure for aligning positive face examples for training. To collect negative examples, we use a bootstrap algorithm, which adds false detections into the training set as training progresses. This eliminates the difficult task of manually selecting nonface training examples, which must be chosen to span the entire space of nonface images. Simple heuristics, such as using the fact that faces rarely overlap in images, can further improve the accuracy. Comparisons with several other state-of-the-art face detection systems are presented, showing that our system has comparable performance in terms of detection and false-positive rates.
CVPR	Similarity Queries in Image Database.	Simone Santini,Ramesh Jain	1996	Query-by-content image database will be based on similarity, rater than on matching, where similarity is a measure that is defined and meaningful for every pair of images in the image space. Since it is the human user that, in the end, has to be satisfied with the results of the query, it is natural to base the similarity measure that we will use on the characteristics of human similarity assessment. In the first part of this paper, we review some of these characteristics and define a similarity measure based on them. Another problem that similarity-based databases will have to face is how to combine different queries into a single complex query. We present a solution based on three operators that are the analogous of the and, or, and not operators one uses in traditional databases. These operators are powerful enough to express queries of unlimited complexity, yet have a very intuitive behavior, making easy for the user to specify a query tailored to a particular need.
CVPR	Vector-Valued Active Contours.	Guillermo Sapiro	1996	A framework for object segmentation in vector-valued images is presented in this paper. The first scheme proposed is based on geometric active contours moving towards the objects to be detected in the vector-valued image. Objects boundaries are obtained as geodesics or minimal weighted distance curves in a Riemannian space. The metric in this space is given by a definition of edges in vector-valued images. The curve flow corresponding to the proposed active contours holds formal existence, uniqueness, stability, and correctness results. The techniques is applicable for example to color and texture images. The scheme automatically handles changes in the deforming curve topology. We conclude the paper presenting an extension of the color active contours which leads to a possible image flow for vector-valued image segmentation. The algorithm is based on moving each one of the image level-sets according to the proposed color active contours. This extension also shows the relation of the color geodesic active contours with a number of partial-differential-equations based image processing algorithms as anisotropic diffusion and shock filters.
CVPR	Complexity analysis of RBF networks for Pattern Recognition.	Lucia Sardo,Josef Kittler	1996	The problem of non-parametric probability density function (PDF) estimation using Radial Basis Function (RBF) Neural Networks is addressed here. We investigate two criteria, based on a modified Kullback-Leibler distance, that lead to an appropriate choice of the network architecture complexity. In the first criterion the modification consists in the addition of a term that penalizes complex architectures (MPL criterion). The second strategy involves the regularization of the network through the imposition of lower bounds on the standard deviation derived from conditions of existence of rejection tests (LBSD criterion). Experimental results indicate that the MPL criterion outperforms the LBSD method.
CVPR	Quantitative Measures of Change based on Feature Organization: Eigenvalues and Eigenvectors.	Sudeep Sarkar,Kim L. Boyer	1996	One important task of site monitoring is change detection from aerial images. Change, in general, can be of various types. In this paper we address the problem of detecting change due to construction activity. Specifically, we would like to know about new construction at a previously undeveloped site and possibly monitor its progress. Model based approaches are not suited for this kind of change as it usually happens in unmodelled areas. Since it is difficult to infer construction activity by predicting and verifying specific local features, we rely on more global statistical indicators. The thesis of this paper is that the change induced by human activity can be inferred from changes in the organization among the visual features. Not only will the attributes of the individual image features change but also the relationships among these features will evolve. With the progress of construction we expect to see increased structure among the image features. We exploit this emerging structure, or organization, to infer change. In this paper, we propose four measures to quantify the global statistical properties of the individual features and the relationships among them. We base these measures on the theory of graph spectra. We provide extensive analysis of the robustness of these measures under various imaging conditions. We also demonstrate the ability of these organization based measures to discriminate between no development, onset of construction, and full development.
CVPR	Stereo Vision for View Synthesis.	Daniel Scharstein	1996	We propose a new method for view synthesis from real images using stereo vision. The method does not explicitly model scene geometry, and enables fast and exact generation of synthetic views. We also re-evaluate the requirements on stereo algorithms for the application of view synthesis and discuss ways of dealing with partially occluded regions of unknown depth and with completely occluded regions of unknown texture. Our experiments demonstrate that it is possible to efficiently synthesize realistic new views even from inaccurate and incomplete depth information.
CVPR	Stereo Matching with Non-Linear Diffusion.	Daniel Scharstein,Richard Szeliski	1996	One of the central problems in stereo matching (and other image registration tasks) is the selection of optimal window sizes for comparing image regions. This paper addresses this problem with some novel algorithms based on iteratively diffusing support at different disparity hypotheses, and locally controlling the amount of diffusion based on the current quality of the disparity estimate. It also develops a novel Bayesian estimation technique which significantly outperforms techniques based on area-based matching (SSD) and regular diffusion. We provide experimental results on both synthetic and real stereo image pairs.
CVPR	Combining greyvalue invariants with local constraints for object recognition.	Cordelia Schmid,Roger Mohr	1996	This paper addresses the problem of recognizing objects in large image databases. The method is based on local characteristics which are invariant to similarity transformations in the image. These characteristics are computed at automatically detected keypoints using the greyvalue signal. The method therefore works on images such as paintings for which geometry based recognition fails.Due to the locality of the method, images can be recognized being given part of an image and in the presence of occlusions. Applying a voting algorithm and semi-local constraints makes the method robust to noise, scene clutter and small perspective deformations. Experiments show an efficient recognition for different types of images. The approach has been validated on an image database containing 1020 images, some of them being very similar by structure, texture or shape.
CVPR	Structure from Multiple 2D Affine Correspondences without Camera Calibration.	Haim Schweitzer,Radha Krishnan	1996	Image motion induced by camera or object motion can be approximated locally by an affine coordinate transformation. We extract 3D information directly from the affine parameters, without camera calibration. The derivation relies on the following assumptions: the object is rigid, locally planar, and its local 3D motion is translation. These assumptions enable complete recovery of 3D structure, whereas it is impossible to compute the direction (and magnitude) of the motion. Still, it is possible to distinguish between objects moving differently. Explicit expressions for the structure and the motion indicators are given in terms of the 6 affine parameters, computed for each image patch. Results of experiments on data with known ground truth are described.
CVPR	Coregistration of Range and Optical Images Using Coplanarity and Orientation Constraints.	Anthony N. A. Schwickerath,J. Ross Beveridge	1996	A least-squares method simultaneously solves for the model-to-sensor-suite pose and sensor-to-sensor registration. The development is for a sensor-suite containing separate range and optical sensors. To address outliers and, more generally, match finding, a statistical method (median filtering) and a search method (local search) are developed. Sensitivity to Gaussian noise and the choice of initial pose estimates is investigated on synthetic data. Both of the matching methods are demonstrated on real data.
CVPR	A Common Framework for Curve Evolution, Segmentation and Anisotropic Diffusio.	Jayant Shah	1996	"In recent years, curve evolution has developed into an important tool in Computer Vision and has been applied to a wide variety of problems such as smoothing of shapes, shape analysis and shape recovery. The underlying principle is the evolution of a simple closed curve whose points move in the direction of the normal with prescribed velocity. A fundamental limitation of the method as it stands is that it cannot deal with important image features such as triple points. The method also requires a choice of an ""edge-strength"" function, defined over the image domain, indicating the likelihood of an object boundary being present at any point in the image domain. This implies a separate preprocessing step, in essence precomputing approximate boundaries in the presence of noise. One also has to choose the initial curve. It is shown here that the different versions of curve evolution used in Computer Vision together with the preprocessing step can be integrated in the form of a new segmentation functional which overcomes these limitations and extends curve evolution models. Moreover, the numerical solutions obtained retain sharp discontinuities or ""shocks"", thus providing sharp demarcation of object boundaries."
CVPR	Subpixel Image Registration by Estimating the Polyphase Decomposition of Cross Power Spectrum.	Hassan Shekarforoush,Marc Berthod,Josiane Zerubia	1996	A method of registering images at subpixel accuracy has been proposed, which does not resort to interpolation. The method is based on the phase correlation method and is remarkably robust to correlated noise and uniform variations of luminance. We have shown that the cross power spectrum of two images, containing subpixel shifts, is a polyphase decomposition of a Dirac delta function. By estimating the sum of polyphase components one can then determine subpixel shifts along each axis.
CVPR	On 3D Shape Similarity.	Heung-Yeung Shum,Martial Hebert,Katsushi Ikeuchi	1996	On 3D Shape Similarity.
CVPR	A shock grammar for recognition.	Kaleem Siddiqi,Benjamin B. Kimia	1996	"We confront the theoretical and practical difficulties of computing a representation for two-dimensional shape, based on shocks or singularities that arise as the shape's boundary is deformed. First, we develop subpixel local detectors for finding and classifying shocks. Second, to show that shock patterns are not arbitrary but obey the rules of a grammar, and in addition satisfy specific topological and geometric constraints. Shock hypotheses that violate the grammar or are topologically or geometrically invalid are pruned to enforce global consistency. Survivors are organized into a hierarchical graph of shock groups computed in the reaction-diffusion space, where diffusion plays a role of regularization to determine the significance of each shock group. The shock groups can be functionally related to the object's parts, protrusions and bends, and the representation is suited to recognition: several examples illustrate its stability with rotations, scale changes, occlusion and movement of parts, even at very low resolutions."
CVPR	Further constraints on visual articulated motions.	David Sinclair,K. Zesar	1996	This paper derives what we term the Euclidean hinge constraint for projective reconstruction of objects displaying articulated motion. A Euclidean hinge is defined here to be an articulation axis with the proviso that any plane perpendicular to the articulation axis in link 1 has a coincident plane which is perpendicular to the articulation ads in link 2 and that these planes remain coincident under articulated motion. This constraint permits the independent projective reconstructions of two adjacent articulated links to be placed in a common frame. The constraint may be expressed mathematically by considering what we define as circular parallax. Additionally the existence of a Euclidean hinge permits the permits the projective frame to be brought nearer to a Euclidean frame for the reconstructed object. A brief reprise of the method of articulation axis estimation is given together with a more extensive series of experimental results.
CVPR	Using a spectral reflectance model for the illumination-invariant recognition of local image structure.	David Slater,Glenn Healey	1996	We represent local spatial structure in a color image using feature matrices that are computed from an image region. Feature matrices contain significantly more information about local image structure than previous representations. Although feature matrices are useful for surface recognition, this representation depends on the spectral properties of the scene illumination. Using a finite dimensional linear model for surface spectral reflectance with the same number of parameters as the number of color bands, we show that illumination changes correspond to linear transformations of the feature matrices and that surface rotations correspond to circular shifts of the matrices. From these relationships we derive an algorithm for illumination and geometry invariant recognition of local surface structure. We demonstrate the algorithm with a series of experiments on images of real objects.
CVPR	Motion from fixation.	Stefano Soatto,Pietro Perona	1996	"We study the problem of estimating rigid motion from a sequence of monocular perspective images obtained by navigating around an object while fixating a particular feature point. We cast the problem in the framework of ""epipolar geometry"", and propose a filter based upon implicit dynamical model for recursively estimating motion under the fixation constraint. This allows us to compare the quality of the estimates directly against the ones obtained assuming a general rigid motion simply by changing the geometry of the parameter space, while maintaining the same structure of the recursive estimator. We also present a closed-form static solution from two views, and a recursive estimator of the relative pose between the viewer and the scene."
CVPR	"Reducing ""structure from motion""."	Stefano Soatto,Pietro Perona	1996	"The literature on recursive estimation of structure and motion from monocular image sequences comprises a large number of different models and estimation techniques. We propose a framework that allows us to derive and compare all models by following the idea of dynamical system reduction. The ""natural"" dynamic model, derived by the rigidity constraint and the perspective projection, is first reduced by explicitly decoupling structure (depth) from motion. Then implicit decoupling techniques are explored, which consist of imposing that some function of the unknown parameters is held constant. By appropriately choosing such a function, not only can we account for all models seen so far in the literature, but we can also derive novel ones. Casting all the different models in a common framework allows us to compare their geometric properties on common experimental grounds."
CVPR	Towards Accurate Recovery of Shape from Shading under Diffuse Lighting.	A. James Stewart,Michael S. Langer	1996	A new surface radiance model for diffuse lighting is presented which incorporates shadows, interreflections, and surface orientation. An algorithm is presented that uses this model to compute shape-from-shading under diffuse lighting. The algorithm is tested on both synthetic and real images, and is found to perform more accurately than the only previous algorithm for this problem.
CVPR	Runway Obstacle Detection by Controlled Spatiotemporal Image Flow Disparity.	Sanghoon Sull,Banavar Sridhar	1996	This paper proposes a method for detecting obstacles on a runway by controlling their expected disparities. By approximating the runway by a planar surface, the initial model flow field (MFF) corresponding to an obstacle-free runway is described by the data from on-board sensors (OBS). The error variance of the initial MFF is computed and used to estimate the MFF. Obstacles are detected by comparing the expected residual flow disparities with the residual flow field (RFF) estimated after warping (or stabilizing) an image using the MFF. Expected temporal and spatial disparities are obtained from the use of the OBS. This allows us to control the residual disparities by increasing the temporal baseline and/or by utilizing the spatial baseline if distant objects cannot be detected for a given temporal baseline. Experimental results for two real flight image sequences are presented.
CVPR	Efficient Image Gradient-Based Object Localisation and Recognitio.	T. N. Tan,Geoffrey D. Sullivan,Keith D. Baker	1996	This paper reports novel algorithms for the efficient localization and recognition of vehicles in traffic scenes, which eliminate the need for explicit symbolic feature extraction and matching. The algorithms make use of two a priori sources of knowledge about the scene and the objects: (i) the ground-plane constraint, and (ii) the fact that road vehicles are strongly rectilinear. The algorithms are demonstrated and tested using routine outdoor traffic images. Success with a variety of vehicles demonstrates the efficiency and robustness of context-based computer vision in road traffic scenes. The limitations of the algorithms are also addressed in the paper.
CVPR	Non-Rigid Matching Using Demons.	Jean-Philippe Thirion	1996	"We present the concept of non-rigid matching based on demons, by reference to Maxwell's demons. We contrast this concept with the more conventional viewpoint of attraction. We show that demons and attractive points are clearly distinct for large deformations, but also that they become similar for small displacements, encompassing techniques close to optical flow. We describe a general iterative matching method based on demons, and derive from it three different non-rigid matching algorithms, one using all the image intensities, one using only contours, and one for already segmented images. At last, we present results with synthesized and real deformations, with applications to Computer Vision and Medical Image Processing."
CVPR	Comparison of Approaches to Egomotion Computation.	Tina Yu Tian,Carlo Tomasi,David J. Heeger	1996	We evaluated six algorithms for computing egomotion from image velocities. We established benchmarks for quantifying bias and sensitivity to noise, and for quantifying the convergence properties of those algorithms that require numerical search. Our simulation results reveal some interesting and surprising results. First, it is often written in the literature that the egomotion problem is difficult because translation (e.g., along the X-axis) and rotation (e.g., about the Y-axis) produce similar image velocities. We found, to the contrary, that the bias and sensitivity of our six algorithms are totally invariant with respect to the axis of rotation. Second, it is also believed by some that fixating helps to make the egomotion problem easier. We found, to the contrary, that fixating does not help when the noise is independent of the image velocities. Fixation does help if the noise is proportional to speed, but this is only for the trivial reason that the speeds are slower under fixation. Third, it is widely believed that increasing the field of view will yield better performance. We found, to the contrary, that this is not necessarily true.
CVPR	Factorization Methods for Projective Structure and Motion.	Bill Triggs	1996	"This paper describes a family of factorization-based algorithms that recover 3D projective structure and motion from multiple uncalibrated perspective images of 3D points and lines. They can be viewed as generalizations of the Tomasi-Kanade algorithm from affine to fully perspective cameras, and from points to lines. They make no restrictive assumptions about scene or camera geometry, and unlike most existing reconstruction methods they do not rely on `privileged' points or images. All of the available image data is used, and each feature in each image is treated uniformly. The key to projective factorization is the recovery of a consistent set of projective depths (scale factors) for the image points: this is done using fundamental matrices and epipoles estimated from the image data. We compare the performance of the new techniques with several existing ones, and also describe an approximate factorization method that gives similar results to SVD-based factorization, but runs much more quickly for large problems."
CVPR	A Framework for Recognizing a Facial Image from a Police Sketch.	Robert G. Uhl Jr.,Niels da Vitoria Lobo	1996	We present a theory and practical computations for automatically matching a police artist sketch to a set of true photographs. We locate facial features in both the sketch as well as the set of photograph images. Then, the sketch is photometrically standardized to facilitate comparison with a photo and then both the sketch and the photos are geometrically standardized. Finally, for matching, eigenanalysis is employed. Results using real police sketches and arrest photos are presented.
CVPR	Structure and motion of curved 3D objects from monocular silhouettes.	B. Vijayakumar,David J. Kriegman,Jean Ponce	1996	"The silhouette of a smooth 3D object observed by a moving camera changes over time. Past work has shown how surface geometry can be recovered using the deformation of the silhouette when the camera motion is known. This paper addresses the problem of estimating both the full Euclidean surface structure and the camera motion from a dense set of silhouettes captured under orthographic or scaled orthographic projection. The approach relies on a viewpoint-invariant representation of curves swept by viewpoint-dependent features such as bitangents, inflections and contour points with parallel tangents. Feature points, which form stereo frontier points between non-consecutive images, are matched using this representation. The camera's angular velocity is computed from constraints derived from this correspondence along with the image velocity of these features. From the angular velocity, the epipolar geometry is ascertained, and infinitesimal motion frontier points can be detected. In turn, the motion of these frontier points constrains the translation component of camera motion. Finally, the surface is reconstructed using established techniques once the camera motion has been estimated."
CVPR	Blurring Strategies for Image Segmentation Using a Multiscale Linking Model.	Koen L. Vincken,Wiro J. Niessen,Max A. Viergever	1996	"Multiscale approaches are an invaluable tool for image segmentation. A vast amount of research has been devoted to the construction of different multiscale representations of an image. In this paper we use the `hyperstack'--a multiscale linking model for image segmentation--for an in-depth comparison of four different scale space generators with respect to segmentation results. We consider the linear (Gaussian) scale space both in the spatial and the Fourier domain, the variable conductance diffusion according to the Perona & Malik equation, and the Euclidean shortening flow. We have done experiments on MR images of the brain, for which a gold standard is available. The hyperstack proofs to be rather insensitive to the underlying scale space generator."
CVPR	Determining Correspondences and Rigid Motion of 3-D Point Sets with Missing Data.	Xiaoguang Wang,Yong-Qing Cheng,Robert T. Collins,Allen R. Hanson	1996	This paper addresses the general 3-D rigid motion problem, where the point correspondences and the motion parameters between two sets of 3-D points are to be recovered. The existence of missing points in the two sets is the most difficult problem. We first show a mathematical symmetry in the solutions of rotation parameters and point correspondences. A closed-form solution based on the correlation matrix eigenstructure decomposition is proposed for correspondence recovery with no missing points. Using a heuristic measure of point pair affinity derived from the eigenstructure, a weighted bipartite matching algorithm is developed to determine the correspondences in general cases where missing points occur. The use of the affinity heuristic also leads to a fast outlier removal algorithm, which can be run iteratively to refine the correspondence recovery. Simulation results and experiments on real images are shown in both ideal and general cases.
CVPR	Illumination and geometry invariant recognition of texture in color images.	Lizhi Wang,Glenn Healey	1996	We develop a method for recognizing color texture independent of rotation, scale, and illumination. Color texture is modeled using spatial correlation functions defined within and between sensor bands. Using a linear model for surface spectral reflectance with the same number of parameters as the number of sensor classes, we show that illumination and geometry changes in the scene correspond to a linear transformation of the correlation functions and a linear transformation of their coordinates. A several step algorithm which includes scale estimation and correlation moment computation is used to achieve the invariance. The key to the method is the new result that illumination and geometry changes in the scene correspond to a specific transformation of correlation function Zernike moment matrices. These matrices can be estimated from a color image. This relationship is used to derive an efficient algorithm for recognition. The algorithm is substantiated using classification results on over 200 images of color textures obtained under various illumination conditions and geometric configurations.
CVPR	Minimal operator set for passive depth from defocus.	Masahiro Watanabe,Shree K. Nayar	1996	A fundamental problem in depth from defocus is the measurement of relative defocus between images. We propose a class of broadband operators that, when used together, provide invariance to scene texture and produce accurate and dense depth maps. Since the operators are broadband, a small number of them are sufficient for depth estimation of scenes with complex textural properties. Experiments are conducted on both synthetic and real scenes to evaluate the performance of the proposed operators. The depth detection gain error is less than 1%, irrespective of texture frequency. Depth accuracy is found to be 0.5/spl sim/1.2% of the distance of the object from the imaging optics.
CVPR	A Unified Mixture Framework for Motion Segmentation: Incorporating Spatial Coherence and Estimating the Number of Models.	Yair Weiss,Edward H. Adelson	1996	"Describing a video sequence in terms of a small number of coherently moving segments is useful for tasks ranging from video compression to event perception. promising approach is to view the motion segmentation problem in a mixture estimation framework. However, existing formulations generally use only the motion data and thus fail to make use of static cues when segmenting the sequence. Furthermore, the number of models is either specified in advance or estimated outside the mixture model framework. In this work we address both of these issues. We show how to add spatial constraints to the mixture formulations and present a variant of the EM algorithm that makes use of both the form and the motion constraints. Moreover this algorithm estimates the number of segments given knowledge about the level of model failure expected in the sequence. The algorithm's performance is illustrated on synthetic and real image sequences."
CVPR	Nonlinear Shape Restoration For Document Images.	Yun Weng,Qiuming Zhu	1996	Previous researches on nonlinear shape restoration are based on the assumptions that the original shapes and distortions of the images have known formulations under certain conditions. The main contribution of this research is the development of a new restoration algorithm, called Multi-Steps Restoration. The algorithm is based on a liner interpolation theory that is able to detect and restore nonlinear shape distortions in any irregular quadrilateral-shaped patterns. The main idea of the algorithm is the use of two-dimensional spline functions in bicubic, biquadratic, and/or bilinear models to approximate the three-dimensional nonlinear distortion curves. The performance of the approach shown by experiment is promising.
CVPR	Attention Control for Robot Vision.	Carl-Fredrik Westin,Carl-Johan Westelius,Hans Knutsson,Gösta H. Granlund	1996	"Focus of attention mechanisms for robot vision are discussed. A new method for neglecting low level filter responses from already modelled structures is presented. The method is based on a filtering technique termed normalized convolution. In one experiment, the robot is continuously moving its arm in the scene while tracking other objects. It is shown how the arm can be made ``invisible'' so that only the moving object of interest is detected. This makes tracking of objects much simpler. In another experiment, the attention of the system is shifted between objects by simply cancelling the mask of the object to be attended to. With this strategy the low level processes do not need to know the difference between a new object entering the scene and a mask being cancelled, and thus a complex communication structure between high and low levels is avoided."
CVPR	Local Parallel Computation of Stochastic Completion Fields.	Lance R. Williams,David W. Jacobs	1996	We describe a local parallel method for computing the stochastic completion field introduced in an earlier paper\cite{williams}. The stochastic completion field represents the likelihood that a completion joining two contour fragments passes through any given position and orientation in the image plane. It is based upon the assumption that the prior probability distribution of completion shape can be modeled as a random walk in a lattice of discrete positions and orientations. The local parallel method can be interpreted as a stable finite difference scheme for solving the underlying Fokker-Planck equation identified by Mumford\cite{mumford}. The resulting algorithm is significantly faster than the previously employed method which relied on convolution with large-kernel filters computed by Monte Carlo simulation. The complexity of the new method is O(n3m) while that of the previous algorithm was O(n4m2) (for an n X n image with m discrete orientations). Perhaps most significantly, the use of a local method allows us to model the probability distribution of completion shape using stochastic processes which are neither homogenous nor isotropic. For example, it is possible to modulate particle decay rate by a directional function of local image brightnesses (i.e., anisotropic decay). The effect is that illusory contours can be made to respect the local image brightness structure. Finally, we note that the new method is more plausible as a neural model since 1) unlike the previous method, it can be computed in a sparse, locally connected network; and 2) the network dynamics are consistent with psychophysical measurements of the time course of illusory contour formation.
CVPR	Gauging Relational Consistency and Correcting Structural Errors.	Richard C. Wilson,Edwin R. Hancock	1996	The aim of this paper is to provide a comparative evaluation of a number of contrasting approaches to relational matching. Unique to this study is the way in which we show how a diverse family of algorithms relate to one-another using a common Bayesian framework. Broadly speaking there are two main aspects to this study. Firstly we focus on the issue of how relational inexactness may be quantified. We illustrate that several popular relational distance measures can be recovered as specific limiting cases of the same Bayesian consistency measure. The second aspect of our comparison concerns the way in which structural inexactness is controlled. We investigate three different realisations of the matching process which draw on contrasting control models. The main conclusion of our study is that the active process of graph-editing outperforms the alternatives in terms of its ability to effectively control a large population of contaminating clutter.
CVPR	Polarization Phase-Based Method for Material Classification and Object Recognition in Computer Vision.	Larry Wolff	1996	A robust and accurate polarization phase-based technique for material classification is presented. The novelty of this technique is three-fold in (i) its theoretical development, (ii) its application, and, (iii) its experimental implementation. The concept of phase of polarization of a light wave is introduced to computer vision for discrimination between materials according to their intrinsic electrical conductivity, such as distinguishing conducting metals, and poorly-conducting dielectrics. Previous work has used intensity, color and polarization component ratios. This new method is based on the physical principle that metals retard orthogonal components of light upon reflection while dielectrics do not. This method has significant complementary advantages with respect to existing techniques, is computationally efficient, and can be easily implemented with existing imaging technology. Experiments for real circuit board inspection, non-conductive and conductive glass, and, outdoor object recognition have been performed to demonstrate its accuracy and potential capabilities.
CVPR	Combination of Multiple Classifiers Using Local Accuracy Estimates.	Kevin Woods,Kevin W. Bowyer,W. Philip Kegelmeyer	1996	Combination of Multiple Classifiers Using Local Accuracy Estimates.
CVPR	Video Browsing Using Edges and Motion.	Ramin Zabih,Justin Miller,Kevin Mai	1996	Video Browsing Using Edges and Motion.
CVPR	FRAME: Filters, Random fields, and Minimax Entropy-- Towards a Unified Theory for Texture Modeling.	Song Chun Zhu,Ying Nian Wu,David Mumford	1996	In this paper, a minimax entropy principle is studied, based on which a novel theory, called FRAME (Filters, Random fields And Minimax Entropy) is proposed for texture modeling. FRAME combines attractive aspects of two important themes in texture modeling: multi-channel filtering and Markov random field (MRF) modeling. It incorporates the responses of a set of well selected filters into the distribution over a random field, and hence has a much stronger descriptive ability than the traditional MRF models. Furthermore, it interprets and clarifies many previous concepts and methods for texture analysis and synthesis from a unified point of view. Algorithms are proposed for probability inference, stochastic simulation and filter selection. Experiments on a variety of textures are described to illustrate our theory and to show the performance of our algorithms. These experiments demonstrate that many textures previously considered as different categories can be modeled and synthesized in a common framework.
CVPR	Isotropic Processing for Gradient Estimation.	Djemel Ziou,Shengrui Wang	1996	This paper concerns the influence of edge direction on the estimation of edge contrast and orientation. We show that the gradient estimated using radial filters is not affected by edge orientation. For non-radial filters the gradient can be affected by edge orientation. For instance, we find that the estimated edge orientation using a non-radial filter may be biased, even if the signal is noise-free. However, there are non-radial filters for which gradient is unaffected by edge orientation as in the case of radial filters. The properties of these functions are given in this paper. The results are illustrated by the study of the Canny, Deriche, and Shen & Castan detectors. We take into account discretization errors. These results give a clear indication of the effect of the rotation invariance property of an edge detector on its response, thus providing a more precise meaning for this property in edge detection.
CVPR	"1996 Conference on Computer Vision and Pattern Recognition (CVPR '96), June 18-20, 1996 San Francisco, CA, USA"		1996	"1996 Conference on Computer Vision and Pattern Recognition (CVPR '96), June 18-20, 1996 San Francisco, CA, USA"
CVPR	Incremental Focus of Attention for Robust Visual Tracking.	Kentaro Toyama,Gregory D. Hager	1996	Incremental Focus of Attention for Robust Visual Tracking.
IEEE Visualization	A Linear Iteration Time Layout Algorithm for Visualising High-Dimensional Data.	Matthew Chalmers	1996	A Linear Iteration Time Layout Algorithm for Visualising High-Dimensional Data.
IEEE Visualization	Cheops: A Compact Explorer for Complex Hierarchies.	Luc Beaudoin,Marc-Antoine Parent,Louis C. Vroomen	1996	Cheops: A Compact Explorer for Complex Hierarchies.
IEEE Visualization	BLaC-Wavelets: A Multiresolution Analysis With Non-Nested Spaces.	Georges-Pierre Bonneau,Stefanie Hahmann,Gregory M. Nielson	1996	BLaC-Wavelets: A Multiresolution Analysis With Non-Nested Spaces.
IEEE Visualization	Visualization of Complex Models Using Dynamic Texture-based Simplification.	Daniel G. Aliaga	1996	Visualization of Complex Models Using Dynamic Texture-based Simplification.
IEEE Visualization	Directional Flow Visualization of Vector Fields.	Ed Boring,Alex Pang	1996	Directional Flow Visualization of Vector Fields.
IEEE Visualization	Breaking the Myth: a Picture is NOT (always) Worth a 1, 000 Words.	Robert Braham,F. David Fracchia,Andrew S. Glassner,Barbara Mones-Hattal,Russ Rose,Nahum D. Gershon	1996	Breaking the Myth: a Picture is NOT (always) Worth a 1, 000 Words.
IEEE Visualization	A Haptic Interaction Method for Volume Visualization.	Ricardo S. Avila,Lisa M. Sobierajski	1996	A Haptic Interaction Method for Volume Visualization.
IEEE Visualization	Time Management, Simultaneity and Time-Critical Computation in Interactive Unsteady Visualization Environments.	Steve Bryson,Sandy Johan	1996	Time Management, Simultaneity and Time-Critical Computation in Interactive Unsteady Visualization Environments.
IEEE Visualization	FEL: The Field Encapsulation Library.	Steve Bryson,David N. Kenwright,Michael J. Gerald-Yamasaki	1996	FEL: The Field Encapsulation Library.
IEEE Visualization	Mathematical Visualization: Standing at the Crossroads.	David Banks,George Francis,Andrew J. Hanson,Loki Jorgenson	1996	Mathematical Visualization: Standing at the Crossroads.
IEEE Visualization	Data Reduction and Interpolation for Visualizing 3D Soil-Quality Data.	David C. Banks,Bernd Hamann,Po-Yu Tsai,Robert J. Moorhead,Jonathan Barlow	1996	Data Reduction and Interpolation for Visualizing 3D Soil-Quality Data.
IEEE Visualization	History Consideration in Reconstructuring Polythedral Surfaces from Parallel Slices.	Gill Barequet,Daniel Shapiro,Ayellet Tal	1996	History Consideration in Reconstructuring Polythedral Surfaces from Parallel Slices.
IEEE Visualization	Case Study: Visual Access for Landscape Event Based Temporal Data.	M. Sheelagh T. Carpendale,Andrew Fall,David J. Cowperthwaite,Joseph Fall,F. David Fracchia	1996	Case Study: Visual Access for Landscape Event Based Temporal Data.
IEEE Visualization	Flexible Information Visualization of Multivariate Data from Biological Sequence Similarity Searches.	Ed Huai-hsin Chi,John Riedl,Elizabeth Shoop,John V. Carlis,Ernest Retzel,Phillip Barry	1996	Flexible Information Visualization of Multivariate Data from Biological Sequence Similarity Searches.
IEEE Visualization	Temporal Continuity of Levels of Detail in Delaunay Triangualted Terrain.	Daniel Cohen-Or,Yishay Levanoni	1996	Temporal Continuity of Levels of Detail in Delaunay Triangualted Terrain.
IEEE Visualization	Contour Blending Using Warp-Guided Distance Field Interpolation.	Daniel Cohen-Or,David Levin,Amira Solomovici	1996	Contour Blending Using Warp-Guided Distance Field Interpolation.
IEEE Visualization	Real-time Slicing of Data Space.	Roger Crawfis	1996	Real-time Slicing of Data Space.
IEEE Visualization	Optimizing Triangle Strips for Fast Rendering.	Francine Evans,Steven Skiena,Amitabh Varshney	1996	Optimizing Triangle Strips for Fast Rendering.
IEEE Visualization	Deformable Volume Rendering by 3D Texture Mapping and Octree Encoding.	Shiaofen Fang,Rajagopalan Srinivasan,Su Huang,Raghu Raghavan	1996	Deformable Volume Rendering by 3D Texture Mapping and Octree Encoding.
IEEE Visualization	Fast Stereo Volume Rendering.	Taosong He,Arie E. Kaufman	1996	Fast Stereo Volume Rendering.
IEEE Visualization	Visualization of Water Quality Data for the Chesapeake Bay.	Adam B. Forgang,Bernd Hamann,Carl F. Cerco	1996	Visualization of Water Quality Data for the Chesapeake Bay.
IEEE Visualization	Raycasting Vector Fields.	Thomas Frühauf	1996	Raycasting Vector Fields.
IEEE Visualization	Two-handed Interactive Stereoscopic Visualization.	David S. Ebert,Christopher D. Shaw,Amen Zwa,Cindy Starr	1996	Two-handed Interactive Stereoscopic Visualization.
IEEE Visualization	Real-Time Incremental Visualization of Dynamic Ultrasound Volumes Using Parallel BSP Trees.	William F. Garrett,Henry Fuchs,Mary C. Whitton,Andrei State	1996	Real-Time Incremental Visualization of Dynamic Ultrasound Volumes Using Parallel BSP Trees.
IEEE Visualization	A System for Measuring Surface Facet Orientation from Atomic Force Microscope Data.	John G. Hagedorn,Holly E. Rushmeier,John Blendell,Mark Vaudin	1996	A System for Measuring Surface Facet Orientation from Atomic Force Microscope Data.
IEEE Visualization	Triangular NURBS Surface Modeling of Scattered Data.	Song Han,Gérard G. Medioni	1996	Triangular NURBS Surface Modeling of Scattered Data.
IEEE Visualization	Generation of Transfer Functions with Stochastic Search Techniques.	Taosong He,Lichan Hong,Arie E. Kaufman,Hanspeter Pfister	1996	Generation of Transfer Functions with Stochastic Search Techniques.
IEEE Visualization	Choosing Effective Colours for Data Visualization.	Christopher G. Healey	1996	Choosing Effective Colours for Data Visualization Christopher G. Healey In this paper we describe a technique for choosing multiple colours for use during data visualization. Our goal is a systematic method for maximizing the total number of colours available for use, while still allowing an observer to rapidly and accurately search a display for any one of the given colours. Previous research suggests that we need to consider three separate effects during colour selection: colour distance, linear separation, and colour category. We describe a simple method for measuring and controlling all of these effects. Our method was tested by performing a set of target identification studies; we analysed the ability of thirty-eight observers to find a colour target in displays that contained differently coloured background elements. Results showed that our method can be used to select a group of colours that will provide good differentiation between data elements during visualization.
IEEE Visualization	Untangling Knots by Stochastic Energy Optimization.	Milana Huang,Robert P. Grzeszczuk,Louis H. Kauffman	1996	Untangling Knots by Stochastic Energy Optimization.
IEEE Visualization	Perceptualisation using a Tactile Mouse.	Robert G. Hughes,A. Robin Forrest	1996	Perceptualisation using a Tactile Mouse.
IEEE Visualization	Illustrating Transparent Surfaces with Curvature-Directed Strokes.	Victoria Interrante,Henry Fuchs,Stephen M. Pizer	1996	Illustrating Transparent Surfaces with Curvature-Directed Strokes.
IEEE Visualization	Volume Thinning for Automatic Isosurface Propagation.	Takayuki Itoh,Yasushi Yamaguchi,Koji Koyamada	1996	Volume Thinning for Automatic Isosurface Propagation.
IEEE Visualization	Using Visualization in the Archaeological Excavations of Pre-Inca Temple in Peru.	Alan D. Kalvin,Alfredo Remy,Orlando Ardito,Kim Morla,Eduardo Nolasco,Jorge Prado,Regulo Franco,Antonio Murga,Guillermo Wiese	1996	Using Visualization in the Archaeological Excavations of Pre-Inca Temple in Peru.
IEEE Visualization	Anatomy-Based Facial Tissue Modeling Using the Finite Element Method.	Erwin Keeve,Sabine Girod,Paula Pfeifle,Bernd Girod	1996	Anatomy-Based Facial Tissue Modeling Using the Finite Element Method.
IEEE Visualization	Multi-Frequency Noise for LIC.	Ming-Hoe Kiu,David C. Banks	1996	Multi-Frequency Noise for LIC.
IEEE Visualization	Mesh Reduction with Error Control.	Reinhard Klein,Gunther Liebich,Wolfgang Straßer	1996	Mesh Reduction with Error Control.
IEEE Visualization	UFLOW: Visualizing Uncertainty in Fluid Flow.	Suresh K. Lodha,Alex Pang,Robert E. Sheehan,Craig M. Wittenbrink	1996	UFLOW: Visualizing Uncertainty in Fluid Flow.
IEEE Visualization	LISTEN: Sounding Uncertainty Visualization.	Suresh K. Lodha,Catherine M. Wilson,Robert E. Sheehan	1996	LISTEN: Sounding Uncertainty Visualization.
IEEE Visualization	Surface Rendering versus Volume Rendering in Medical Imaging.	William E. Lorensen,Ron Kikinis,Sandy Napel,Arie E. Kaufman,John Flynn	1996	Surface Rendering versus Volume Rendering in Medical Imaging.
IEEE Visualization	A Visualization Tool for Mechanical Design.	Shao-Chiung Lu,Alec B. Rebello,D. H. Cui,Roni Yagel,R. A. Miller,G. L. Kinzel	1996	A Visualization Tool for Mechanical Design.
IEEE Visualization	A 3D Contextual Shading Method for Visualization of Diecasting Defects.	Shao-Chiung Lu,Alec B. Rebello,D. H. Cui,Roni Yagel,R. A. Miller,G. L. Kinzel	1996	A 3D Contextual Shading Method for Visualization of Diecasting Defects.
IEEE Visualization	Fast Perspective Volume Rendering with Splatting by Utilizing a Ray-Driven Approach.	Klaus Mueller,Roni Yagel	1996	Fast Perspective Volume Rendering with Splatting by Utilizing a Ray-Driven Approach.
IEEE Visualization	Three Dimensional Visualization of Proteins in Cellular Interactions.	Colin R. F. Monks,Patricia Crossno,George S. Davidson,Constantine J. Pavlakos,Abraham Kupfer,Cláudio T. Silva,Brian N. Wylie	1996	Three Dimensional Visualization of Proteins in Cellular Interactions.
IEEE Visualization	Interactive Visulization of Ocean Circulation Models.	Scott Nations,Robert J. Moorhead,Kelly P. Gaither,Steve Aukstakalnis,Rhonda Vickery,Warren Carl Couvillion Jr.,Daniel N. Fox,Peter Flynn,Alan J. Wallcraft,Patrick Hogan,Ole Martin Smedstad	1996	Interactive Visulization of Ocean Circulation Models.
IEEE Visualization	Virtual Workbench - A Non-Immersive Virtual Environment for Visualizing and Interacting with 3D Objects for Scientific Visualization.	Upul Obeysekare,Chas Williams,Jim Durbin,Lawrence J. Rosenblum,Robert Rosenberg,Fernando Grinstein,Ravi Ramamurthi,Alexandra Landsberg,William Sandberg	1996	Virtual Workbench - A Non-Immersive Virtual Environment for Visualizing and Interacting with 3D Objects for Scientific Visualization.
IEEE Visualization	Dynamic View-Dependent Simplification for Polygonal Models.	Julie C. Xia,Amitabh Varshney	1996	Dynamic View-Dependent Simplification for Polygonal Models.
IEEE Visualization	Mantle Convection Visualization on the Cray T3D.	James S. Painter,Hans-Peter Bunge,Yarden Livnat	1996	Mantle Convection Visualization on the Cray T3D.
IEEE Visualization	Electrical Energy Absorption in the Human Head from a Cellular Telephone.	Vishram Pandit,Robert McDermott,Gianluca Lazzi,Cynthia Furse,Om Gandhi	1996	Electrical Energy Absorption in the Human Head from a Cellular Telephone.
IEEE Visualization	Real-Time Accelerators for Volume Rendering.	Hanspeter Pfister,Günter Knittel,Jürgen Hesser,John C. Goble	1996	Real-Time Accelerators for Volume Rendering.
IEEE Visualization	Information Exploration Shootout.	Gregory Piatetsky-Shapiro,Graham J. Wills	1996	Information Exploration Shootout.
IEEE Visualization	Opacity-modulating Triangular Textures for Irregular Surfaces.	Penny Rheingans	1996	Opacity-modulating Triangular Textures for Irregular Surfaces.
IEEE Visualization	A System for the Complementary Visualization of 3D Volume Images Using 2D and 3D Binaurally Processed Sonification Representations.	David Rossiter,Wai-Yin Ng	1996	A System for the Complementary Visualization of 3D Volume Images Using 2D and 3D Binaurally Processed Sonification Representations.
IEEE Visualization	Flow Visualization for Turbomachinery Design.	Martin Roth,Ronald Peikert	1996	Flow Visualization for Turbomachinery Design.
IEEE Visualization	Visualization of Laser Confocal Microscopy Datasets.	Georgios Sakas,Michael G. Vicker,Peter Jörg Plath	1996	Visualization of Laser Confocal Microscopy Datasets.
IEEE Visualization	The Design and Implementation of an Object-Oriented Toolkit for 3D Graphics and Visualization.	William J. Schroeder,Ken Martin,William E. Lorensen	1996	The Design and Implementation of an Object-Oriented Toolkit for 3D Graphics and Visualization.
IEEE Visualization	Octree-Based Decimation of Marching Cubes Surfaces.	Raj Shekhar,Elias Fayyad,Roni Yagel,J. Fredrick Cornhill	1996	Octree-Based Decimation of Marching Cubes Surfaces.
IEEE Visualization	Isosurfacing in Span Space with Utmost Efficiency (ISSUE).	Han-Wei Shen,Charles D. Hansen,Yarden Livnat,Christopher R. Johnson	1996	Isosurfacing in Span Space with Utmost Efficiency (ISSUE).
IEEE Visualization	Volume Tracking.	Deborah Silver,Xin Wang	1996	Volume Tracking.
IEEE Visualization	Keynote Address: Bringing Visualization to the User (Abstract).	Alvy Ray Smith	1996	Keynote Address: Bringing Visualization to the User (Abstract).
IEEE Visualization	Visualization of Roughness in Musical Consonance.	Florian Sobieczky	1996	Visualization of Roughness in Musical Consonance.
IEEE Visualization	Data Level Comparative Visualization in Aircraft Design.	Jens C. Trapp,Hans-Georg Pagendarm	1996	Data Level Comparative Visualization in Aircraft Design.
IEEE Visualization	Wavelets Applied to Lossless Compression and Progressive Transmission of Floating Point Data in 3-D Curvilinear Grids.	Aaron Trott,Robert J. Moorhead,John McGinley	1996	Wavelets Applied to Lossless Compression and Progressive Transmission of Floating Point Data in 3-D Curvilinear Grids.
IEEE Visualization	A Fast Gibbs Sampler for Synthesizing Constrained Fractals.	Baba C. Vemuri,Chhandomay Mandal	1996	A Fast Gibbs Sampler for Synthesizing Constrained Fractals.
IEEE Visualization	Capstone Address: Color, Pattern, and the Human Visual System (Abstract).	Brian A. Wandell	1996	Capstone Address: Color, Pattern, and the Human Visual System (Abstract).
IEEE Visualization	Complex-Valued Contour Meshing.	Chris Weigle,David C. Banks	1996	Complex-Valued Contour Meshing.
IEEE Visualization	Interactive Exploration and Modeling of Large Data Sets: A Case Study with Venus Light Scattering Data.	Jarke J. van Wijk,Hans J. W. Spoelder,Willem-Jan Knibbe,Kamran Eftekhari Shahroudi	1996	We present a system where visualization and the control of the simulation are integrated to facilitate interactive exploration and modeling of large data sets. The system was developed to estimate properties of the atmosphere of Venus from comparison between measured and simulated data. Reuse of results, distributed computing, and multiple views on the data were the major ingredients to create an effective environment.
IEEE Visualization	Hierarchical and Parallelizable Direct Volume Rendering for Irregular and Multiple Grids.	Jane Wilhelms,Allen Van Gelder,Paul Tarantino,Jonathan Gibbs	1996	Hierarchical and Parallelizable Direct Volume Rendering for Irregular and Multiple Grids.
IEEE Visualization	Multiresolution Multidimensional Wavelet Brushing.	Pak Chung Wong,R. Daniel Bergeron	1996	Multiresolution Multidimensional Wavelet Brushing.
IEEE Visualization	Visualization Over the World Wide Web and its Application to Environmental Data.	Jason Wood,Ken Brodlie,Helen Wright	1996	Visualization Over the World Wide Web and its Application to Environmental Data.
IEEE Visualization	The Challenges of Visualizing and Modeling Environmental Data.	Yingcai Xiao,John P. Ziebarth,Chuck Woodbury,Eric Bayer,Bruce Rundell,Jeroen van der Zijp	1996	The Challenges of Visualizing and Modeling Environmental Data.
IEEE Visualization	Clinical Evaluation of Interactive Volume Visualization.	Karel J. Zuiderveld,Peter M. A. van Ooijen,John W. C. Chin-A-Woeng,Pieter C. Buijs,Marco Olree,Frits H. Post	1996	Clinical Evaluation of Interactive Volume Visualization.
IEEE Visualization	Interactive Visualiztion of 3D-Vector Fields using Illuminated Streamlines.	Malte Zöckler,Detlev Stalling,Hans-Christian Hege	1996	Interactive Visualiztion of 3D-Vector Fields using Illuminated Streamlines.
IEEE Visualization	"Visualization '96, Proceedings, San Francisco, CA, USA, October 27 - November 1, 1996."	Roni Yagel,Gregory M. Nielson	1996	"Visualization '96, Proceedings, San Francisco, CA, USA, October 27 - November 1, 1996."
IEEE Trans. Vis. Comput. Graph.	Structured Penumbral Irradiance Computation.	George Drettakis,Eugene Fiume	1996	A definitive understanding of irradiance behavior in penumbral regions has been hard to come by, mainly due to the computational expense of determining the visible parts of an area light source. Consequently, sampling strategies have been mostly ad hoc, and evaluation of the resulting approximations has been difficult. In this paper, the structure of penumbral irradiance is investigated empirically and numerically. This study has been made feasible by the use of the discontinuity mesh and the backprojection, an efficient data structure representing visibility in regions of partial occlusion. Regions of penumbrae in which irradiance varies nonmonotonically are characterized empirically, and numerical tests are performed to determine the frequency of their occurrence. This study inspired the development of two algorithms for the construction of interpolating approximations to irradiance: One algorithm reduces the number of edges in the mesh defining the interpolant domain, and the other algorithm chooses among linear, quadratic, and mixed interpolants based on irradiance monotonicity. Results from numerical tests and images are presented that demonstrate good performance of the new algorithms for various realistic test configurations.
IEEE Trans. Vis. Comput. Graph.	Frequency Analysis of Gradient Estimators in Volume Rendering.	Mark J. Bentum,Barthold B. A. Lichtenbelt,Thomas Malzbender	1996	Gradient information is used in volume rendering to classify and color samples along a ray. In this paper, we present an analysis of the theoretically ideal gradient estimator and compare it to some commonly used gradient estimators. A new method is presented to calculate the gradient at arbitrary sample positions, using the derivative of the interpolation filter as the basis for the new gradient filter. As an example, we will discuss the use of the derivative of the cubic spline. Comparisons with several other methods are demonstrated. Computational efficiency can be realized since parts of the interpolation computation can be leveraged in the gradient estimation.
IEEE Trans. Vis. Comput. Graph.	Splatting of Non Rectilinear Volumes Through Stochastic Resampling.	Xiaoyang Mao	1996	This paper extends the conventional splatting algorithm for volume rendering non rectilinear grids. A stochastic sampling technique called Poisson sphere/ellipsoid is employed to adaptively resample a non rectilinear grid with a set of randomly distributed points whose energy support extents are well approximated by spheres or ellipsoids. Then volume rendered images can be generated by splatting the scalar values at the new sample points with filter kernels corresponding to these spheres and ellipsoids. Experiments have been carried out to investigate the image quality as well as the time/space efficiency of the new approach, and the results suggest that our approach can be regarded as an alternative for existing fast volume rendering techniques of non rectilinear grids.
IEEE Trans. Vis. Comput. Graph.	Volumetric Data Exploration Using Interval Volume.	Issei Fujishiro,Yuji Maeda,Hiroshi Sato,Yuriko Takeshima	1996	A new type of geometric model called Interval volume for volumetric data exploration is presented. An interval volume represents a three-dimensional subvolume for which the associate scalar values lie within a user-specified interval, and provides one of the promising approaches to solid fitting, which is an extended concept of traditional surface fitting. A well known isosurfacing algorithm called Marching Cubes is extended to obtain a solid fitting algorithm, which extracts from a given volumetric data set a high resolution, polyhedral solid data structure of an interval volume. Branch-on-Need Octree is used as an auxiliary data structure to accelerate the extraction process. A variety of interval volume rendering methods and principal related operations, including measurements and focusing, are also presented. The effectiveness of measurement-coupled visualization capabilities of the presented approach is demonstrated by application to visualizing a four-dimensional simulated data from atomic collision research.
IEEE Trans. Vis. Comput. Graph.	Efficient Triangular Surface Approximations Using Wavelets and Quadtree Data Structures.	Markus H. Gross,Oliver G. Staadt,Roger Gatti	1996	"We present a new method for adaptive surface meshing and triangulation which controls the local level-of-detail of the surface approximation by local spectral estimates. These estimates are determined by a wavelet representation of the surface data. The basic idea is to decompose the initial data set by means of an orthogonal or semi-orthogonal tensor product wavelet transform (WT) and to analyze the resulting coefficients. In surface regions, where the partial energy of the resulting coefficients is low, the polygonal approximation of the surface can be performed with larger triangles without loosing too much fine grain details. However, since the localization of the WT is bound by the Heisenberg principle the meshing method has to be controlled by the detail signals rather than directly by the coefficients. The dyadic scaling of the WT stimulated us to build an hierarchical meshing algorithm which transforms the initially regular data grid into a quadtree representation by rejection of unimportant mesh vertices. The optimum triangulation of the resulting quadtree cells is carried out by selection from a look-up table. The tree grows recursively as controlled by detail signals which are computed from a modified inverse WT.In order to control the local level-of-detail, we introduce a new class of wavelet space filters acting as ""magnifying glasses"" on the data.We show that our algorithm performs a low algorithmic complexity, so that surface meshing can be achieved at interactive rates, such as required by flight simulators. However, other applications are possible as well, such as mesh reduction in complex data, FEM or radiosity meshing.The method is applied on different types of data comprising both digital terrain models and laser range scans. In addition, quantitative investigations on error analysis are carried out."
IEEE Trans. Vis. Comput. Graph.	Fractal Volume Compression.	Wayne O. Cochran,John C. Hart,Patrick J. Flynn	1996	This research explores the principles, implementation, and optimization of a competitive volume compression system based on fractal image compression. The extension of fractal image compression to volumetric data is trivial in theory. However, the simple addition of a dimension to existing fractal image compression algorithms results in infeasible compression times and noncompetitive volume compression results. This paper extends several fractal image compression enhancements to perform properly and efficiently on volumetric data, and introduces a new 3D edge classification scheme based on principal component analysis. Numerous experiments over the many parameters of fractal volume compression suggest aggressive settings of its system parameters. At this peak efficiency, fractal volume compression surpasses vector quantization and approaches within 1 dB PSNR of the discrete cosine transform. When compared to the DCT, fractal volume compression represents surfaces in volumes exceptionally well at high compression rates, and the artifacts of its compression error appear as noise instead of deceptive smoothing or distracting ringing.
IEEE Trans. Vis. Comput. Graph.	A Real-Time Photo-Realistic Visual Flythrough.	Daniel Cohen-Or,Eran Rich,Uri Lerner,Victor Shenkar	1996	In this paper we present a comprehensive flythrough system which generates photo-realistic images in true real-time. The high performance is due to an innovative rendering algorithm based on a discrete ray casting approach, accelerated by ray coherence and multiresolution traversal. The terrain as well as the 3D objects are represented by a textured mapped voxel-based model. The system is based on a pure software algorithm and is thus portable. It was first implemented on a workstation and then ported to a general-purpose parallel architecture to achieve real-time performance.
IEEE Trans. Vis. Comput. Graph.	Editorial.		1996	Editorial.
IEEE Trans. Vis. Comput. Graph.	Visualizing Unstructured Flow Data Using Dual Stream Functions.	David Knight,Gordon Mallinson	1996	One of the most important ways of visualizing fluid flow is the construction of streamlines, which are lines that are everywhere tangential to the local fluid velocity. Stream surfaces are defined as surfaces through which no fluid penetrates. Streamlines can therefore be computed from the intersection of two nonparallel stream surfaces. This paper presents new algorithms for the computation of dual stream functions from Computational Fluid Dynamics data that is defined on an unstructured tetrahedral mesh. These algorithms are compared with standard numerical routines for computing streamlines, and are shown to be quicker and more accurate than techniques involving numerical integration along the streamline.
IEEE Trans. Vis. Comput. Graph.	Interactive Display of Large NURBS Models.	Subodh Kumar,Dinesh Manocha,Anselmo Lastra	1996	We present algorithms for interactive rendering of large-scale NURBS models. The algorithms convert the NURBS surfaces to Bézier surfaces, tessellate each Bézier surface into triangles, and render them using the triangle-rendering capabilities common to current graphics systems. This paper presents algorithms for computing tight bounds on surface properties in order to generate high quality tessellation of Bézier surfaces. We introduce enhanced visibility determination techniques and present methods to make efficient use of coherence between successive frames. In addition, we also discuss issues in parallelization of these techniques. The algorithm also avoids polygonization anomalies like cracks. Our algorithms work well in practice and, on high-end graphics systems, are able to display models described using thousands of Bézier surfaces at interactive frame rates.
IEEE Trans. Vis. Comput. Graph.	Analysis of a Parallel Volume Rendering System Based on the Shear-Warp Factorization.	Philippe Lacroute	1996	This paper presents a parallel volume rendering algorithm that can render a 256 × 256 × 225 voxel medical data set at over 15 Hz and a 512 × 512 × 334 voxel data set at over 7 Hz on a 32-processor Silicon Graphics Challenge. The algorithm achieves these results by minimizing each of the three components of execution time: computation time, synchronization time, and data communication time. Computation time is low because the parallel algorithm is based on the recently-reported shear-warp serial volume rendering algorithm which is over five times faster than previous serial algorithms. The algorithm uses run-length encoding to exploit coherence and an efficient volume traversal to reduce overhead. Synchronization time is minimized by using dynamic load balancing and a task partition that minimizes synchronization events. Data communication costs are low because the algorithm is implemented for shared-memory multiprocessors, a class of machines with hardware support for low-latency fine-grain communication and hardware caching to hide latency.We draw two conclusions from our implementation. First, we find that on shared-memory architectures data redistribution and communication costs do not dominate rendering time. Second, we find that cache locality requirements impose a limit on parallelism in volume rendering algorithms. Specifically, our results indicate that shared-memory machines with hundreds of processors would be useful only for rendering very large data sets.
IEEE Trans. Vis. Comput. Graph.	Controlled Topology Simplification.	Taosong He,Lichan Hong,Amitabh Varshney,Sidney W. Wang	1996	We present a simple, robust, and practical method for object simplification for applications where gradual elimination of high frequency details is desired. This is accomplished by converting an object into multi-resolution volume rasters using a controlled filtering and sampling technique. A multiresolution triangle-mesh hierarchy can then be generated by applying the Marching Cubes algorithm. We further propose an adaptive surface generation algorithm to reduce the number of triangles generated by the standard Marching Cubes. Our method simplifies the topology of objects in a controlled fashion. In addition, at each level of detail, multilayered meshes can be used for an efficient antialiased rendering.
IEEE Trans. Vis. Comput. Graph.	Solving Geometric Constraints By Homotopy.	Hervé Lamure,Dominique Michelucci	1996	Solving Geometric Constraints By Homotopy.
IEEE Trans. Vis. Comput. Graph.	Image Composition Schemes for Sort-Last Polygon Rendering on 2D Mesh Multicomputers.	Tong-Yee Lee,C. S. Raghavendra,John B. Nicholas	1996	"In a sort-last polygon rendering system, the efficiency of image composition is very important for achieving fast rendering. In this paper, the implementation of a sort-last rendering system on a general purpose multicomputer system is described. A two-phase sort-last-full image composition scheme is described first, and then many variants of it are presented for 2D mesh message-passing multicomputers, such as the Intel Delta and Paragon. All the proposed schemes are analyzed and experimentally evaluated on Caltech's Intel Delta machine for our sort-last parallel polygon renderer. Experimental results show that sort-last-sparse strategies are better suited than sort-last-full schemes for software implementation on a general purpose multicomputer system. Further, interleaved composition regions perform better than coherent regions. In a large multicomputer system, performance can be improved by carefully scheduling the tasks of rendering and communication. Using 512 processors to render our test scenes, the peak rendering rate achieved on a 262, 144 triangle dataset is close to 4.6 million triangles per second which is comparable to the speed of current state-of-the-art graphics workstations."
IEEE Trans. Vis. Comput. Graph.	A Road Map To Solid Modeling.	Christoph M. Hoffmann,Jarek Rossignac	1996	The objective of solid modeling is to represent, manipulate, and reason about, the three-dimensional shape of solid physical objects, by computer. Such representations should be unambiguous.Solid modeling is an application-oriented field that began in earnest in the early 1970s. [46]. Major application areas include design, manufacturing, computer vision, graphics, and virtual reality. Technically, the field draws on diverse sources including numerical analysis, symbolic algebraic computation, approximation theory, applied mathematics, point set topology, algebraic geometry, computational geometry, and data bases. Monographs and major surveys of solid modeling include [13], [19], [27], [37], [44], [45], [46].In this road map article, we begin with some mathematical foundations of the field. We review next the major representation schemata of solids. Then, major layers of abstraction in a typical solid modeling system are characterized: The lowest level of abstraction comprises a substratum of basic service algorithms. At an intermediate level of abstraction there are algorithms for larger, more conceptual operations. Finally, a yet higher level of abstraction presents to the user a functional view that is typically targeted towards solid design. Here, we will look at some applications and at user interaction concepts.The classical design paradigms of Solid Modeling concentrated on obtaining one specific final shape. Those paradigms are becoming supplanted by feature-based, constraint-based design paradigms that are oriented more toward the design process and define classes of shape instances. These new paradigms venture into territory that has yet to be explored systematically. Concurrent with this paradigm shift, there is also a shift in the system architecture towards modularized confederations of plug-compatible functional components. We explore these trends lightly in the last section.
IEEE Trans. Vis. Comput. Graph.	Image Metamorphosis with Scattered Feature Constraints.	Seungyong Lee,George Wolberg,Kyung-Yong Chwa,Sung Yong Shin	1996	This paper describes an image metamorphosis technique to handle scattered feature constraints specified with points, polylines, and splines. Solutions to the following three problems are presented: feature specification, warp generation, and transition control. We demonstrate the use of snakes to reduce the burden of feature specification. Next, we propose the use of multilevel free-form deformations (MFFD) to compute C2-continuous and one-to-one mapping functions among the specified features. The resulting technique, based on B-spline approximation, is simpler and faster than previous warp generation methods. Furthermore, it produces smooth image transformations without undesirable ripples and foldovers. Finally, we simplify the MFFD algorithm to derive transition functions to control geometry and color blending. Implementation details are furnished and comparisons among various metamorphosis techniques are presented.
IEEE Trans. Vis. Comput. Graph.	RIVA: A Versatile Parallel Rendering System for Interactive Scientific Visualization.	Peggy Li,William H. Duquette,David W. Curkendall	1996	"JPL's Remote Interactive Visualization and Analysis System (RIVA) is described in detail. The RIVA system integrates workstation graphics, massively parallel computing technology, and gigabit communication networks to provide a flexible interactive environment for scientific data perusal, analysis, and visualization. RIVA's kernel is a highly scalable parallel perspective renderer tailored especially for the demands of large datasets beyond the sensible reach of workstations. Early experience with using RIVA to interactively explore and process multivariate, multiresolution datasets is reported; several examples using data from a variety of remote sensing instruments are discussed in detail and the results shown. Particular attention is placed on describing the algorithmic details of RIVA's parallel renderer kernel, with emphasis on the key aspects of achieving the algorithm's overall scalability. The paper summarizes the performance achieved for machine sizes up to more than 500 nodes and for initial input image/terrain bases in the 2 Gbyte range."
IEEE Trans. Vis. Comput. Graph.	Editorial.		1996	Editorial.
IEEE Trans. Vis. Comput. Graph.	A Near Optimal Isosurface Extraction Algorithm Using the Span Space.	Yarden Livnat,Han-Wei Shen,Christopher R. Johnson	1996	"We present the ""Near Optimal IsoSurface Extraction"" (NOISE) algorithm for rapidly extracting isosurfaces from structured and unstructured grids. Using the span space, a new representation of the underlying domain, we develop n isosurface extraction algorithm with a worst case complexity of $O\left ({\sqrt n} + k\right )$ for the search phase, where n is the size of the data set and k is the number of cells intersected by the isosurface. The memory requirement is kept at O(n) while the preprocessing step is O(n log n). We utilize the span space representation as a tool for comparing isosurface extraction methods on structured and unstructured grids. We also present a fast triangulation scheme for generating and displaying unstructured tetrahedral grids."
IEEE Trans. Vis. Comput. Graph.	Correction: A Near Optimal Isosurface Extraction Algorithm Using the Span Space.	Yarden Livnat,Han-Wei Shen,Christopher R. Johnson	1996	Correction: A Near Optimal Isosurface Extraction Algorithm Using the Span Space.
IEEE Trans. Vis. Comput. Graph.	Editorial.	Arie E. Kaufman	1996	Editorial.
IEEE Trans. Vis. Comput. Graph.	Reconstruction Error Characterization and Control: A Sampling Theory Approach.	Raghu Machiraju,Roni Yagel	1996	"Reconstruction is prerequisite whenever a discrete signal needs to be resampled as a result of transformation such as texture mapping, image manipulation, volume slicing, and rendering. We present a new method for the characterization and measurement of reconstruction error in spatial domain. Our method uses the Classical Shannon's Sampling Theorem as a basis to develop error bounds. We use this formulation to provide, for the first time, an efficient way to guarantee an error bound at every point by varying the size of the reconstruction filter. We go further to support position-adaptive reconstruction and data-adaptive reconstruction which adjust filter size to the location of reconstruction point and to the data values in its vicinity. We demonstrate the effectiveness of our methods with 1D signals, 2D signals (images), and 3D signals (volumes)."
IEEE Trans. Vis. Comput. Graph.	Interactive Time-Dependent Particle Tracing Using Tetrahedral Decomposition.	David N. Kenwright,David A. Lane	1996	Streak lines and particle traces are effective visualization techniques for studying unsteady fluid flows. For real-time applications, accuracy is often sacrificed to achieve interactive frame rates. Physical space particle tracing algorithms produce the most accurate results although they are usually too expensive for interactive applications. An efficient physical space algorithm is presented in this paper which was developed for interactive investigation and visualization of large, unsteady, aeronautical simulations. Performance has been increased by applying tetrahedral decomposition to speed up point location and velocity interpolation in curvilinear grids. Preliminary results from batch computations [1] showed that this approach was up to six times faster than the most common algorithm which uses the Newton-Raphson method and trilinear interpolation. Results presented here show that the tetrahedral approach also permits interactive computation and visualization of unsteady particle traces. Statistics are given for frame rates and computation times on single and multiprocessors. The benefits of interactive feature detection in unsteady flows are also demonstrated.
IEEE Trans. Vis. Comput. Graph.	A Digital Brain Atlas for Surgical Planning, Model-Driven Segmentation, and Teaching.	Ron Kikinis,Martha Elizabeth Shenton,Dan V. Iosifescu,Robert W. McCarley,Pairash Saiviroonporn,Hiroto H. Hokama,Andre Robatino,David Metcalf,Cindy Wible,Chiara M. Portas,Robert M. Donnino,Ferenc A. Jolesz	1996	We developed a three-dimensional (3D) digitized atlas of the human brain to visualize spatially complex structures. It was designed for use with magnetic resonance (MR) imaging data sets. Thus far, we have used this atlas for surgical planning, model-driven segmentation, and teaching. We used a combination of automated and supervised segmentation methods to define regions of interest based on neuroanatomical knowledge. We also used 3D surface rendering techniques to create a brain atlas that would allow us to visualize complex 3D brain structures. We further linked this information to script files in order to preserve both spatial information and neuroanatomical knowledge. We present here the application of the atlas for visualization in surgical planning for model-driven segmentation and for the teaching of neuroanatomy. This digitized human brain has the potential to provide important reference information for the planning of surgical procedures. It can also serve as a powerful teaching tool, since spatial relationships among neuroanatomical structures can be more readily envisioned when the user is able to view and rotate the structures in 3D space. Moreover, each element of the brain atlas is associated with a name tag, displayed by a user-controlled pointer. The atlas holds a major promise as a template for model-driven segmentation. Using this technique, many regions of interest can be characterized simultaneously on new brain images.
IEEE Trans. Vis. Comput. Graph.	Theme Issue Introduction: Challenges in Visualization Research.	Gregory M. Nielson	1996	Theme Issue Introduction: Challenges in Visualization Research.
IEEE Trans. Vis. Comput. Graph.	Implementation and Analysis of an Image-Based Global Illumination Framework for Animated Environments.	Jeffry Nimeroff,Julie Dorsey,Holly E. Rushmeier	1996	"We describe a new framework for efficiently computing and storing global illumination effects for complex, animated environments. The new framework allows the rapid generation of sequences representing any arbitrary path in a ""view space"" within an environment in which both the viewer and objects move. The global illumination is stored as time sequences of range-images at base locations that span the view space. We present algorithms for determining locations for these base images, and the time steps required to adequately capture the effects of object motion. We also present algorithms for computing the global illumination in the base images that exploit spatial and temporal coherence by considering direct and indirect illumination separately. We discuss an initial implementation using the new framework. Results and analysis of our implementation demonstrate the effectiveness of the individual phases of the approach; we conclude with an application of the complete framework to a complex environment that includes object motion."
IEEE Trans. Vis. Comput. Graph.	D-NURBS: A Physics-Based Framework for Geometric Design.	Hong Qin,Demetri Terzopoulos	1996	This paper presents dynamic NURBS, or D-NURBS, a physics-based generalization of Non-Uniform Rational B-Splines. NURBS have become a de facto standard in commercial modeling systems because of their power to represent both free-form shapes and common analytic shapes. Traditionally, however, NURBS have been viewed as purely geometric primitives, which require the designer to interactively adjust many degrees of freedom (DOFs)¿control points and associated weights¿to achieve desired shapes. The conventional shape modification process can often be clumsy and laborious. D-NURBS are physics-based models that incorporate mass distributions, internal deformation energies, forces, and other physical quantities into the NURBS geometric substrate. Their dynamic behavior, resulting from the numerical integration of a set of nonlinear differential equations, produces physically meaningful, hence intuitive shape variation. Consequently, a modeler can interactively sculpt complex shapes to required specifications not only in the traditional indirect fashion, by adjusting control points and setting weights, but also through direct physical manipulation, by applying simulated forces and local and global shape constraints. We use Lagrangian mechanics to formulate the equations of motion for D-NURBS curves, tensor-product D-NURBS surfaces, swung D-NURBS surfaces, and triangular D-NURBS surfaces. We apply finite element analysis to reduce these equations to efficient numerical algorithms computable at interactive rates on common graphics workstations. We implement a prototype modeling environment based on D-NURBS and demonstrate that D-NURBS can be effective tools in a wide range of CAGD applications such as shape blending, scattered data fitting, and interactive sculpting.
IEEE Trans. Vis. Comput. Graph.	Volume-Preserving Free-Form Solids.	Ari Rappoport,Alla Sheffer,Michel Bercovier	1996	Volume-Preserving Free-Form Solids.
IEEE Trans. Vis. Comput. Graph.	Shape Description By Medial Surface Construction.	Damian J. Sheehy,Cecil G. Armstrong,Desmond J. Robinson	1996	The medial surface is a skeletal abstraction of a solid that provides useful shape information, which compliments existing model representation schemes. The medial surface and its associated topological entities are defined, and an algorithm for computing the medial surface of a large class of B-rep solids is then presented. The algorithm is based on the domain Delaunay triangulation of a relatively sparse distribution of points, which are generated on the boundary of the object. This strategy is adaptive in that the boundary point set is refined to guarantee a correct topological representation of the medial surface.
IEEE Trans. Vis. Comput. Graph.	An Algorithm for the Medial Axis Transform of 3D Polyhedral Solids.	Evan C. Sherbrooke,Nicholas M. Patrikalakis,Erik Brisson	1996	The medial axis transform (MAT) is a representation of an object which has been shown to be useful in design, interrogation, animation, finite element mesh generation, performance analysis, manufacturing simulation, path planning, and tolerance specification. In this paper, an algorithm for determining the MAT is developed for general 3D polyhedral solids of arbitrary genus without cavities, with nonconvex vertices and edges. The algorithm is based on a classification scheme which relates different pieces of the medial axis (MA) to one another even in the presence of degenerate MA points. Vertices of the MA are connected to one another by tracing along adjacent edges, and finally the faces of the axis are found by traversing closed loops of vertices and edges. Representation of the MA and associated radius function is addressed, and pseudocode for the algorithm is given along with recommended optimizations. A connectivity theorem is proven to show the completeness of the algorithm. Complexity estimates and stability analysis for the algorithms are presented. Finally, examples illustrate the computational properties of the algorithm for convex and nonconvex 3D polyhedral solids with polyhedral holes.
IEEE Trans. Vis. Comput. Graph.	Function Representation for Sweeping by a Moving Solid.	Alexei Sourin,Alexander A. Pasko	1996	Function Representation for Sweeping by a Moving Solid.
IEEE Trans. Vis. Comput. Graph.	Efficient Streamline, Streamribbon, and Streamtube Constructions on Unstructured Grids.	Shyh-Kuang Ueng,Christopher A. Sikorski,Kwan-Liu Ma	1996	Streamline construction is one of the most fundamental techniques for visualizing steady flow fields. Streamribbons and streamtubes are extensions for visualizing the rotation and the expansion of the flow. This paper presents efficient algorithms for constructing streamlines, streamribbons, and streamtubes on unstructured grids. A specialized Runge-Kutta method is developed to speed up the tracing of streamlines. Explicit solutions are derived for calculating the angular rotation rates of streamribbons and the radii of streamtubes. In order to simplify mathematical formulations and reduce computational costs, all calculations are carried out in the canonical coordinate system instead of the physical coordinate system. The resulting speed-up in overall performance helps explore large flow fields.
IEEE Trans. Vis. Comput. Graph.	Feature Extraction and Iconic Visualization.	Theo van Walsum,Frits H. Post,Deborah Silver,Frank J. Post	1996	We present a conceptual framework and a process model for feature extraction and iconic visualization. The features are regions of interest extracted from a dataset. They are represented by attribute sets, which play a key role in the visualization process. These attribute sets are mapped to icons, or symbolic parametric objects, for visualization. The features provide a compact abstraction of the original data, and the icons are a natural way to visualize them. We present generic techniques to extract features and to calculate attribute sets, and describe a simple but powerful modeling language which was developed to create icons and to link the attributes to the icon parameters. We present illustrative examples of iconic visualization created with the techniques described, showing the effectiveness of this approach.
IEEE Trans. Vis. Comput. Graph.	Glyphs for Visualizing Uncertainty in Vector Fields.	Craig M. Wittenbrink,Alex Pang,Suresh K. Lodha	1996	Environmental data have inherent uncertainty which is often ignored in visualization. Meteorological stations and doppler radars, including their time series averages, have a wealth of uncertainty information that traditional vector visualization methods such as meteorological wind barbs and arrow glyphs simply ignore. We have developed a new vector glyph to visualize uncertainty in winds and ocean currents. Our approach is to include uncertainty in direction and magnitude, as well as the mean direction and length, in vector glyph plots. Our glyph shows the variation in uncertainty, and provides fair comparisons of data from instruments, models, and time averages of varying certainty. We also define visualizations that incorporate uncertainty in an unambiguous manner as verify visualization. We use both quantitative and qualitative methods to compare our glyphs to traditional ones. Subjective comparison tests with experts are provided, as well as objective tests, where the information density of our new glyphs and traditional glyphs are compared. The design of the glyph and numerous examples using environmental data are given. We show enhanced visualization, data together with their uncertainty information, that may improve understanding of environmental vector field data quality.
IEEE Trans. Vis. Comput. Graph.	A Fast Method for Estimating Discrete Field Values in Early Engineering Design.	Jovan Zagajac	1996	A Fast Method for Estimating Discrete Field Values in Early Engineering Design.
CVPR	Pictorial Recognition Using Affine-Invariant Spectral Signatures.	Jezekiel Ben-Arie,Zhiqian Wang	1997	This paper describes an efficient approach to pose invariant object recognition employing pictorial recognition of image patches. A complete affine invariance is achieved by a representation which is based on a new sampling configuration in the frequency domain. Employing Singular Value Decomposition (SVD), the affine transform is decomposed into slant, tilt, swing, scale and 2D translation. From this decomposition, we derive an affine invariant representation that allows to recognize image patches that correspond to object surfaces which are roughly planar -- invariant to their pose in space. The representation is in the form of Spectral Signatures that are derived from a set of Cartesian logarithmic-logarithmic (log-log) sampling configuration in the frequency domain. Unlike previous log-polar representations which are not invariant to slant (i.e. foreshortening only in one direction), our new configuration yields complete affine invariance. The proposed log-log configuration can be employed both globally or locally by a Gabor or Fourier transforms. Local representation enables to recognize separately several objects in the same image. The actual signature recognition is performed by multi-dimensional indexing in a pictorial dataset represented in the frequency domain. The recognition also provides 3D pose information.
CVPR	Novel view synthesis in tensor space.	Shai Avidan,Amnon Shashua	1997	We present a new method for synthesizing novel views of a 3D scene from few model images in full correspondence. The core of this work is the derivation of a tensorial operator that describes the transformation from a given tensor of three views to a novel tensor of a new configuration of three views. By repeated application of the operator on a seed tensor with a sequence of desired virtual camera positions we obtain a chain of warping functions (tensors) from the set of model images to create the desired virtual views.
CVPR	MDL estimation for small sample sizes and its application to segmenting binary strings.	Byron Dom	1997	Minimum Description Length (MDL) estimation has proven itself of major importance in a large number of applications many of which are in the fields of computer vision and pattern recognition. A problem is encountered in applying the associated formulas, however, especially those associated with model cost. This is because most of these are asymptotic forms appropriate only for large sample sizes. J. Rissanen has recently derived sharper code-length formulas valid for much smaller sample sizes. Because of the importance of these results, it is our intent here to present a tutorial description of them. In keeping with this goal we have chosen a simple application whose relative tractability allows it to be explored more deeply than most problems: the segmentation of binary strings based on a piecewise Bernoulli assumption. By that we mean that the strings are assumed to be divided into substrings, the bits of which are assumed to have been generated by a single (within a substring) Bernoulli source.
CVPR	Optic flow calculation using robust statistics.	Alireza Bab-Hadiashar,David Suter	1997	A method for calculating optic flow, using robust statistics, is developed. The method generally out-performs all competing methods in terms of accuracy. One of the key features in the success of this method, is that we use least median of squares, which is known to be robust to outliers. The computational cost is kept very low by using an approximate solution to the least median of squares only in a first stage that detects outliers. The essential ingredients of our method should be applicable in a wide range of other computer vision problems.
CVPR	3D Reconstruction of the Human Jaw from A Sequence of Images.	Moumen T. Ahmed,Sameh M. Yamany,Elsayed E. Hemayed,S. Ahmed,S. Roberts,Aly A. Farag	1997	"A novel approach is proposed to obtain a record of the patient's occlusion using computer vision. Data acquisition is obtained using intra-oral video cameras. The technique utilizes shape from shading to extract 3D information from 2D views of the jaw, and a novel technique for 3D data registration using genetic algorithms. The resulting 3D model can be used for diagnosis, treatment planning, and implant purposes. The overall purpose of this research is to develop a model-based vision system for orthodontics to replace traditional approaches. This system will be flexible, accurate, and will reduce the cost of orthodontic treatments."
CVPR	Modelling of single mode distributions of colour data using directional statistics.	Daniel C. Alexander,Bernard F. Buxton	1997	Three different statistical models of colour data for use in segmentation or tracking algorithms are proposed. Results of a performance comparison of a tracking algorithm, applied to two separate applications, using each of the three different types of underlying model of the data are presented. From these a comparison of the performance of the statistical colour models themselves is obtained.
CVPR	Verifying model-based alignments in the presence of uncertainty.	T. D. Alter,W. Eric L. Grimson	1997	This paper introduces a unified approach to the problem of verifying alignment hypotheses in the presence of substantial amounts of uncertainty in the predicted locations of projected model features. Our approach is independent of whether the uncertainty is distributed or bounded, and, moreover, incorporates information about the domain in a formally correct manner. Information which can be incorporated includes the error model, the distribution of background features, and the positions of the data features near each predicted model feature. Experiments are described that demonstrate the improvement over previously used methods. Furthermore, our method is efficient in that the number of operations is on the order of the number of image features that lie nearby the predicted model features.
CVPR	Model-Based approach to Accurate and Consistent 3--D Modeling of Drainage and Surrounding Terrain.	Pascal Fua	1997	"We propose an automated approach to modeling drainage channels---and, more generally, linear features that lie on the terrain---from multiple images, which results not only in high-resolution, accurate and consistent models of the features, but also of the surrounding terrain.In our specific case, we have chosen to exploit the fact that rivers flow downhill and lie at the bottom of local depressions in the terrain, valley floors tend to be ``U'' shaped, and the drainage pattern appears as a network of linear features that can be visually detected in single gray-level images.Different approaches have explored individual facets of this problem. Ours unifies these elements in a common framework. We accurately model terrain and features as 3--dimensional objects from several information sources that may be in error and inconsistent with one another. This approach allows us to generate models that are faithful to sensor data, internally consistent and consistent with physical constraints. The ACM Portal is published by the Association for Computing Machinery. Copyright © 2010 ACM, Inc. Terms of Usage Privacy Policy Code of Ethics Contact Us Useful downloads: Adobe Acrobat QuickTime Windows Media Player Real Player"
CVPR	Photometric Computation of the Sign of Gaussian Curvature Using a Curve-Orientation Invariant.	Elli Angelopoulou,Lawrence B. Wolff	1997	We compute the sign of Gaussian curvature using a purely geometric definition. Consider a point p on a smooth surface S and a closed curve g on S which encloses p. The image of g on the unit normal Gaussian sphere is a new curve b. The Gaussian curvature at p is defined as the ratio of the area enclosed by g over the area enclosed by b as g contracts to p. The sign of Gaussian curvature at p is determined by the relative orientations of the closed curves g and b. We directly compute the relative orientation of two such curves from intensity data. We employ three unknown illumination conditions to create a photometric scatter plot. This plot is in one-to-one correspondence with the subset of the unit Gaussian sphere containing the mutually illuminated surface normals. This permits direct computation of the sign of Gaussian curvature without the recovery of surface normals. Our method is albedo invariant. We assume diffuse reflectance, but the nature of the diffuse reflectance can be general and unknown. Simulations, as well as empirical results, demonstrate the accuracy of our technique.
CVPR	A Curvature Based Descriptor Invariant to Pose and Albedo Derived from Photometric Data.	Elli Angelopoulou,James P. Williams,Lawrence B. Wolff	1997	Gaussian curvature is an invariant local descriptor of smooth surfaces. We present an object signature which is a condensed representation of the distribution of Gaussian curvature information at visible object points. An invariant related to Gaussian curvature at a point is derived from the covariance matrix of the photometric values in a neighborhood about that point. In addition, we introduce an albedo-normalization method that is capable of cancelling albedo on Lambertian surfaces. We use three illumination conditions, two of which are unknown. The three-tuple of intensity values at a point is related via a one-to-one mapping to the surface normal at that point. The determinant of the covariance matrix of the local three-tuples is invariant to albedo, rotation and translation. The collection of determinants over mutually illuminated object points is combined into a signature distribution which is albedo, rotation, translation, and scale invariant. An object recognition methodology using these signatures is proposed.
CVPR	Stereo-Motion that Complements Stereo and Motion Analyses.	Pui-Kuen Ho,Ronald Chung	1997	Recovering three-dimensional (3D) information of a scene from its images is a fundamental problem in computer vision. There exists two major multi-ocular cues for it, namely the motion cue and the stereo cue. This paper presents a new approach of integrating the two cues when two cameras that move through the scene while taking pictures repeatedly are available. The approach is based on the Singular Value Decomposition (SVD) technique, with which the 3D structure of the scene, the image projection parameters, the motion parameters, and the stereo geometry are all separated. The approach offers the advantages of both cues: simple correspondence as well as accurate reconstruction. It can also work with relatively short image sequences.
CVPR	Shape Indexing Using Approximate Nearest-Neighbour Search in High-Dimensional Spaces.	Jeffrey S. Beis,David G. Lowe	1997	Shape indexing is a way of making rapid associations between features detected in an image and object models that could have produced them. When model databases are large, the use of high-dimensional features is critical, due to the improved level of discrimination they can provide. Unfortunately, finding the nearest neighbour to a query point rapidly becomes inefficient as the dimensionality of the feature space increases. Past indexing methods have used hash tables for hypothesis recovery, but only in low-dimensional situations. In this paper, we show that a new variant of the k-d tree search algorithm makes indexing in higher-dimensional spaces practical. This Best Bin First, or BBF, search is an approximate algorithm which finds the nearest neighbour for a large fraction of the queries, and a very close neighbour in the remaining cases. The technique has been integrated into a fully developed recognition system, which is able to detect complex objects in real, cluttered scenes in just a few seconds.
CVPR	Independent 3D Motion Detection Based on Depth Elimination in Normal Flow Fields.	Antonis A. Argyros,Stelios C. Orphanoudakis	1997	"This paper considers a specific problem of visual perception of motion, namely the problem of visual detection of independent 3D motion. Most of the existing techniques for solving this problem rely on restrictive assumptions about the environment, the observer's motion, or both. Moreover, they are based on the computation of optical flow, which amounts to solving the ill-posed correspondence problem. In this work, independent motion detection is formulated as robust parameter estimation applied to the visual input acquired by a binocular, rigidly moving observer. Depth and motion measurements are combined in a linear model. The parameters of this model are related to the parameters of self-motion (egomotion) and the parameters of the stereoscopic configuration of the observer. The robust estimation of this model leads to a segmentation of the scene based on 3D motion. The method avoids the correspondence problem by employing only normal flow fields. Experimental results demonstrate the effectiveness of this method in detecting independent motion in scenes with large depth variations, without any constraints imposed on observer motion."
CVPR	The Bas-Relief Ambiguity.	Peter N. Belhumeur,David J. Kriegman,Alan L. Yuille	1997	When an unknown object with Lambertian reflectance is viewed orthographically, there is an implicit ambiguity in determining its 3-d structure: we show that the object&lsquo;s visible surface f(x, y) is indistinguishable from a &ldquo;generalized bas-relief&rdquo; transformation of the object&lsquo;s geometry, \bar f (x, y) &equals; &lambda;f(x, y) + &mu;x + &nu;y, and a corresponding transformation on the object&lsquo;s albedo. For each image of the object illuminated by an arbitrary number of distant light sources, there exists an identical image of the transformed object illuminated by similarly transformed light sources. This result holds both for the illuminated regions of the object as well as those in cast and attached shadows. Furthermore, neither small motion of the object, nor of the viewer will resolve the ambiguity in determining the flattening (or scaling) &lambda; of the object&lsquo;s surface. Implications of this ambiguity on structure recovery and shape representation are discussed.
CVPR	Representation Of Objects In A Volumetric Frequency Domain With Application To Face Recognition.	Jezekiel Ben-Arie,Dibyendu Nandy	1997	A novel method for representing 3-D objects that unifies viewer and model centered object representations is presented. A unified 3-D frequency-domain representation (called Volumetric/Iconic Spectral Signatures - V/ISS) encapsulates both the spatial structure of the object and a continuum of its views in the same data structure. The frequency-domain image of an object viewed from any direction can be directly extracted employing an extension of the Projection Slice Theorem, where each Fourier-transformed view is a planar slice of the volumetric frequency representation. The V/ISS representation can be employed for pose-invariant recognition of complex objects such as faces. The recognition and pose estimation is based on an efficient matching algorithm in a four dimensional Fourier space. Experimental examples of pose estimation and recognition of faces are also presented.
CVPR	Resolving Occlusion in Augmented Reality : a Contour Based Approach without 3D Reconstruction.	Marie-Odile Berger	1997	We present a new approach for resolving occlusions in augmented reality. The main interest is that it does not require 3D reconstruction of the considered scene. Our idea is to use a contour based approach and to label each contour point as being behind or in front of, depending on whether it is in front of or behind the virtual object. This labeling step only requires that the contours can be tracked from frame to frame. A proximity graph is then built in order to group the contours that belong to the same occluding object. Finally, we use some kind of active contours to accurately recover the mask of the occluding object.
CVPR	Deformable Multi Template Matching with Application to Portal Images.	Martin Berger,Gaudenz Danuser	1997	The exact positioning of patients during radiotherapy is essentialfor high precision treatment. The registration of portal image sequences can help to control the patient position. The particular problem of such megavoltage X-ray imagery is its extremely low contrast, rendering accurate feature extraction a difficult task. To circumvent the step offeature extraction, the algorithm presented in this paper relies on anarea-bused matching of the image signal using deformable templates. This strategy contrasts with most state of the art registration algorithms for portal imagery. The paper includes the mathematical formalism of the least squares template matching method, as well as the framework for automated quality control, together yielding a fast, robust and very accurate image matching procedure. Tests on I7 portal image series with more than 100 images in total have shown very satisfying results. Artificially rotated and shifted images demonstrate the performance of the method with respect to a ground truth.
CVPR	Deformable Multi Template Matching with Application to Portal Images.	Martin Berger,Gaudenz Danuser	1997	The exact positioning of patients during radiotherapy is essentialfor high precision treatment. The registration of portal image sequences can help to control the patient position. The particular problem of such megavoltage X-ray imagery is its extremely low contrast, rendering accurate feature extraction a difficult task. To circumvent the step offeature extraction, the algorithm presented in this paper relies on anarea-bused matching of the image signal using deformable templates. This strategy contrasts with most state of the art registration algorithms for portal imagery. The paper includes the mathematical formalism of the least squares template matching method, as well as the framework for automated quality control, together yielding a fast, robust and very accurate image matching procedure. Tests on I7 portal image series with more than 100 images in total have shown very satisfying results. Artificially rotated and shifted images demonstrate the performance of the method with respect to a ground truth.
CVPR	Reconstruction of 3D-Curves from 2D-Images Using Affine Shape Methods for Curves.	Rikard Berthilsson,Kalle Astrijm	1997	In this paper, we propose an algorithm for doing reconstruction of general SD-curves from a number of 2D-images taken by uncalibrated cameras. No point correspondences between the images are assumed. The curve and the view points are uniquely reconstructed, module common projective transformations and the point correspondence problem is solved. Furthermore, the algorithm is independent of the choice of coordinates, as it is based on orthogonal projections and aligning subspaces. The algorithm is based on an extension of afine shape of finite point configurations to more general objects.
CVPR	Recursive Structure and Motion from Image Sequences using Shape and Depth Spaces.	Rikard Berthilsson,Anders Heyden,Gunnar Sparr	1997	In this paper a novel recursive method for estimating structure and motion from image sequences is presented. The novelty lies in the fact that the output of the algorithm is independent of the chosen coordinate systems in the images as well as the ordering of the points. It relies on subspace methods and is derived from both ordinary coordinate representations and camera matrices and from a so called depth and shape analysis. Furthermore, no initial phase is needed to start up the algorithm. It starts directly with the first two images and incorporates new images as soon as new corresponding points are obtained. The performance of the algorithm is shown on simulated data.Moreover, the two different approaches, one using camera matrices and the other using the concepts of affine shape and depth, are unified into a general theory of structure and motion from image sequences.
CVPR	Motion estimation using ordinal measures.	Dinkar N. Bhat,Shree K. Nayar,Alok Gupta	1997	We present a method for motion estimation using ordinal measures. Ordinal measures are based on relative ordering of intensity values in an image region called rank permutation. While popular measures like the sum-of-squared-difference (SSD) and normalized correlation (NCC) rely on linearity between corresponding intensity values, ordinal measures only require them to be monotonically related so that rank permutations between corresponding regions are presented. This property turns out to be useful for motion estimation in tagged magnetic resonance images. We study the imaging equation involved in two methods of tagging and observe temporal monotonicity in intensity under certain conditions though the tags themselves fade. We compare our method to SSD and NCC in a rotating ring phantom image sequence. We present an experiment on a real heart image sequence which suggests the suitability of our method.
CVPR	Deterioration detection for digital film restoration.	Olivier Buisson,Bernard Besserer,Samia Boukir,F. Helt	1997	This paper presents a robust technique to detect local deteriorations of old cinematographic films. This method relies on spatio-temporal information and combines two different detectors: a morphological detector which uses spatial properties of deteriorations, and a dynamic detector based on motion estimation techniques. Our deterioration detector has been validated on several film sequences and turned out to be a powerful tool for digital film restoration.
CVPR	Learning Parameterized Models of Image Motion.	Michael J. Black,Yaser Yacoob,Allan D. Jepson,David J. Fleet	1997	A framework for learning parameterized models of optical flow from image sequences is presented. A class of motions is represented by a set of orthogonal basis flow fields that are computed from a training set using principal component analysis. Many complex image motions can be represented by a linear combination of a small number of these basis flows. The learned motion models may be used for optical flow estimation and for model-based recognition. For optical flow estimation we describe a robust, multi-resolution scheme for directly computing the parameters of the learned flow models from image derivatives. As examples we consider learning motion discontinuities, non-rigid motion of human mouths, and articulated human motion.
CVPR	Controlling view-based algorithms using approximate world models and action information.	Aaron F. Bobick,Claudio S. Pinhanez	1997	Most view-based vision algorithms are based on strong assumptions about the disposition of the objects in the image. To safely apply those algorithms in real world image sequences, we propose that a vision system should be divided into two components. The first component contains an approximate world model of the scene-a low accuracy, coarse description of the objects and actions in the world. Approximate world models are constructed and updated by simple vision routines and by the use of action information provided by an external source. The second component employs view-based algorithms to perform required perceptual tasks; the selection and control of the view-based methods are determined by the information provided by the approximate world model. We demonstrate the approximate world model approach in a project to control cameras in a TV studio where the external context is provided by a script.
CVPR	Experimental Performance Evaluation of Feature Grouping Modules.	Sudhir Borra,Sudeep Sarkar	1997	"We present five performance measures to evaluate grouping modules in the context of constrained search and indexing based object recognition. Using these measures, we demonstrate a sound experimental framework based on statistical ANOVA tests to compare and contrast three edge based organization modules, namely those of Etemadi et al., Jacobs, and Sarkar-Boyer in the domain of aerial objects using 50 images. With adapted parameters, the Jacobs module is overall the best choice for constraint based recognition. For fixed parameters, the Sarkar-Boyer module is the best in terms of recognition accuracy and indexing speedup. Etemadi et al.'s module performs equally well with fixed and adapted parameters while the Jacobs module is most sensitive to fixed and adapted parameter choices. The overall performance ranking of the modules is Jacobs, Sarkar-Boyer, and Etemadi et al.."
CVPR	Global Training of Document Processing Systems Using Graph Transformer Networks.	Léon Bottou,Yoshua Bengio,Yann LeCun	1997	We propose a new machine learning paradigm called Graph Transformer Networks that extends the applicability of gradient-based learning algorithms to systems composed of modules that take graphs as inputs and produce graphs as output. Training is performed by computing gradients of a global objective function with respect to all the parameters in the system using a kind of back-propagation procedure.A complete check reading system based on these concepts is described. The system uses convolutional neural network character recognizers, combined with global training techniques to provide record accuracy on business and personal checks. It is presently deployed commercially and reads million of checks per month.
CVPR	Corner Detection with Covariance Propagation.	Qiang Ji,Robert M. Haralick	1997	This paper presents a statistical approach for detecting corners from chain encoded digital arcs. An arc point is declared as a corner if the estimated parameters of the two fitted lines of the two arc segments immediately to the right and left of the arc point are statistically significantly different. The corner detection algorithm consists of two steps: corner detection and optimization. While corner detection involves statistically identifying the most likely corner points along an arc sequence, corner optimization deals with improving the locational errors of the detected corners.The major contributions of this research include developing a method for analytically estimating the covariance matrix of the fitted line parameters and developing a hypothesis test statistic to statistically test the difference between the parameters of two fitted lines. This paper discusses the theory and performance characterization of the proposed corner detector.
CVPR	Disparity Component Matching for Visual Correspondence.	Yuri Boykov,Olga Veksler,Ramin Zabih	1997	"We present a method for computing dense visual correspondence based on general assumptions about scene geometry. Our algorithm does not rely on correlation, and uses a variable region of support. We assume that images consist of a number of connected sets of pixels with the same disparity, which we call {\em disparity components}. At each pixel we compute a small set of plausible disparities, each of which is more likely to be the pixel's true disparity than not. A pixel is assigned a disparity $d$ based on connected components of pixels, where each pixel in a component considers $d$ to be plausible. Our implementation chooses the largest plausible disparity component; however, global contextual constraints can also be applied. While the algorithm was originally designed for visual correspondence, it can also be used for other early vision problems such as image restoration. It runs in a few seconds on traditional benchmark images with standard parameter settings, and gives quite promising results."
CVPR	Model-based multi-objective analysis of ultrasound image sequences in prenatal diagnosis.	H. Isil Bozma,Erdem Kemik	1997	In this study, ultrasound image sequences of fetus head are examined for the bi-parietal diameter measurement. There are five parts of the fetus head that have to be seen clearly before such a measurement can be made. Our approach is based on model-based multi-objective analysis. For each part, an objective function encoding both data and anatomic constraints is formulated. The resulting multi-objective problem is then transformed into a problem of coupled differential equations and is then solved using numerical integration. Throughout the process, continuity principle is used between the frames so that the results of analysis on one frame can be used on the next frame-whenever possible.
CVPR	Stereo Coupled Active Contours.	Tat-Jen Cham,Roberto Cipolla	1997	We consider how tracking in stereo may be enhanced by coupling pairs of active contours in different views via affine epipolar geometry and various subsets of planar affine transformations, as well as by implementing temporal constraints imposed by curve rigidity. 3D curve tracking is achieved using a submanifold model, where it is shown how the coupling mechanisms can be decomposed to cater for fixed and variable epipolar geometries. In the case of tracking planar curves, the canonical frame model is developed such that the various geometrical constraints needed in different situations may be efficiently selected. The results show that coupled active contours add consistency and robustness to tracking in stereo.
CVPR	Coupled hidden Markov models for complex action recognition.	Matthew Brand,Nuria Oliver,Alex Pentland	1997	We present algorithms for coupling and training hidden Markov models (HMMs) to model interacting processes, and demonstrate their superiority to conventional HMMs in a vision task classifying two-handed actions. HMMs are perhaps the most successful framework in perceptual computing for modeling and classifying dynamic behaviors, popular because they offer dynamic time warping, a training algorithm and a clear Bayesian semantics. However the Markovian framework makes strong restrictive assumptions about the system generating the signal-that it is a single process having a small number of states and an extremely limited state memory. The single-process model is often inappropriate for vision (and speech) applications, resulting in low ceilings on model performance. Coupled HMMs provide an efficient way to resolve many of these problems, and offer superior training speeds, model likelihoods, and robustness to initial conditions.
CVPR	Learning and Recognizing Human Dynamics in Video Sequences.	Christoph Bregler	1997	This paper describes a probabilistic decomposition of human dynamics at multiple abstractions, and shows how to propagate hypotheses across space, time, and abstraction levels. Recognition in this framework is the succession of very general low level grouping mechanisms to increased specific and learned model based grouping techniques at higher levels. Hard decision thresholds are delayed and resolved by higher level statistical models and temporal context. Low-level primitives are areas of coherent motion found by EM clustering, mid-level categories are simple movements represented by dynamical systems, and high-level complex gestures are represented by Hidden Markov Models as successive phases of simple movements. We show how such a representation can be learned from training data, and apply it to the example of human gait recognition.
CVPR	Local Blur Estimation and Super-Resolution.	Ming-Chao Chiang,Terrance E. Boult	1997	Until now, all super-resolution algorithms have presumed that the images were taken under the same illumination conditions. This paper introduces a new approach to super-resolution---based on edge models and a local blur estimate---which circumvents these difficulties. The paper presents the theory and the experimental results using the new approach.
CVPR	A Divide-and-conquer Strategy in Shape from Shading Problem.	Seong Ik Cho,Hideo Saito,Shinji Ozawa	1997	A divide-and-conquer strategy in shape from shading problem under fully perspective condition is proposed for the information recovery of book surfaces. The whole recovery process is composed of three sequential steps : preprocessing, apparent shape recovery, and ortho-image generation. Pure shade images are extracted at preprocessing step by introducing phenomenological model of interreflection and by removing pigment parts from observed images. Using existing invariances, equations of slope and that of ortho-image being explicit are derived from equation of shading and that of observation being implicit. Recurrence relation is derived from definition of mean slope in discrete image. Theoretically, it become possible to recover unique shape without iteration using derived equations in case of Lambertian cylinder. However, a feed-back shape recovery process is implemented as practical algorithm in order to overcome self-shadows. Results of simulations and real experiments show the properness and acceptability of the proposed strategy and implemented algorithms.
CVPR	Multi-Image Focus of Attention for Rapid Site Model Construction.	Robert T. Collins	1997	A multi-image focus of attention mechanism has been developed that can quickly distinguish raised objects like buildings from structured background clutter typical to many aerial image scenarios. The underlying approach is the space-sweep stereo method, in which features from multiple images are backprojected onto a virtual, horizontal plane that is methodically swept through the scene. Backprojected gradient orientations from multiple images are highly correlated when they come from scene locations containing structural edges that are roughly horizontal, like building roofs and terrain; otherwise, they tend to be uniformly distributed. These observations are used to define a structural salience measure that can determine whether a given volume of space contains a statistically signijicant number of structural edges, without first pe$otrning precise reconstruction of those edges. The utility of structural salience for computing focus of attention regions is illustrated on sample data from Ft.Hood, Texas.
CVPR	Face Detection With Information-Based Maximum Discrimination.	Antonio Colmenarez,Thomas S. Huang	1997	Antonio J. Colmenarez and Thomas S. Huang In this paper we present a visual learning technique that maximizes the discrimination between positive and negative examples in a training set. We demonstrate our technique in the context of face detection with complex background without color or motion information, which has proven to be a challenging problem. We use a family of discrete Markov processes to model the face and background patterns and estimate the probability models using the data statistics. Then, we convert the learning process into an optimization, selecting the Markov process that optimizes the information-based discrimination between the two classes. The detection process is carried out by computing the likelihood ratio using the probability model obtained from the learning procedure. We show that because of the discrete nature of these models, the detection process is, by almost two orders of magnitude, less computationally expensive than neural network approaches. However, no improvement in terms of correct-answer/false-alarm tradeoff is achieved.
CVPR	Robust analysis of feature spaces: color image segmentation.	Dorin Comaniciu,Peter Meer	1997	A general technique for the recovery of significant image features is presented. The technique is based on the mean shift algorithm, a simple nonparametric procedure for estimating density gradients. Drawbacks of the current methods (including robust clustering) are avoided. Feature space of any nature can be processed, and as an example, color image segmentation is discussed. The segmentation is completely autonomous, only its class is chosen by the user. Thus, the same program can produce a high quality edge image, or provide, by extracting all the significant colors, a preprocessor for content-based query systems. A 512/spl times/512 color image is analyzed in less than 10 seconds on a standard workstation. Gray level images are handled as color images having only the lightness coordinate.
CVPR	Depth from Scattering.	Fabio Gagliardi Cozman,Eric Krotkov	1997	Light power is affected when it crosses the atmosphere; there is a simple, albeit non-linear, relationship between the radiance of an image at any given wavelength and the distance between object and viewer. This phenomenon is called atmospheric scattering and has been extensively studied by physicists and meterologists. We present the first analysis of this phenomenon from an image understanding perspective: we investigate a group of techniques for extraction of depth cues solely from the analysis of atmospheric scattering effects in images. Depth from scattering techniques are discussed for indoor and outdoor environments, and experimental tests with real images are presented. We have found that depth cues in outdoor scenes can be recovered with surprising accuracy and can be used as an additional information source for autonomous vehicles.
CVPR	Scale-Space Vector Fields for Feature Analysis.	Andrew D. J. Cross,Edwin R. Hancock	1997	This paper describes a vectorial representation that can be used to assess the symmetry of objects in 2D images. The method exploits a magneto-static analogy. Commencing from the gradient-field extracted from filtered grey-scale images we construct a vector-potential. Our magneto-static analogy is that tangential gradient vectors represent the elements of a current distribution on the image plane. By embedding the image plane in an augmented 3-dimensional space, we compute the vector potential by performing volume integration over the current distribution. The associated magnetic field is computed by taking the curl of the vector-potential. The auxiliary spatial dimension provides a natural scale-space sampling of the generating current-distribution; as the height above the image plane is increased, so the volume over which averaging is effected also increases. We extract edge and symmetry lines through a topographic analysis of the vector-field at various heights above the image plane. Symmetry axes are lines of where the curl of the vector-potential vanishes; at edges the divergence of the vector-potential vanishes.
CVPR	Multi-Modal Tracking of Faces for Video Communications.	James L. Crowley,François Bérard	1997	This paper describes a system which uses multiple visual processes to detect and track faces for video compression and transmission. The system is based on an architecture in which a supervisor selects and activates visual processes in cyclic manner. Control of visual processes is made possible by a confidence factor which accompanies each observation. Fusion of results into a unified estimation for tracking is made possible by estimating a covariance matrix with each observation. Visual processes for face tracking are described using blink detection, normalized color histogram matching, and cross correlation (SSD and NCC). Ensembles of visual processes are organized into processing states so as to provide robust tracking. Transition between states is determined by events detected by processes. The result of face detection is fed into recursive estimator (Kalman filter). The output from the estimator drives a PD controller for a pan/tilt/zoom camera. The resulting system provides robust and precise tracking which operates continuously at approximately 20 images per second on a 150 megahertz computer work-station.
CVPR	Character extraction of license plates from video.	Yuntao Cui,Qian Huang	1997	In this paper, we present a new approach to extract characters on a license plate of a moving vehicle given a sequence of perspective distortion corrected license plate images. We model the extraction of characters as a Markov random field (MRF). With the MRF modeling, the extraction of characters is formulated as the problem of maximizing the a posteriori probability based on given prior and observations. A genetic algorithm with local greedy mutation operator is employed to optimize the objective function. Experiments and comparison study were conducted. It is shown that our approach provides better performance than other single frame methods.
CVPR	Reflectance and Texture of Real-World Surfaces Authors.	Kristin J. Dana,Shree K. Nayar,Bram van Ginneken,Jan J. Koenderink	1997	In this work, we investigate the visual appearance of real-world surfaces and the dependence of appearance on imaging conditions. We present a BRDF (bidirectional reflectance distribution function) database with reflectance measurements for over 60 different samples, each observed with over 200 different combinations of viewing and source directions. We fit the BRDF measurements to two recent models to obtain a BRDF parameter database. These BRDF parameters can be directly used for both image analysis and image synthesis. Finally, we present a BTF (bidirectional texture function) database with image textures from over 60 different samples, each observed with over 200 different combinations of viewing and source directions. Each of these unique databases has important implications for a variety of vision algorithms and each is made publicly available.
CVPR	FOCUS: Searching for Multi-colored Objects in a Diverse Image Database.	Madirakshi Das,Edward M. Riseman,Bruce A. Draper	1997	We describe a new multi-phase, color-based image retrieval system, FOCUS (Fast Object Color-based qUery System), with an online user interface which is capable of identifying multi-colored query objects in an image in the presence of significant, interfering backgrounds. The query object may occur in arbitrary sizes, orientations and locations in the database images. The color features used to describe an image have been developed based on the need for speed in matching and ease of computation on complex images while maintaining the scale and rotation invariance properties. The first phase matches the color content of an image computed as the peaks in the color histogram of the image, with the query object colors using an efficient indexing mechanism. The second phase matches the spatial relationships between color regions in the image with the query using a spatial proximity graph (SPG) structure designed for the purpose. The method is fast and has low storage overhead. Test results with multi-colored query objects from artificial and natural domains show that FOCUS is quite effective in handling interfering backgrounds and large variations in scale. The experimental results on a database of diverse images highlights the capabilities of the system.
CVPR	The Representation and Recognition of Human Movement Using Temporal Templates.	James W. Davis,Aaron F. Bobick	1997	A new view-based approach to the representation and recognition of action is presented. The basis of the representation is a temporal template -- a static vector-image where the vector value at each point is a function of the motion properties at the corresponding spatial location in an image sequence. Using 18 aerobics exercises as a test domain, we explore the representational power of a simple, two component version of the templates: the first value is a binary value indicating the presence of motion, and the second value is a function of the recency of motion in a sequence. We then develop a recognition method which matches these temporal templates against stored instances of views of known actions. The method automatically performs temporal segmentation, is invariant to linear changes in speed, and runs in real-time on a standard platform. We recently incorporated this technique into the KidsRoom: an interactive, narrative play-space for children.
CVPR	Appearance Matching of Occluded Objects Using Coarse-to-fine Adaptive Masks.	Jeff L. Edwards,Hiroshi Murase	1997	In this paper, we discuss an appearance matching technique for the interpretation of color scenes containing occluded objects. Dealing with occlusions is very difficult, and we have explored the use of an iterative, coarse-to-fine correlation-based method that uses hypothesized occlusion events to modify the scene-to-template similarity measure at run-time. Specifically, a binary mask is used to adaptively exclude regions of the template image from the correlation computation. At each iteration, these masks are adjusted based on higher resolution scene data and the occluding interactions between multiple object hypotheses. We present results which demonstrate the technique is reasonably robust over a large database of color test scenes containing objects at a variety of scales, and tolerates minor object rotations and global illumination variations.
CVPR	A robust and convergent iterative approach for determining the dominant plane from two views without correspondence and calibration.	Pär Fornland,Christoph Schnörr	1997	A robust, iterative approach is introduced for finding the dominant plane in a scene using binocular vision. Neither camera calibration nor stereo correspondence is required. Recently Cohen formalized a framework guaranteeing (local) convergence of iterative two-step methods. In this paper, the framework is adopted, with a global step using tentative matches to estimate the planar projectivity, and a local step attempting to solve the stereo correspondence. A detected point in the first image is matched to an auxiliary point in the second image, on the line joining the transformed first image point, and its closest detected second image point. Convergence is assured, while achieving robustness to both mismatching and non-coplanar points.
CVPR	Body plans.	David A. Forsyth,Margaret M. Fleck	1997	This paper describes a representation for people and animals, called a body plan, which is adapted to segmentation and to recognition in complex environments. The representation is an organized collection of grouping hints obtained from a combination of constraints on color and texture and constraints on geometric properties such as the structure of individual parts and the relationships between parts. Body plans can be learned from image data, using established statistical learning techniques. The approach is illustrated with two examples of programs that successfully use body plans for recognition: one example involves determining whether a picture contains a scantily clad human, using a body plan built by hand; the other involves determining whether a picture contains a horse, using a body plan learned from image data. In both cases, the system demonstrates excellent performance on large, uncontrolled test sets and very large and diverse control sets.
CVPR	Learning bilinear models for two-factor problems in vision.	William T. Freeman,Joshua B. Tenenbaum	1997	"In many vision problems, we want to infer two (or more) hidden factors which interact to produce our observations. We may want to disentangle illuminant and object colors in color constancy; rendering conditions from surface shape in shape-from-shading; face identity and head pose in face recognition; or font and letter class in character recognition. We refer to these two factors generically as ``style'' and ``content''.Bilinear models offer a powerful framework for extracting the two-factor structure of a set of observations, and are familiar in computational vision from several well-known lines of research. This paper shows how bilinear models can be used to learn the style-content structure of a pattern analysis or synthesis problem, which can then be generalized to solve related tasks using different styles and/or content. We focus on three tasks: extrapolating the style of data to unseen content classes, classifying data with known content under a novel style, and translating data from novel content classes and style to a known style or content. We show examples from color constancy, face pose estimation, shape-from-shading, typography and speech."
CVPR	Efficient Stereo with Multiple Windowing.	Andrea Fusiello,Vito Roberto,Emanuele Trucco	1997	We present a new, efficient stereo algorithm addressing robust disparity estimation in the presence of occlusions. The algorithm is an adaptive, multi-window scheme using left-right consistency to compute disparity and its associated uncertainty. We demonstrate and discuss performances with both synthetic and real stereo pairs, and show how our results improve on those of closely related techniques for both robustness and efficiency.
CVPR	Efficient Regularity-Based Grouping.	Jacob Feldman	1997	"This paper summarizes a novel logic-based approach to grouping and perceptual organization (presented more thoroughly in \cite{Feldman_ci}), and presents novel efficient methods for computing interpretations in this framework. Grouping interpretations are first defined as logical structures, built out of atomic premises (``regularities'') that are derived from considerations of non-accidentalness. These interpretations can then be partially ordered by their degree of regularity or constraint (measured numerically by their {\it codimension\/}). The Genericity Constraint---the principle that interpretations should minimize coincidences in the observed configuration---dictates that the preferred interpretation will be the minimum in this partial order, i.e. the interpretation with {\it maximum codimension.\/} The preferred interpretation, called the {\it qualitative parse\/}, corresponds neatly to the interpretation intuitively preferred by human observers. As a side-effect, the ``most salient'' or most structured part of the scene can be identified, as the highest-codimension subtree of the qualitative parse. An efficient ($O(n^2)$) method for computing the maximum codimension interpretation is presented, along with examples."
CVPR	Adaptive B-Splines and Boundary Estimation.	Mário A. T. Figueiredo,José M. N. Leitão,Anil K. Jain	1997	This paper describes a boundary estimation scheme based on a new adaptive approach to B-spline curve fitting. The number of control points of the spline, their locations, and the observation parameters, are all considered unknown. The optimal number of control points is estimated via a new minimum description length (MDL) type criterion. The result is an adaptive parametrically deformable contour which also estimates the observation model parameters. Experiments on synthetic and real (medical) images confirm the adequacy and good performance of the approach.
CVPR	The confounding of translation and rotation in reconstruction from multiple views.	Cornelia Fermüller,Yiannis Aloimonos	1997	If 3D rigid motion is estimated with some error a distorted version of the scene structure will in turn be computed. Of computational interest are these regions in space where the distortions are such that the depths become negative, because in order to be visible the scene has to lie in front of the image. The stability analysis for the structure-from-motion problem presented in this paper investigates the optimal relationship between the errors in the estimated translational and rotational parameters of a rigid motion, that results in the estimation of a minimum number of negative depth values. The input used is the value of the flow along some direction, which is more general than optic flow or correspondence. For a planar retina it is shown that the optimal configuration is achieved when the projections of the translational and rotational errors on the image plane are perpendicular. Furthermore, the projection of the actual and the estimated translation lie on a line passing through the image center. For a spherical retina given a rotational error, the optimal translation is the correct one, while given a translational error. The optimal rotational error is normal to the translational one at an equal distance from the real and estimated translations. The proofs, besides illuminating the confounding of translation and rotation in structure from motion, have an important application to ecological optics, explaining differences of planar and spherical eye or camera designs in motion and shape estimation.
CVPR	Efficient Approximation of Range Images Through Data-Dependent Adaptive Triangulations.	Miguel Angel García,Angel Domingo Sappa,Luis Basañez	1997	"This paper presents an efficient algorithm for generating adaptive triangular meshes from dense range images. The proposed technique consists of two stages. First, a quadrilateral mesh is generated from the given range image. The points of this mesh adapt to the surface shapes represented in the range image by grouping in areas of high curvature and dispersing in low-variation regions. The second stage splits each quadrilateral cell obtained before into two triangles. Between the two possible flips, it is chosen the one whose diagonal' s direction is closest to the orientation of the discontinuities present in that cell. Both stages avoid costly iterative optimization techniques. Results with real range images are presented. They show low CPU times and accurate triangular approximations of the given images."
CVPR	Color-Based Tracking of Heads and Other Mobile Objects at Video Frame Rates.	Paul W. Fieguth,Demetri Terzopoulos	1997	We develop a simple and very fast method for object tracking based exclusively on color information in digitized video images. Running on a Silicon Graphics R4600 Indy system with an IndyCam, our algorithm is capable of simultaneously tracking objects at full frame size ($640\times480$ pixels) and video frame rate (30 fps). Robustness with respect to occlusion is achieved via an explicit hypothesis-tree model of the occlusion process. We demonstrate the efficacy of our technique in the challenging task of tracking people, especially tracking human heads and hands.
CVPR	A region-level graph labeling approach to motion-based segmentation.	Marc Gelgon,Patrick Bouthemy	1997	This paper deals with the problem of motion-based segmentation of image sequences. Such partitions are multiple-purpose in dynamic scene analysis. We first extract a texture-based partition using an unsupervised MRF approach. The regions obtained are then grouped according to a motion-based criterion. This grouping process relies on two motion estimation techniques and exploits contextual information between regions. In contrast with clustering techniques, region grouping is formalized as a motion-based graph labeling process, within a Markovian framework. Results on real-world image sequences are shown and validate the proposed method.
CVPR	Combining Region Splitting and Edge Detection through Guided Delaunay Image Subdivision.	Theo Gevers,Arnold W. M. Smeulders	1997	In this paper, an adaptive split-and-merge segmentation method is proposed. The splitting phase of the algorithm employs the incremental Delaunay triangulation competent of forming grid edges of arbitrary orientation and position. The tessellation grid, defined by the Delaunay triangulation, is adjusted to the semantics of the image data by combining similarity and difference information among pixels. Experimental results on synthetic images show that the method is robust to different object edge orientations, partially weak object edges and very noisy homogeneous regions. Experiments on a real image indicate that the method yields good segmentation results even when there is a quadratic sloping of intensities particularly suited for segmenting natural scenes of man-made objects.
CVPR	Projective registration with difference decomposition.	Michael Gleicher	1997	Current methods for registering image regions perform well for simple transformations or large image regions. The author presents a new method that is better able to handle small image regions as they deform with nonlinear transformations. He introduces difference decomposition, a novel approach to solving the registration problem. The method is a generalization of previous methods and can better handle nonlinear transforms. Although the methods are general, he focuses on projective transformations and introduces piecewise-projective transformations for modeling the motions of non-planar objects. He concludes with examples from a prototype implementation.
CVPR	Motion of Disturbances: Detection and Tracking of multi-Body non-Rigid Motion.	Gilad Halevy,Daphna Weinshall	1997	"We present a new approach to the tracking of very non rigid patterns of motion, such as water flowing down a stream. The algorithm is based on a ""disturbance map,"" which is obtained by linearly subtracting the temporal average of the previous frames from the new frame. Every local motion creates a disturbance having the form of a wave, with a ""head"" at the present position of the motion and a historical ""tail"" that indicates the previous locations of that motion. These disturbances serve as loci of attraction for ""tracking particles"" that are scattered throughout the image. The algorithm is very fast and can be performed in real time. We provide excellent tracking results on various complex sequences, using both stabilized and moving cameras, showing: a busy ant column, waterfalls, rapids and flowing streams, shoppers in a mall, and cars in a traffic intersection."
CVPR	Lip reading from scale-space measurements.	Richard Harvey,Iain Matthews,J. Andrew Bangham,Stephen Cox	1997	Systems that attempt to recover the spoken word from image sequences usually require complicated models of the mouth and its motions. Here we describe a new approach based on a fast mathematical morphology transform called the sieve. We form statistics of scale measurements in one and two dimensions and these are used as a feature vector for standard Hidden Markov Models (HMMs).
CVPR	Hierarchical Recognition of Articulated Objects from Single Perspective Views.	Alexa Hauck,Stefan Lanser,Christoph Zierl	1997	This paper presents an approach to the recognition of articulated 3D objects in monocular video images. A hierarchical object representation models objects as a composition of rigid components which are explicitly connected by specific kinematic constraints, e.g., rotational and/or translational joints. The recognition task follows this tree-like structure by first estimating the 3D pose of the static component (root) and afterwards determining the relative 3D pose of the remaining components recursively. This method limits the search space for the actual correspondences between image and model features and copes with the problem of self-occlusion. Experiments in the context of autonomous, mobile robots show the practicability of this approach.
CVPR	Object-Based Video Indexing for Virtual Studio Productions.	Bilge Günsel,A. Murat Tekalp,Peter J. L. van Beek	1997	"This paper introduces an object-based approach for temporal video partitioning and content-based indexing, where the basic indexing unit is ``lifespan of a video object,"" rather than a ``camera shot"" or ``story unit."" We propose a system to extract content-based features of video objects (VOs), based on a compact 2D triangular mesh representation of them. An adaptive mesh-based video object tracking scheme is then employed to compute the motion trajectori of all node points. A set of ``key snapshots"" which constitute a visual summary of the lifespan of the object are automatically selected using motion and shape information. The system provides direct access to the VOs and gives the functionalities such as object-based search, manipulation, animation, and tracking."
CVPR	A Four-step Camera Calibration Procedure with Implicit Image Correction.	Janne Heikkilä,Olli Silvén	1997	In geometrical camera calibration the objective is to determine a set of camera parameters that describe the mapping between 3-D reference coordinates and 2-D image coordinates. Various methods for camera calibration can be found from the literature. However, surprisingly little attention has been paid to the whole calibration procedure, i.e., control point extraction from images, model fitting, image correction, and errors originating in these stages. The main interest has been in model fitting, although the other stages are also important. In this paper we present a four-step calibration procedure that is an extension to the two-step method. There is an additional step to compensate for distortion caused by circular features, and a step for correcting the distorted image coordinates. The image correction is performed with an empirical inverse model that accurately compensates for radial and tangential distortions. Finally, a linear method for solving the parameters of the inverse model is presented.
CVPR	Tracking non-rigid, moving objects based on color cluster flow.	Bernd Heisele,Ulrich Kressel,W. Ritter	1997	In this contribution we present an algorithm for tracking non-rigid, moving objects in a sequence of colored images, which were recorded by a non-stationary camera. The application background is vision-based driving assistance in the inner city. In an initial step, object parts are determined by a divisive clustering algorithm, which is applied to all pixels in the first image of the sequence. The feature space is defined by the color and position of a pixel. For each new image the clusters of the previous image are adapted iteratively by a parallel k-means clustering algorithm. Instead of tracking single points, edges, or areas over a sequence of images, only the centroids of the clusters are tracked. The proposed method remarkably simplifies the correspondence problem and also ensures a robust tracking behaviour.
CVPR	Real-time closed-world tracking.	Stephen S. Intille,James W. Davis,Aaron F. Bobick	1997	A real-time tracking algorithm that uses contextual information is described. The method is capable of simultaneously tracking multiple, non-rigid objects when erratic movement and object collisions are common. A closed-world assumption is used to adaptively select and weight image features used for correspondence. Results of algorithm testing and the limitations of the method are discussed. The algorithm has been used to track children in an interactive, narrative playspace.
CVPR	Euclidean Reconstruction from Image Sequences with Varying and Unknown Focal Length and Principal Point.	Anders Heyden,Kalle Åström	1997	In this paper the special case of reconstruction from image sequences taken by cameras with skew equal to 0 and aspect ratio equal to has been treated. These type of cameras, here called cameras with Euclidean image planes, represent rigid projections where neither the principal point nor the focal length is known. It will be shown that it is possible to reconstruct an unknown object from images taken by a camera with Euclidean image plane up to similarity transformations, i.e., Euclidean transformations plus changes in the global scale. An algorithm, using bundle adjustment techniques, has been implemented. The performance of the algorithm is shown on simulated data.
CVPR	Real-Time Estimation of Human Body Posture from Monocular Thermal Images.	Shoichiro Iwasawa,Kazuyuki Ebihara,Jun Ohya,Shigeo Morishima	1997	This paper introduces a new real-time method to estimate the posture of a human from thermal images acquired by an infrared camera regardless of the background and lighting conditions. Distance transformation is performed for the human body area extracted from the thresholded thermal image for the calculation of the center of gravity. After the orientation of the upper half of the body is obtained by calculating the moment of inertia, significant points such as the top of the head, the tips of the hands and foot are heuristically located. In addition, the elbow and knee positions are estimated from the detected (significant) points using a genetic algorithm based learning procedure. The experimental results demonstrate the robustness of the proposed algorithm and real-time (faster than 20 frames per second) performance.
CVPR	Calibration of a Structured Light System: A Projective Approach.	Du Q. Huynh	1997	We present in this paper a novel calibration method that uses cross ratio to compute world points falling onto any given light stripe plane of a structured light system. We show that, by using 4 known non-coplanar sets of 3 collinear world points, the direct 4 x 3 image to-world transformation matrix for each light stripe plane can also be recovered from plane-to-plane homography. Preliminary experiments conducted with a calibration target and a mannequin suggest that this novel calibration method is robust and is applicable to many shape measurement tasks.
CVPR	Linear Fitting with Missing Data: Applications to Structure-from-Motion and to Characterizing Intensity Images.	David W. Jacobs	1997	Several vision problems can be reduced to the problem of fitting a linear surface of low dimension to data, including the problems of structure-from-affine-motion, and of characterizing the intensity images of a Lambertian scene by constructing the {\em intensity manifold}. For these problems, one must deal with a data matrix with some missing elements. In structure-from-motion, missing elements will occur if some point features are not visible in some frames. To construct the intensity manifold missing matrix elements will arise when the surface normals of some scene points do not face the light source in some images. We propose a novel method for fitting a low rank matrix to a matrix with missing elements. We show experimentally that our method produces good results in the presence of noise. These results can be either used directly, or can serve as an excellent starting point for an iterative method.
CVPR	3-D to 2-D Recognition with Regions.	David W. Jacobs,Ronen Basri	1997	This paper presents a novel approach to parts-based object recognition in the presence of occlusion. We focus on the problem of determining the pose of a 3-D object from a single 2-D image when convex parts of the object have been matched to corresponding regions in the image. We consider three types of occlusions: self-occlusion, occlusions whose locus is identified in the image, and completely arbitrary occlusions. We derive efficient algorithms for the first two cases, and characterize their performance. For the last case, we prove that the problem of finding valid poses is computationally hard, but provide an efficient, approximate algorithm. This work generalizes our previous work on region-based object recognition, which focused on the case of planar models.
CVPR	Image Based View Synthesis of Articulated Agents.	Martin Jägersand	1997	Using a combination of techniques from visual representations, view synthesis, and visual-motor model estimation, we present a method for animating movements of an articulated agent (e.g. human or robot arm), without the use of any prior models or explicit 3D information. The information needed to generate simulated images can be acquired either on or off line, by watching the agent doing an arbitrary, possibly unrelated task.We present experimental results synthesizing image sequences of the simulated movement of a human arm and a PUMA 760 robot arm. Control is in either image (camera), motor (joint), or Cartesian world coordinates. We have created a user interface, where a user can input a movement program, and then upon execution, view movies of the (simulated) agent executing the program, along with the instantaneous values of the dynamics variables
CVPR	Building Reconstruction from Optical and Range Images.	Christopher O. Jaynes,Allen R. Hanson,Edward M. Riseman,Howard J. Schultz	1997	A technique is introduced for extracting and reconstructing a wide class of building types from a registered range image and optical image. An attentional focus stage, followed by model indexing, allows top-down robust surface fittting to reconstruct the 3D nature of the buildings in the data. Because of the effectiveness of model selection, top-down processing of noisy range data still succeeds and the algorithm is capable of detecting and reconstructing several different building roof classes, including fiat single level, fiat multi-leveled, peaked, and curued rooftops. The algorithm is applicable to range data that may have been collected from several different range sensor types. We demonstrate reconstructions of different buildings classes in the presence of large amounts of noise. Our results underline the usefuless of range data when processed in the context of a focus-of-attention area derived from the monocular optical image.
CVPR	Parametrized structure from motion for 3D adaptive feedback tracking of faces.	Tony Jebara,Alex Pentland	1997	"A real-time system is described for automatically detecting, modeling and tracking faces in 3D. A closed loop approach is proposed which utilizes structure from motion to generate a 3D model of a face and then feed back the estimated structure to constrain feature tracking in the next frame. The system initializes by using skin classification, symmetry operations, 3D warping and eigenfaces to find a face. Feature trajectories are then computed by SSD or correlation-based tracking. The trajectories are simultaneously processed by an extended Kalman filter to stably recover 3D structure, camera geometry and facial pose. Adaptively weighted estimation is used in this filter by modeling the noise characteristics of the 2D image patch tracking technique. In addition, the structural estimate is constrained by using parametrized models of facial structure (eigen-heads). The Kalman filter's estimate of the 3D state and motion of the face predicts the trajectory of the features which constrains the search space for the next frame in the video sequence. The feature tracking and Kalman filtering closed loop system operates at 25 Hz."
CVPR	Object Recognition Using Appearance-Based Parts and Relations.	Chien-Yuan Huang,Octavia I. Camps,Tapas Kanungo	1997	The recognition of general three-dimensional objects in cluttered scenes is a challenging problem. In particular, the design of a good representation suitable to model large numbers of generic objects that is also robust to occlusion has been an stumbling block in achieving success. In this paper, we propose a representation using appearance-based parts and relations to overcome these problems. Appearance-based parts and relations are defined in terms of closed regions and the union of these regions, respectively. The regions are segmented using the MDL principle, and their appearance is obtained from collection of images and compactly represented by parametric manifolds in the two eigenspaces spanned by the parts and the relations.
CVPR	Image Indexing Using Color Correlograms.	Jing Huang,Ravi Kumar,Mandar Mitra,Wei-Jing Zhu,Ramin Zabih	1997	We define a new image feature called the color correlogram and use it for image indexing and comparison. This feature distills the spatial correlation of colors, and is both effective and inexpensive for content-based image retrieval. The correlogram robustly tolerates large changes in appearance and shape caused by changes in viewing positions, camera zooms, etc. Experimental evidence suggests that this new feature outperforms not only the traditional color histogram method but also the recently proposed histogram refinement methods for image indexing/retrieval.
CVPR	Recognizing Objects by Matching Oriented Points.	Andrew Edie Johnson,Martial Hebert	1997	We present an approach to recognition of complex objects in cluttered 3-D scenes that does not require feature extraction or segmentation. Our object representation comprises descriptive images associated with each oriented point on the surface of an object. Using a single point basis constructed from an oriented point, the position of other points on the surface of the object can be described by two parameters. The accumulation of these parameters for many points on the surface of the object results in an image at each oriented point. These images, localized descriptions of the global shape of the object, are invariant to rigid transformations. Through correlation of images, point correspondences between a model and scene data are established and then grouped using geometric consistency. The effectiveness of our algorithm is demonstrated with results showing recognition of complex objects in cluttered scenes with occlusion.
CVPR	Analysis of Gesture and Action in Technical Talks for Video Indexing.	Shanon X. Ju,Michael J. Black,Scott L. Minneman,Don Kimber	1997	"In this paper, we present an automatic system for analyzing and annotating video sequences of technical talks. Our method uses a robust motion estimation technique to detect key frames and segment the video sequence into subsequences containing a single overhead slide. The subsequences are stabilized to remove motion that occurs when the speaker adjusts their slides. Any changes remaining between frames in the stabilized sequences may be due to speaker gestures such as pointing or writing and we use active contours to automatically track these potential gestures. Given the constrained domain we define a simple ``vocabulary'' of actions which can easily be recognized based on the active contour shape and motion. The recognized actions provide a rich annotation of the sequence that can be used to access a condensed version of the talk from a web page."
CVPR	Matching 3-D arcs.	Behrooz Kamgar-Parsi,Behzad Kamgar-Parsi	1997	Matching 3-D arcs.
CVPR	Characterization of errors in compositing panoramic images.	Sing Bing Kang,Richard S. Weiss	1997	In this paper, we describe the effect of errors in the intrinsic camera parameters on reconstructed panoramic images. A panoramic image is created by first capturing a sequence of images while rotating the camera about a vertical axis a full 360 degrees. The subsequent steps are projecting the original rectilinear images onto cylindrical surfaces and compositing them to form the panoramic image.Our analysis has led to a technique that allows simultaneous recovery of the camera focal length and properly composited panoramic images. The correct focal length can be determined by iterating the processes of projecting the original rectilinear images onto cylindrical surfaces given an estimate of the focal length and compositing the resulting images to yield an increasingly better estimate of the focal length. This paper shows that the convergence towards the correct focal length is exponential.
CVPR	Content-based trademark retrieval system using visually salient features.	Yong-Sung Kim,Whoi-Yul Kim	1997	An ever increasing number of registered trademarks has created greater demand for an automatic trademark retrieval system. We present a method for such a system based on the image content using shape features. Zernike or pseudo-Zernike moments of the image are employed as a feature set. To retrieve similar shapes, we take into account visually salient features that dominantly affect the global shape of the trademarks and ignore their minor detail. Experimental results on a database of 3,000 trademark images demonstrate that the proposed method retrieves visually similar trademarks which agree well with human perception.
CVPR	Images as embedding maps and minimal surfaces: movies, color, and volumetric medical images.	Ron Kimmel,Ravi Malladi,Nir A. Sochen	1997	"A general geometrical framework for image processing is presented. We consider intensity images as surfaces in the (x, I) space. The image is thereby a two dimensional surface in three dimensional space for gray level images. The new formulation unifies many classical schemes, algorithms, and measures via choices of parameters in ""master"" geometrical measure. More important, it is a simple and efficient tool for the design of natural schemes for image enhancement, segmentation, and scale space. Here we give the basic motivation and apply the scheme to enhance images. We present the concept of an image as a surface in dimensions higher than the three dimensional intuitive space. This will help us handle movies, color, and volumetric medical images."
CVPR	Facial Expression Recognition and Its Degree Estimatio.	Satoshi Kimura,Masahiko Yachida	1997	The purpose of this study is not only to recognize some kind of facial expressions which is associated with human emotion but also to estimate its degree. Our method is based on the idea that facial expression recognition can be achieved by extracting a variation from expressionless face with considering face area as a whole pattern. For the purpose of extracting subtle changes in the face such as the degree of expressions, it is necessary to eliminate the individuality appearing in the facial image. Using a elastic net model, a variation of facial expression is represented as motion vectors of the deformed Net from a facial edge image. Then, applying K-L expansion, the change of facial expression represented as the motion vectors of nodes is mapped into low dimensional eigen space, and estimation is achieved by projecting input images on to the Emotion Space. In this paper we have constructed three kinds of expression models: happiness, anger, surprise, and experimental results are evaluated.
CVPR	Interaction with On-Screen Objects using Visual Gesture Recognition.	Rick Kjeldsen,John R. Kender	1997	"This paper will review the design of a working system that visually recognizes hand gestures for the control of a window based user interface. After an overview of the system, it will explore one aspect of gestural interaction in depth, hand tracking, and what is needed for the user to be able to interact comfortably with on-screen objects. We describe how the location of the hand is mapped to a location on the screen, and how it is both necessary and possible to smooth the camera input using a non-linear physical model of the cursor. The performance of the system is examined, especially with respect to object selection. We show how a standard HCI model of object selection (Fitts' Law) can be extended to model the selection performance of free-hand pointing."
CVPR	Non-linear operators in image restoration.	Pierre Kornprobst,Rachid Deriche,Gilles Aubert	1997	We firstly present a variational approach such that during image restoration, edges detected in the original image are being preserved, and then we compare in a second part, the mathematical foundation of this method with respect to some of the well known methods recently proposed in the literature within the class of PDE based algorithms (anisotropic diffusion, mean curvature motion, min/max flow technique,... ). The performance of our approach is carefully examined and compared to the classical methods. Experimental results on synthetic and real images will illustrate the capabilities of all the studied approaches.
CVPR	Object detection with vector quantized binary features.	John Krumm	1997	"This paper presents a new algorithm for detecting objects in images, one of the fundamental tasks of computer vision. The algorithm extends the representational efficiency of eigenimage methods to binary features, which are less sensitive to illumination changes than gray-level values normally used with eigenimages. Binary features (square subtemplates) are automatically chosen on each training image. Using features rather than whole templates makes the algorithm more robust to background clutter and partial occlusions. Instead of representing the features with real-valued eigenvector principle components, we use binary vector quantization to avoid floating point computations. The object is detected in the image using a simple geometric hash table and Hough transform. On a test of 1000 images, the algorithm works on 99.3%. We present a theoretical analysis of the algorithm in terms of the receiver operating characteristic, which consists of the probabilities of detection and false alarm. We verify this analysis with the results of our 1000-image test and we use the analysis as a principled way to select some of the algorithm's important operating parameters."
CVPR	An image-based visual-motion-cue for autonomous navigatio.	Sridhar R. Kundur,Daniel Raviv,Ernest Kent	1997	This paper presents a novel time-based visual motion cue called the Hybrid Visual Threat Cue (HVTC) that provides some measure for a change in relative range as well as absolute clearances, between a 3D surface and a moving observer. It is shown that the HVTC is a linear combination of Time-To-Contact (TTC), visual looming and the Visual Threat Cue (VTC). The visual field associated with the HVTC can be used to demarcate the regions around a moving observer into safe and danger zones of varying degree, which may be suitable for autonomous navigation tasks. The HVTC is independent of the 3D environment and needs almost no a-priori information about it. It is rotation independent, and is measured in ~time/sup -1/\ units Several approaches to extract the HVTC, are suggested. Also a practical method to extract it from a sequence of images of a 3D textured surface obtained by a visually fixating, fixed-focus monocular camera in motion is presented. This approach of extracting the HVTC is independent of the type of 3D surface texture and needs no optical flow information, 3D reconstruction, segmentation, feature tracking.
CVPR	Shape from the Light Field Boundary.	Kiriakos N. Kutulakos	1997	Ray-based representations of shape have received little attention in computer vision. In this paper we show that the problem of recovering shape from silhouettes becomes considerably simplified if it is formulated as a reconstruction problem in the space of oriented rays that intersect the object. The method can be used with both calibrated and uncalibrated cameras, does not rely on point correspondences to compute shape, and does not impose restrictions on object topology or smoothness.
CVPR	What is a light source?	Michael S. Langer,Steven W. Zucker	1997	Traditional light source modelling is concerned with specific types of light sources, the two most common of which are point sources and daylight. Little attempt has been made, however, to relate different types of sources to each other. For example, how may the lighting from an overcast sky be compared to that from a lamp? Having a theoretical framework to compare different types of light sources is important for computer vision, in particular for understanding shading and shadow cues. A vision system needs to take account of the light source in order to interpret these cues. In this paper, we present a framework for comparing types of light sources which is based on a dimensional analysis of the set of light rays in a free space. Specifically, we introduce a 4-D light source hypercube in which the different types of sources may be embedded and compared. We also present a novel definition for light sources which generalizes the standard definition of a source as an emitter.
CVPR	Using Differential Constraints to Reconstruct Complex Surfaces from Stereo.	Richard Lengagne,Olivier Monga,Pascal Fua	1997	Stereo reconstruction algorithms often fail to properly deal with complex surfaces, because there is not enough image information. To overcome this problem, we propose to guide the reconstruction process using a priori information about the differential geometry of the object surfaces. We use both linear structures such as crest lines or scalar fields such as curvature values to generate a reconstruction of the surface which is consistent with the differential properties. This method improves the accuracy of the reconstruction around the discontinuities and increases the compactness of the surface representation.
CVPR	On Perpendicular Texture: Why do we see more flowers in the distance?	Thomas K. Leung,Jitendra Malik	1997	Almost all work on texture in the computer vision and graphics communities has modeled the texture as tangential, i.e. lying in the tangent plane to the surface. This is equivalent to thinking of the texture as a pattern painted on the surface. Three-dimensional textures, where the elements may point out of the surface, have largely been ignored. We study a special class of 3D textures, perpendicular textures where we can model the elements as being normal to the surface. The perspective projection of perpendicularly textured surfaces results in several interesting phenomena, which do not occur in the much-studied tangential texture case. These include occlusion, foreshortening and illumination.In this paper, we study the geometry of the problem, modeling the locations of the elements of the texture as being a realization of a spatial point process. Relations between slant and tilt of the surface, density and height of elements and occlusions are derived. Occlusions can now be used as a cue to infer shape, instead of being treated as a source of error.
CVPR	Dynamic Appearance-Based Recognition.	Rajesh P. N. Rao	1997	"We describe a hierarchical appearance-based method for learning, recognizing, and predicting arbitrary spatiotemporal sequences of images. The method, which implements a robust hierarchical form of the Kalman filter derived from the Minimum Description Length (MDL) principle, includes as a special case several well-known object encoding techniques including eigenspace methods for static recognition. Successive levels of the hierarchical filter implement dynamic models operating over successively larger spatial and temporal scales. Each hierarchical level predicts the recognition state at a lower level and modifies its own recognition state using the residual error between the prediction and the actual lower-level state. Simultaneously, on a longer time scale, the filter learns an internal model of input dynamics by adapting its generative and state transition matrices at each level to minimize prediction errors. The resulting prediction/learning scheme thereby implements an on-line form of the well-known Expectation-Maximization (EM) algorithm from statistics. We present experimental results demonstrating the method's efficacy in mediating robust spatiotemporal recognition in a variety of scenarios containing varying degrees of occlusions and clutter."
CVPR	Using Chromaticity Distributions and Eigenspace Analysis for Pose, Illumination, and Specularity-Invariant Recognition of 3D Objects.	Stephen Lin,Sang Wook Lee	1997	The distribution of object colors can be effectively utilized for recognition and indexing. Difficulties arise in the recognition of object color distributions when there are variations in illumination color, changes in object pose with respect to illumination direction, and specular reflections. However, most of the recent approaches to color-based recognition focus mainly on illumination color invariance. We propose an approach that identifies object color distributions influenced by: (1) illumination pose, (2) illumination color and (3) specularity. We suggest the use of chromaticity distributions to achieve illumination pose invariance. To characterize changes in chromaticity distribution due to illumination color, a set of chromaticity histograms of each object is generated for a range of lighting colors based on linear models of illumination and reflectance, and the histograms are represented using a small number of eigen basis vectors constructed from principal components analysis. Since specular reflections may alter the chromaticity distributions of test objects, a model-based specularity detection/rejection algorithm, called chromaticity differencing, is developed to reduce these effects.
CVPR	Edge Flow: A Framework of Boundary Detection and Image Segmentation.	Wei-Ying Ma,B. S. Manjunath	1997	A novel boundary detection scheme based on edgejow is proposed in this paper. This scheme utilizes a predictive coding model to identify the direction of change in color and texture at each image location at a given scale, and constructs an edge flow vector: By iteratively propagating the edge flow, the boundaries can be detected at image locations which encounter two opposite directions offrow in the stable state. A user defined image scale is the only significant control parameter that is needed by the algorithm. The scheme facilitates integration of color and texture into a single framework for boundary detection.
CVPR	Configuration based scene classification and image indexing.	Pamela Lipson,Eric Grimson,Pawan Sinha	1997	"Scene classification is a major open challenge in machine vision. Most solutions proposed so far such as those based on color histograms and local texture statistics cannot capture a scene's global configuration, which is critical in perceptual judgments of scene similarity. We present a novel approach, ""configural recognition"", for encoding scene class structure. The approach's main feature is its use of qualitative spatial and photometric relationships within and across regions in low resolution images. The emphasis on qualitative measures leads to enhanced generalization abilities and the use of low-resolution images renders the scheme computationally efficient. We present results on a large database of natural scenes. We also describe how qualitative scene concepts may be learned from examples."
CVPR	Object tracking using affine structure for point correspondences.	Gurmeet Singh Manku,Pankaj Jain,Amit Aggarwal,Lalit Kumar,Subhashis Banerjee	1997	A new object tracking algorithm based on affine structure has been developed and it is shown that its performance is better than that of a Kalman filter based correlation tracker. The algorithm is fast, reliable, viewpoint invariant, and insensitive to occlusion and/or individual corner disappearance or reappearance. Detailed experimental analysis on a long real image sequence is also presented.
CVPR	A class of probabilistic shape models.	Jorge S. Marques,Arnaldo J. Abrantes	1997	Deformable models are related to other data representation methods. It was recently proposed a class of models based on a fuzzy energy function which includes many well known algorithms (snakes, elastic nets, fuzzy and hard c-means and Kohonen maps). This paper describes a probabilistic extension of these algorithms in a Bayesian framework, using Gibbs-Boltzman distributions. It is shown that the new class of models minimizes an energy function with an additional term: the log partition function. The role of the log partition function in probabilistic versions of snakes, c-means and elastic nets is studied and analytic expressions are derived in the case of probabilistic snakes. The log partition function produces an additional force field which improves the performance of these algorithms in some applications.
CVPR	Edge Localization in Surface Reconstruction Using Optimal Estimation Theory.	Shailendra Mathur,Frank P. Ferrie	1997	"Many relaxation based smoothing methods used in surface reconstruction algorithms filter out the effect of noise in image data, but result in the elimination of important discontinuity information as well. In this paper the inter-pixel interaction during relaxation is shown to be equivalent to a multiple measurement fusion problem which can be solved using optimal estimation theory. Pixels in a given neighbourhood act as noisy information sources, combining their information to update the state of that neighbourhood. By formulating discontinuities as another ``noise'' source in the image, and by using the so-called Curvature Consistency reconstruction algorithm on range images, it is shown that optimal estimation theory offers a method for the automatic and adaptive localization of discontinuities while providing a smooth piece-wise continuous surface description."
CVPR	A Real-time Computer Vision System for Measuring Traffic Parameters.	Philip F. McLauchlan,David Beymer,Benjamin Coifman,Jitendra Malik	1997	For the problem of tracking vehicles on freeways using machine vision, existing systems work well in free-flowing traffic. Traffic engineers, however, are more interested in monitoring freeways when there is congestion, and current systems break down for congested traffic due to the problem of partial occlusion. We have developed a feature-based tracking approach for the task of tracking vehicles under congestion. Instead of tracking entire vehicles, vehicle sub-features are tracked to make the system robust to partial occlusion. In order to group together sub-features that come from the same vehicle, the constraint of common motion is used. Here we describe the real-time implementation of the system using a network of DSP chips.
CVPR	Prediction Intervals for Surface Growing Range Segmentation.	James V. Miller,Charles V. Stewart	1997	The surface growing framework presented by Besl and Jain has served as the basis for many range segmentation techniques. It has been augmented with alternative fitting techniques, model selection criteria, and solid modelling components. All of these approaches, however, require global thresholds and large isolated seed regions. Range scenes typically do not satisfy the global threshold assumption since it requires data noise characteristics to be constant throughout the scene. Furthermore, as scene complexity increases, the number of surfaces, discontinuities, and outliers increase, hindering the identification of large seed regions.We present statistical criteria based on multivariate regression to replace the traditional decision criteria used in surface growing. We use local estimates and their uncertainties to construct criteria which capture the uncertainty in extrapolating estimated fits. We restrict surface expansion to very localized extrapolations, increasing the sensitivity to discontinuities and allowing regions to refine their estimates and uncertainties. Our approach uses a small number of parameters which are either statistical thresholds or cardinality measures, i.e. we do not use thresholds defined by specific range distances or orientation angles.
CVPR	A Hybrid Framework for Surface Registration and Deformable Models.	Johan Montagnat,Hervé Delingette	1997	In computer vision, two complementary approaches have been widely used to perform object reconstruction and registration. The deformable model framework locally applies internal and external forces to fit 3D data. The non-rigid registration framework iteratively computes the best global transformation in order to minimize the distance between a template and the data. In this paper, we first show that applying a global transformation on a surface model, is equivalent to applying an external force on a deformable model without any regularizing force. Second we propose a hybrid framework which combines the registration framework and the deformable models scheme. Our hybrid deformation approach allows to control the scale at which the model is deformed. This is clearly beneficial for performing both reconstruction and registration tasks. We show many examples of this approach on active contours and deformable surfaces. Furthermore, a global transformation based on axial symmetry is introduced.
CVPR	Fast 3D Stabilization and Mosaic Construction.	Carlos Morimoto,Rama Chellappa	1997	We present a fast electronic image stabilization system that compensates for 3D rotation. The extended Kalman filter framework is employed to estimate the rotation between frames, which is represented using unit quaternions. A small set of automatically selected and tracked feature points are used as measurements. The effectiveness of this technique is also demonstrated by constructing mosaic images from the motion estimates, and comparing them to mosaics built from 2D stabilization algorithms. Two different stabilization schemes are presented. The first, implemented in a real-time platform based on a Datacube MV200 board, estimates the motion between two consecutive frames and is able to process gray level images of resolution 128x120 at 10 Hz. The second scheme estimates the motion between the current frame and an inverse mosaic; this allows better estimation without the need for indexing the new image frames. Experimental results for both schemes using real and synthetic image sequences are presented.
CVPR	Registering Multiple Cartographic Models with the Hierarchical Mixture of Experts Algorithm.	Simon Moss,Edwin R. Hancock	1997	This paper describes an application of the hierarchical mixture of experts algorithm to the registration of multiple cartographic models to noisy radar data According to the HME algorithm each model is represented by a set of maximum likelihood registration parameters together with a set of matching probabilities. This architecture can be viewed as providing simultaneous registration and hypothesis verification. The maps in the cartographic data-base compete to account for radar data through the imposed probability normalization. The resulting matching algorithm can be regarded as a generic tool for model retrieval from a data-base. Our evaluation on radar images illustrates some of the characteristics of the algorithm. Our main conclusions are that the method is both robust to added image noise and poor initialization.
CVPR	Detection and Description of Buildings from Multiple Aerial Images.	Sanjay Noronha,Ramakant Nevatia	1997	A method for detection and description of rectangular buildings from two or more registered aerial intensity images is proposed. The output is a 3D description of the buildings, with an associated confidence measure for each building. Hierarchical perceptual grouping and matching across views is employed to increase the robustness of the system. Verification of selected building hypotheses is done using shadow and wall evidence of the buildings. The system is largely feature-based. Grouping and matching are performed in a hierarchical manner, utilizing primitives of increasing complexity, starting with line segments and junctions, and proceeding to higher level features. Binocular and trinocular epipolar constraints are used to reduce the search space for matching features.
CVPR	Catadioptric Omnidirectional Camera.	Shree K. Nayar	1997	Conventional video cameras have limited fields of view that make them restrictive in a variety of vision applications. There are several ways to enhance the field of view of an imaging system. However, the entire imaging system must have a single effective viewpoint to enable the generation of pure perspective images from a sensed image. A new camera with a hemispherical field of view is presented. Two such cameras can be placed back-to-back, without violating the single viewpoint constraint, to arrive at a truly omnidirectional sensor. Results are presented on the software generation of pure perspective images from an omnidirectional image, given any user-selected viewing direction and magnification. The paper concludes with a discussion on the spatial resolution of the proposed camera.
CVPR	LAFTER: Lips and Face Real-Time Tracker.	Nuria Oliver,Alex Pentland,François Bérard	1997	This paper describes an active-camera real-time system for tracking, shape description, and classification of the human face and mouth using only an SGI Indy computer. The system is based on use of 2-D blob features, which are spatially-compact clusters of pixels that are similar in terms of low-level image properties. Patterns of behavior (e.g., facial expressions and head movements) can be classified in real-time using Hidden Markov Model (HMM) methods. The system has been tested on hundreds of users and has demonstrated extremely reliable and accurate performance. Typical classification accuracies are near 100 %
CVPR	Pedestrian Detection Using Wavelet Templates.	Michael Oren,Constantine Papageorgiou,Pawan Sinha,Edgar Osuna,Tomaso Poggio	1997	This paper presents a trainable object detection architecture that is applied to detecting people in static images of cluttered scenes. This problem poses several challenges. People are highly non-rigid objects with a high degree of variability in size, shape, color, and texture. Unlike previous approaches, this system learns from examples and does not rely on any a priori (handcrafted) models or on motion. The detection technique is based on the novel idea of the wavelet template that defines the shape of an object in terms of a subset of the wavelet coeficients of the image. It is invariant to changes in color and texture and can be used to robustly define a rich and complex class of objects such as people. We show how the invariant properties and computational eficiency of the wavelet template make it an effective tool for object detection.
CVPR	Training Support Vector Machines: an Application to Face Detection.	Edgar Osuna,Robert Freund,Federico Girosi	1997	"We investigate the application of Support Vector Machines (SVMs) in computer vision. SVM is a learning technique developed by V. Vapnik and his team (AT\&T Bell Labs.) that can be seen as a new method for training polynomial, neural network, or Radial Basis Functions classifiers. The decision surfaces are found by solving a linearly constrained quadratic programming problem. This optimization problem is challenging because the quadratic form is completely dense and the memory requirements grow with the square of the number of data points.We present a decomposition algorithm that guarantees global optimality, and can be used to train SVM's over very large data sets. The main idea behind the decomposition is the iterative solution of sub-problems and the evaluation of optimality conditions which are used both to generate improved iterative values, and also establish the stopping criteria for the algorithm.We present experimental results of our implementation of SVM, and demonstrate the feasibility of our approach on a face detection problem that involves a data set of 50,000 data points."
CVPR	Image velocity estimation from trajectory surface in spatiotemporal space.	Kazuhiro Otsuka,Tsutomu Horikoshi,Satoshi Suzuki	1997	A new framework and method, based on image motion trajectories in spatiotemporal space (x-y-t space), are proposed to estimate image velocity from an image sequence. We focus on the surfaces of the trajectories in the x-y-t space formed by the edges and contours of moving objects and obtain image velocity from the orientation of the intersection line formed by tangent planes on the trajectories. The proposed method includes two Hough transforms to detect the most dominant orientation in all possible intersection lines and reliably produces the dominant translational image velocity semi-locally. Also, the confidence measure of estimates is defined to decide the optimal size of patch that suppresses the aperture problem. Experimental results from several synthetic and real image sequences are presented to verify the effectiveness of the method and to confirm its robustness against noise and occlusion.
CVPR	Non-parametric Similarity Measures for Unsupervised Texture Segmentation and Image Retrieval.	Jan Puzicha,Thomas Hofmann,Joachim M. Buhmann	1997	In this paper we propose and examine non-parametric statistical tests to define similarity and homogeneity measures for textures. The statistical tests are applied to the coefficients of images filtered by a multi-scale Gabor filter bank. We will demonstrate that these similarity measures are useful for both, texture based image retrieval and for unsupervised texture segmentation, and hence offer an unified approach to these closely related tasks. We present results on Brodatz-like micro-textures and a collection of real-word images.
CVPR	Object Detection Using Hierarchical MRF and MAP Estimation.	Richard J. Qian,Thomas S. Huang	1997	This paper presents a new scale, position and orientation invariant approach to object detection. The proposed method first chooses attention regions in an image based on the region detection result on the image. Within the attention regions, the method then detects targets using a novel object detection algorithm that combines template matching methods with feature-based methods via hierarchical MRF and MAP estimation. Hierarchical MRF and MAP estimation provide a flexible framework to incorporate various visual clues. The combination of template matching and feature detection helps to achieve robustness against complex backgrounds and partial occlusions in object detection. Experimental results are given in the paper.
CVPR	Uncalibrated 1D projective camera and 3D affine reconstruction of lines.	Long Quan	1997	"We describe a linear algorithm to recover 3D affine shape/motion from line correspondences over three views with uncalibrated affine cameras. The key idea is the introduction of a one-dimensional projective camera. This converts the 3D affine reconstruction of ""lines"" into 2D projective reconstruction of ""points"". Using the full tensorial representation of three uncalibrated 1D views, we prove that the 3D affine reconstruction of lines from minimal data is unique up to a re-ordering of the views. 3D affine line reconstruction can be performed by properly rescaling image coordinates instead of using projection matrices. The algorithm is validated on both simulated and real image sequences."
CVPR	Panoramic mosaics by manifold projection.	Shmuel Peleg,Joshua Herman	1997	"As the field of view of a picture is much smaller than our own visual field of view, it is common to paste together several pictures to create a panoramic mosaic having a larger field of view. Images with a wider field of view can be generated by using fish-eye lens, or panoramic mosaics can be created by special devices which rotate around the camera's optical center (Quicktime VR, Surround Video), or by aligning, and pasting, frames in a video sequence to a single reference frame. Existing mosaicing methods have strong limitations on imaging conditions, and distortions are common. Manifold projection enables the creation of panoramic mosaics from video sequences under more general conditions, and in particular the unrestricted motion of a hand-held camera. The panoramic mosaic is a projection of the scene into a virtual manifold whose structure depends on the camera's motion. This manifold is more general than the customary projections onto a single image plane or onto a cylinder. In addition to being more general than traditional mosaics, manifold projection is also computationally efficient, as the only image deformations used are image plane translations and rotations. Real-time, software only, implementation on a Pentium-PC, proves the superior quality and speed of this approach."
CVPR	Orientation diffusions.	Pietro Perona	1997	Diffusions provide a convenient way of smoothing noisy brightness images, of analyzing images at multiple scales, and of enhancing discontinuities. Some quantities of interest in computer vision are defined on curved manifolds; typical examples are orientation and hue that are defined on the circle. Generalizing diffusions to orientation is not straightforward, especially in the case where a discrete implementation is sought. An example of what may go wrong is presented. A method is proposed to define diffusions of orientation-like quantities. First a definition in the continuum is discussed, then a discrete orientation diffusion is proposed. The behavior of such diffusions is explored both analytically and experimentally. It is shown how such orientation diffusions contain a nonlinearity that is reminiscent of edge-process and anisotropic diffusion. A number of open questions are proposed at the end.
CVPR	Optimal Selection of Camera Parameters for Recovery of Depth from Defocused Images.	A. N. Rajagopalan,Subhasis Chaudhuri	1997	In the depth from defocus (DFD) method, two defocused images of a scene are obtained by capturing the scene with different sets of camera parameters. An arbitrary selection of the camera settings can result in observed images whose relative blurring is insufficient to yield a good estimate of the depth. In this paper, we study the effect of the degree of relative blurring on the accuracy of the estimate of the depth by addressing the DFD problem in a maximum likelihood-based framework. We propose a criterion for optimal selection of camera parameters to obtain an improved estimate of the depth. The optimality criterion is based on the Cramer-Rao bound of the variance of the error in the estimate of blur. Simulations as well as experimental results on real images are presented for validation.
CVPR	The FERET Evaluation Methodology for Face-Recognition Algorithms.	P. Jonathon Phillips,Hyeonjoon Moon,Patrick J. Rauss,Syed A. Rizvi	1997	Two of the most critical requirements in support of producing reliable face-recognition systems are a large database of facial images and a testing procedure to evaluate systems. The Face Recognition Technology (FERET) program has addressed both issues through the FERET database of facial images and the establishment of the FERET tests. To date, 14,126 images from 1199 individuals are included in the FERET database, which is divided into development and sequestered portions. In September 1996, the FERET program administered the third in a series of FERET face-recognition tests. The primary objectives of the third test were to (1) assess the state of the art, (2) identify future areas of research, and (3) measure algorithm performance on large databases.
CVPR	A direct method for stereo correspondence based on singular value decomposition.	Maurizio Pilu	1997	This paper proposes a new algorithm for matching point features across pairs of images. Despite the well-known combinatorial complexity of the problem, this work shows that an acceptably good solution can be obtained directly by singular value decomposition of an appropriate correspondence strength matrix. The approach draws from the method proposed previously but, besides suggesting its usefulness for stereo matching, in this work a correlation-weighted proximity function is used as correspondence strength to specifically cater for real images.
CVPR	A Stratified Approach to Metric Self-Calibration.	Marc Pollefeys,Luc J. Van Gool	1997	Camera calibration is essential to many computer vision applications. In practice this often requires cumbersome calibration procedures to be carried out regularly. In the last few years a lot of work has been done on self-calibration of cameras, ranging from weak calibration to metric calibration. It has been shown that a metric calibration of the camera setup (up to scale) was possible based on the rigidity of the scene only. In this paper a stratified approach is proposed which gradually retrieves the metric calibration of the camera setup. Starting from an uncalibrated image sequence the projective calibration is retrieved first. In projective space the plane at infinity is then identified yielding the affine calibration. This is achieved using a constraint which can be formulated between any two arbitrary images of the sequence. Once the affine calibration is known the upgrade to metric is easily obtained through linear equations.
CVPR	Automated Model Acquisition from Range Images with View Planning.	Michael K. Reed,Peter K. Allen,Ioannis Stamos	1997	"We present an incremental system that builds accurate CAD models of objects from multiple range images. Using a hybrid of surface mesh and volumetric representations, the system creates a ""water-tight"" 3D model at each step of the modeling process, allowing reasonable models to be built from a small number of views. We also present a method that can be used to plan the next view and reduce the number of scans needed to recover the object. Results are presented for the creation of 3D models of a computer game controller, a hip joint prosthesis, and a mechanical strut."
CVPR	Vision for a smart kiosk.	James M. Rehg,Maria Loughlin,Keith Waters	1997	We describe a novel computer vision application: vision-based human sensing for a Smart Kiosk interface. A Smart Kiosk is a free-standing information dispensing computer appliance capable of engaging in public interactions with multiple people. Vision sensing is a critical component of the kiosk interface, where it is used to determine the context for the interaction. We present a taxonomy of vision problems for a kiosk interface and describe a prototype kiosk which uses color stereo tracking and graphical output to interact with several users.
CVPR	Fast Binary Image Processing Using Binary Decision Diagrams.	Luc Robert,Grégoire Malandain	1997	Many classical image processing tasks can be realized as evaluations of a boolean function over subsets of an image. For instance, the simplicity test used in 3D thinning requires examining the 26 neighbors of each voxel and computing a single boolean function of these inputs. In this article, we show how Binary Decision Diagrams can be used to produce automatically very efficient and compact code for such functions. The total number of operations performed by a generated function is at most one test and one branching for each input value (e.g., in the case of 3D thinning, 26 tests and branchings). At each stage, the function is guaranteed to examine only the pertinent input data, i.e., the values which affect the result. As an example, we consider the 3D simplicity test in digital topology, and thinning processes. We produce functions much faster than our previously optimized implementations and than any other implementation we know of. In the case of 3D simplicity test, on average, at each voxel only 8.7 neighboring voxel values are examined.
CVPR	A general filter for measurements with any probability distribution.	Yoav Rosenberg,Michael Werman	1997	"The Kalman filter is a very efficient optimal filter, however it has the precondition that the noises of the process and of the measurement are Gaussian. The authors introduce 'the general distribution filter' which is an optimal filter that can be used even where the distributions are not Gaussian. An efficient practical implementation of the filter is possible where the distributions are discrete and compact or can be approximated as such."
CVPR	Velocity and Disparity Cues for Robust Real-Time Binocular Tracking.	Sebastien Rougeaux,Yasuo Kuniyoshi	1997	We have designed and implemented a real-time binocular tracking system which uses two independent cues commonly found in the primary functions of biological visual systems to robustly track moving targets in complex environments, without a-priori knowledge of the target shape or texture: a fast optical flow segmentation algorithm quickly locates independently moving objects for target acquisition and provides a reliable velocity estimate for smooth tracking. In parallel, target position is generated from the output of a zero-disparity filter where a phase-based disparity estimation technique allows dynamic control of the camera vergence to adapt the horopter geometry to the target location. The system takes advantage of the optical properties of our custom-designed foveated wide-angle lenses, which exhibit a wide field of view along with a high resolution fovea. Methods to cope with the distortions introduced by the space-variant resolution, and a robust real-time implementation on a high performance active vision head are presented.
CVPR	Analyzing Articulated Motion Using Expectation-Maximization.	Henry A. Rowley,James M. Rehg	1997	We present a novel application of the Expectation-Maximization algorithm to the global analysis of articulated motion. The approach utilizes a kinematic model to constrain the motion estimates, producing a segmentation of the flow field into parts with different articulated motions. Experiments with synthetic and real images are described.
CVPR	Cylindrical rectification to minimize epipolar distortion.	Sébastien Roy,Jean Meunier,Ingemar J. Cox	1997	We propose anew rectification method for aligning epipolar lines of a pair of stereo images taken under any camera geometry. It effectively remaps both images onto the surface of a cylinder instead of a plane, which is used in common rectification methods. For a large set of camera motions, remapping to a plane has the drawback of creating rectified images that are potentially infinitely large and presents a loss of pixel information along epipolar lines. In contrast, cylindrical rectification guarantees that the rectified images are bounded for all possible camera motions and minimizes the loss of pixel information along epipolar line. The processes (e.g., stereo matching, etc.) subsequently applied to the rectified images are thus more accurate and general since they can accommodate any camera geometry.
CVPR	Efficient Guaranteed Search for Gray-level Patterns.	William Rucklidge	1997	We address the problem of locating a gray-level pattern in a gray-level image. The pattern can have been transformed by an affine transformation, and may have undergone some additional changes. We define a difference function based on comparing each pixel of the pattern with a window in the image, and search efficiently for transformations that minimize the difference function. The search is guaranteed: it will always find the transformation minimizing the difference function, and not get fooled by a local minimum; it is also efficient, in that it does not need to examine every transformation in order to achieve this guarantee. This technique can be applied to object location, motion tracking, optical flow, or block-based motion compensation in video image sequence compression (e.g., MPEG).
CVPR	On Occluding Contour Artifacts in Stereo Vision.	Radim Sára,Ruzena Bajcsy	1997	We study occluding contour artifacts in area-based stereo matching: they are false responses of the matching operator to the occlusion boundary and cause the objects extend beyond their true boundaries in disparity maps. Most of the matching methods suffer from these artifacts; the effect is so strong that it cannot be ignored. We show what gives rise to the artifacts and design a matching criterion that accommodates the presence of occlusions as opposed to methods that identify and remove the artifacts. This approach leads to the problem of measurement contamination studied in statistics. We show that such a problem is hard given finite computational resources, unless more independent measurements directly related to occluding contours is available. What can be achieved is a substantial reduction of the artifacts, especially for large matching templates. Reduced artifacts allow for easier hierarchical matching and for easy fusion of reconstructions from different viewpoints into a coherent whole.
CVPR	Name-It: Association of Face and Name in Video.	"Shin'ichi Satoh,Takeo Kanade"	1997	This paper proposes a novel approach to extract meaningful content information from video by collaborative integration of image understanding and natural language processing. As an actual example, we developed a system that associates faces and names in videos, called Name-It, which is given news videos as a knowledge source, then automatically extracts face and name association as content information. The system can infer the name of a given unknown face image, or guess faces which are likely to have the name given to the system. This paper explains the method with several successful matching results which reveal effectiveness in integrating heterogeneous techniques as well as the importance of real content information extraction from video, especially face-name association.
CVPR	True Multi-Image Alignment and its Application to Mosaicing and Lens Distortion Correction.	Harpreet S. Sawhney,Rakesh Kumar	1997	Multiple images of a scene are related through 2D/3D view transformations and linear and non-linear camera transformations. In the traditional techniques to compute these transformations, especially the ones relying on direct intensity gradients, one image and its coordinate system have been assumed to be ideal and distortion free. In this paper, we present an algorithm for true multi-image alignment that does not rely on the measurements of a reference image being distortion free. The algorithm is developed to specifically align and mosaic images using parametric transformations in the presence of lens distortion. When lens distortion is present, none of the images can be assumed to be ideal. In our formulation, all the images are modeled as intensity measurements represented in their respective coordinate systems, each of which is related to an ideal coordinate system through an interior camera transformation and an exterior view transformation. The goal of the accompanying algorithm is to compute an image in the ideal coordinate system while solving for the transformations that relate the ideal system with each of the data images. Key advantages of the technique presented in this paper are: (i) no reliance on one distortion free image, (ii) ability to register images and compute coordinate transformations even when the multiple images are of an extended scene with no overlap between the first and last frame of the sequence, and (iii) ability to handle linear and non-linear transformations within the same framework. Results of applying the algorithm are presented for the correction of lens distortion, and creation of video mosaics.
CVPR	Automatic line matching across views.	Cordelia Schmid,Andrew Zisserman	1997	The paper presents a new method for matching individual line segments between images. The method uses both grey-level information and the multiple view geometric relations between the images. For image pairs epipolar geometry facilitates the computation of a cross-correlation based matching score for putative line correspondences. For image triplets cross-correlation matching scores are used in conjunction with line transfer based on the trifocal geometry. Algorithms are developed for both short and long range motion. In the case of long range motion the algorithm involves evaluating a one parameter family of plane induced homographies. The algorithms are robust to deficiencies in the line segment extraction and partial occlusion. Experimental results are given for image pairs and triplets, for varying motions between views, and for different scene types. The three view algorithm eliminates all mismatches.
CVPR	Photorealistic Scene Reconstruction by Voxel Coloring.	Steven M. Seitz,Charles R. Dyer	1997	A novel scene reconstruction technique is presented, different from previous approaches in its ability to cope with large changes in visibility and its modeling of intrinsic scene color and texture information. The method avoids image correspondence problems by working in a discretized scene space whose voxels are traversed in a fixed visibility ordering. This strategy takes full account of occlusions and allows the input cameras to be far apart and widely distributed about the environment. The algorithm identifies a special set of invariant voxels which together form a spatial and photometric reconstruction of the scene, fully consistent with the input images. The approach is evaluated with images from both inward-facing and outward-facing cameras.
CVPR	A Bayesian Segmentation Framework for Textured Visual Images.	Shishir Shah,J. K. Aggarwal	1997	This paper presents a new framework for segmentation of textured visual imagery. The proposed method consists of a Bayesian formulation for labeling similar regions. Similarity is defined via texture features obtained by Gabor Wavelets. Multivariate Gaussian distributions are employed to model the feature class-conditional densities, while the Markov process is used to characterize the distributions of the region labeling due to each feature. A coarse nearest neighbor clustering is performed over the feature space to estimate the initial labelings. An iterative solution to the Maximum A Posteriori(MAP) estimation is developed, where the parameters of the prior distribution of region labels are estimated using the Expectation-Maximization(EM) algorithm. Finally, for man-made object segmentation, a region-growing procedure is used to analyze the classified texture regions by incorporating measures of local shape characteristics to obtain smooth boundaries and region homogeneity. Results of the developed algorithm on real scene images are presented.
CVPR	Completion Energies and Scale.	Eitan Sharon,Achi Brandt,Ronen Basri	1997	The detection of smooth curves in images and their completion over gaps are two important problems in perceptual grouping. In this study we examine the notion of completion energy of curve elements, showing and exploiting its intrinsic dependence on length and width scales. We introduce a fast method for computing the most likely completion between two elements, by developing novel analytic approximations and a fast numerical procedure for computing the curve of least energy. We then use our newly developed energies to find the most likely completions in images through a generalized summation of induction fields. This is done through multiscale procedures, i.e., separate processing at different scales with some interscale interactions. Such procedures allow the summation of all induction fields to be done in a total of only O(N logN) operations, where N is the number of pixels in the image. More important, such procedures yield a more realistic dependence of the induction field on the length and width scales: the field of a long element is very different from the sum of the fields of its composing short segments.
CVPR	Normalized Cuts and Image Segmentation.	Jianbo Shi,Jitendra Malik	1997	We propose a novel approach for solving the perceptual grouping problem in vision. Rather than focusing on local features and their consistencies in the image data, our approach aims at extracting the global impression of an image. We treat image segmentation as a graph partitioning problem and propose a novel global criterion, the normalized cut, for segmenting the graph. The normalized cut criterion measures both the total dissimilarity between the different groups as well as the total similarity within the groups. We show that an efficient computational technique based on a generalized eigenvalue problem can be used to optimize this criterion. We have applied this approach to segmenting static images, as well as motion sequences, and found the results to be very encouraging.
CVPR	"Ridge's corner detection and correspondence."	Erez Shilat,Michael Werman,Yoram Gdalyahu	1997	Traditionally, corners are found along step edges. In this paper we present an alternative approach-corners along ridges/troughs and local minima points. These features seem to be more reliable for tracking. A new approach for sub-pixel localization of these corners is suggested, using a local approximation of the image surface.
CVPR	Area and Length Minimizing Flows for Shape Segmentation.	Kaleem Siddiqi,Steven W. Zucker,Yves Bérubé Lauzière,Allen Tannenbaum	1997	Several active contour models have been proposed to unify the curve evolution framework with classical energy minimization techniques for segmentation, such snakes. The essential idea is to evolve a curve (in 20) or a surface (in 30) under constraints from image forces so that it clings to features of interest in an intensity image. Recently the evolution equation has. been derived from first principles as the gradient flow that minimizes a modified length functional, tailored fo features such as edges. However, because the jlow may slow to converge in practice, a constant (hyperbolic) term is added to keep the curve/surface moving in the desired direction. In this paper, we provide a justification for this term based on the gradient flow derived from a weighted area functional, with image dependent weighting factor. When combined with the earlier modified length gradient flow we obtain a pde which offers number of advantages, as illustrated by several examples of shape segmentation on medical images. In many cases the weighted area flow may be used on its own, with significant computational savings.
CVPR	Object recognition using invariant profiles.	David Slater,Glenn Healey	1997	We derive a sensitivity analysis for moment invariants of multidimensional distributions. These invariants have many uses in computational systems and have recently been used for illumination-invariant recognition in color images. In this context, the sensitivity analysis predicts the response of moment invariants to partial occlusion. Using the results of the sensitivity analysis, we develop a novel surface representation called the invariant profile which captures color distribution and spatial information while remaining invariant to the spectral content of the scene illumination. Unlike previous representations, the recognition of invariant profiles does not require illumination correction. We demonstrate the sensitivity analysis and the use of invariant profiles for recognition with a set of experiments on color images.
CVPR	Video Skimming and Characterization through the Combination of Image and Language Understanding Techniques.	Michael A. Smith	1997	"Digital video is rapidly becoming important for education, entertainment, and a host of multimedia applications. With the size of the video collections growing to thousands of hours, technology is needed to effectively browse seg ments in a short time without losing the content of the video. We propose a method to extract the significant audio and video information and create a ""skim"" video which represents a very short synopsis of the original. The goal of this work is to show the utility of integrating lan guage and image understanding techniques for video skimming by extraction of significant information, such as specific objects, audio keywords and relevant video struc ture. The resulting skim video is much shorter, where com paction is as high as 20:1, and yet retains the essential content of the original segment."
CVPR	Removing the bias from line detection.	Carsten Steger	1997	The extraction of curvilinear structures is an important low-level operation in computer vision. Most existing operators use a simple model for the line that is to be extracted, i.e., they do not take into account the surroundings of a line. Therefore, they will estimate a wrong line position whenever a line with different lateral contrast is extracted. In contrast, the algorithm proposed in this paper uses an explicit model for lines and their surroundings. By analyzing the scale-space behavior of a model line profile, it is shown how the bias that is induced by asymmetrical lines can be removed. Thus, the algorithm is able to extract an unbiased line position and width, both with sub-pixel accuracy.
CVPR	Lens distortion calibration using point correspondences.	Gideon P. Stein	1997	This paper describes a new method for lens distortion calibration using only point correspondences in multiple views, without the need to know either the 3D location of the points or the camera locations. The standard lens distortion model is a model of the deviations of a real camera from the ideal pinhole or projective camera model.Given multiple views of a set of corresponding points taken by ideal pinhole cameras there exist epipolar and trilinear constraints among pairs and triplets of these views. In practice, due to noise in the feature detection and due to lens distortion these constraints do not hold exactly and we get some error. The calibration is a search for the lens distortion parameters that minimize this error. Using simulation and experimental results with real images we explore the properties of this method. We describe the use of this method with the standard lens distortion model, radial and decentering, but it could also be used with any other parametric distortion models. Finally we demonstrate that lens distortion calibration improves the accuracy of 3D reconstruction.
CVPR	Model-based brightness constraints: on direct estimation of structure and motion.	Gideon P. Stein,Amnon Shashua	1997	We describe a new direct method for estimating structure and motion from image intensities of multiple views. We extend the direct methods of Horn and Weldon [18] to three views. Adding the third view enables us to solve for motion and compute a dense depth map of the scene, directly from image spatio-temporal derivatives in a linear manner without first having to find point correspondences or compute optical flow. We describe the advantages and limitations of this method which are then verified with experiments using real images
CVPR	Critical Motion Sequences for Monocular Self-Calibration and Uncalibrated Euclidean Reconstruction.	Peter F. Sturm	1997	In this paper, sequences of camera motions that lead to inherent ambiguities in uncalibrated Euclidean reconstruction or self-calibration are studied. Our main contribution is a complete, detailed classification of these critical motion sequences (CMS). The practically important classes are identified and their degrees of ambiguity are derived. We also discuss some practical issues, especially concerning the reduction of the ambiguity of a reconstruction.
CVPR	Are Textureless Scenes Recoverable?	Hari Sundaram,Shree K. Nayar	1997	It is widely accepted that textureless surfaces cannot be recovered using passive sensing techniques. The problem is approached by viewing image formation as a fully three-dimensional mapping. It is shown that the lens encodes structural information of the scene within a compact three-dimensional space behind it. After analyzing the information content of this space and by using its properties we derive necessary and sufficient conditions for the recovery of textureless scenes. Based on these conditions, a simple procedure for recovering textureless scenes is described. We experimentally demonstrate the recovery of three textureless surfaces, namely, a line, a plane, and a paraboloid. Since textureless surfaces represent the worst case recovery scenario, all the results and the recovery procedure are naturally applicable to scenes with texture.
CVPR	Shocks from images: propagation of orientation elements.	Hüseyin Tek,Perry A. Stoll,Benjamin B. Kimia	1997	The extraction of figure symmetry from image contours faces a number of fundamental difficulties: object symmetries are distorted due to (i) gaps in the bounding contour of a shape due to figure-ground blending, weak contrast edges, highlights, noise, etc.; (ii) an introduction of parts and occluders, and (iii) spurious edge elements due to surface markings, texture, etc. A framework for extracting such symmetries from real images is proposed based on the propagation of contour orientation information and the detection of four types of singularities (shocks) arising from the collision of propagating elements. In this paper, we show that an additional labeling of shocks based on whether the colliding wavefronts carry true orientation information (regular vs. rarefaction waves) allows a division of shocks into three sets: regular shocks are the partial shocks of partial contours as they remain invariant to the completion of the contour; semi-degenerate and degenerate shocks depict potential parts and gaps. Finally, shocks altered due to spurious edges, occlusion, and gaps are recovered via a simulation of inter-penetrating waves generated at select shock groups which with the aid of the above shock labels leads to second and further generations of shocks.
CVPR	A Computational Approach to Steerable Functions.	Patrick C. Teo,Yacov Hel-Or	1997	We present a computational, group-theoretic approach to steerable functions. The approach is group-theoretic in that the treatment involves continuous transformation groups for which elementary Lie group theory may be applied. The approach is computational in that the theory is constructive and leads directly to a procedural implementation. For functions that are steerable with $n$ finite number of basis functions under a $k$-parameter group, the procedure is efficient and is guaranteed to return the minimum number of basis functions. If the function is not steerable, a numerical implementation of the procedure could also be used to compute basis functions that approximately steer the function over a range of transformation parameters. Examples of both applications are demonstrated.
CVPR	An Assessment of Information Criteria for Motion Model Selection.	Philip H. S. Torr	1997	Rigid motion imposes constraints on the motion of image points between the two images. The matched points must conform to one of several possible constraints, such as that given by the fundamental matrix or image-image homography, and it is essential to know which model to fit to the data before recovery of structure, matching or segmentation can be performed successfully. This paper compares several model selection methods with a particular emphasis on providing a method that will work fully automatically on real imagery.
CVPR	Autocalibration and the absolute quadric.	Bill Triggs	1997	The author describes a new method for camera autocalibration and scaled Euclidean structure and motion, from three or more views taken by a moving camera with fixed but unknown intrinsic parameters. The motion constancy of these is used to rectify an initial projective reconstruction. Euclidean scene structure is formulated in terms of the absolute quadric-the singular dual 3D quadric (4/spl times/4 rank 3 matrix) giving the Euclidean dot-product between plane normals. This is equivalent to the traditional absolute conic but simpler to use. It encodes both affine and Euclidean structure, and projects very simply to the dual absolute image conic which encodes camera calibration. Requiring the projection to be constant gives a bilinear constraint between the absolute quadric and image conic, from which both can be recovered nonlinearly from m/spl ges/3 images, or quasi-linearly from m/spl ges/4. Calibration and Euclidean structure follow easily. The nonlinear method is stabler, faster, more accurate and more general than the quasi-linear one. It is based on a general constrained optimization technique-sequential quadratic programming-that may well be useful in other vision problems.
CVPR	Ego-Motion Estimation Using Optical Flow Fields Observed from Multiple Cameras.	An-Ting Tsao,Chiou-Shann Fuh,Yi-Ping Hung,Yong-Sheng Chen	1997	In this paper, we consider a multi-camera vision system mounted on a moving object in a static three-dimensional environment. By using the motion flow fields seen by all of the cameras, an algorithm which does not need to solve the point-correspondence problem among the cameras is proposed to estimate the 3D ego-motion parameters of the moving object. Our experiments have shown that using multiple optical flow fields obtained from different cameras can be very helpful for ego-motion estimation.
CVPR	Feature Tracking from an Image Sequence Using Geometric Invariants.	Hung-Tat Tsui,Zhong-Ying Zhang,Shao-Hua Kong	1997	"In this paper two new feature tracking algorithms are proposed. In the first algorithm, a perspective camera model is used. Making use of the projective invariant of Barrett, and assuming the image feature points corresponding to 8 general points in space are tracked by a conventional method in the image sequence, the other feature points in the sequence can be tracked using a Hough technique. Correspondence between two reference images as required by the original Barrett's invariant is not necessary. In the second algorithm, an affine camera model is assumed and the image feature points corresponding to 4 non-coplanar points in space are assumed tracked in the image sequence using a conventional method. These image points form the basis of affine coordinates in each image. After the correspondence of a fifth point is established between the first two images, the affine coordinates of all image points in the first images can be computed. As far as we know, this is the only algorithm which can transfer a point knowing only a single image. Experiments showed that both algorithms gave highly accurate tracking results."
CVPR	Empirical Bayesian EM-based Motion Segmentation.	Nuno Vasconcelos,Andrew Lippman	1997	A recent trend in motion-based segmentation has been to rely on statistical procedures derived from Expectation-Maximization (EM) principles. EM-based approaches have various attractives for segmentation, such as proceeding by taking non-greedy soft decisions with regards to the assignment of pixels to regions, or allowing the use of sophisticated priors capable of imposing spatial coherence on the segmentation. A practical difficulty with such priors is, however, the determination of appropriate values for their parameters. In this work, we exploit the fact that the EM framework is itself suited for empirical Bayesian data analysis to develop an algorithm that finds the estimates of the prior parameters which best explain the observed data. Such an approach maintains the Bayesian appeal of incorporating prior beliefs, but requires only a qualitative description of the prior, avoiding the requirement of a quantitative specification of its parameters. This eliminates the need for trial-and-error strategies for parameter determination and leads to better segmentations in less iterations.
CVPR	A bootstrapping algorithm for learning linear models of object classes.	Thomas Vetter,Michael J. Jones,Tomaso Poggio	1997	Flexible models of object classes, based on linear combinations of prototypical images, are capable of matching novel images of the same class and have been shown to be a powerful tool to solve several fundamental vision tasks such as recognition, synthesis and correspondence. The key problem in creating a specific flexible model is the computation of pixelwise correspondence between the prototypes, a task done until now in a semiautomatic way. In this paper we describe an algorithm that automatically bootstraps the correspondence between the prototypes. The algorithm - which can be used for 2D images as well as for 3D models - is shown to synthesize successfully a flexible model of frontal face images and a flexible model of handwritten digits.
CVPR	Extracting Surface Textures and Microstructures from Multiple Aerial Images.	Xiaoguang Wang,Allen R. Hanson	1997	We present a surface texture and microstructure extraction system to provide added realism in visualization and virtual reality applications. The system uses multiple images, 3-D models and camera information, in addition to knowledge about man-made structures, to cope with problems such as perspective distortion, data deficiency, and corruption caused by shadows and occlusions. Combined with the Ascender site modeling system~\cite{Collins_IUW} and scene rendering algorithms, the system is typically useful for urban site model refinement and visualization.
CVPR	Using Three-Dimensional Features to Improve Terrain Classification.	Xiaoguang Wang,Frank Stolle,Howard J. Schultz,Edward M. Riseman,Allen R. Hanson	1997	Using Three-Dimensional Features to Improve Terrain Classification.
CVPR	Smoothness in Layers: Motion segmentation using nonparametric mixture estimation.	Yair Weiss	1997	"Grouping based on common motion, or ``common fate'' provides a powerful cue for segmenting image sequences. Recently a number of algorithms have been developed that successfully perform motion segmentation by assuming that the motion of each group can be described by a low dimensional parametric model (e.g. affine). Typically the assumption is that motion segments correspond to planar patches in 3D undergoing rigid motion. Here we develop an alternative approach, where the motion of each group is described by a smooth dense flow field and the stability of the estimation is ensured by means of a prior distribution on the class of flow fields. We present a variant of the EM algorithm that can segment image sequences by fitting multiple smooth flow fields to the spatiotemporal data. Using the method of Green's functions, we show how the estimation of a single smooth flow field can be performed in closed form, thus making the multiple model estimation computationally feasible. Furthermore, the number of models is estimated automatically using similar methods to those used in the parametric approach. We illustrate the algorithm's performance on synthetic and real image sequences."
CVPR	Using Local 3D Structure for Segmentation of Bone from Computer Tomography Images.	Carl-Fredrik Westin,Abhir Bhalerao,Ron Kikinis,Hans Knutsson	1997	In this paper we focus on using local 3D structure for segmentation. A tensor descriptor is estimated for each neighbourhood, i.e. for each voxel in the data set. The tensors are created from a combination of the outputs form a set of 3D quadrature filters. The shape of the tensors describe locally the structure of the neighborhood in terms of how much it is like a plane, a line, and a sphere. We apply this to segmentation of bone from Computer Tomography data (CT). Traditional methods are based purely on gray-level value discrimination and have difficulties in recovering thin bone structures due to so called partial voluming, a problem which is present in all such sampled data. We illuminate the partial voluming problem by showing that thresholding creates complicated artifacts even if the signal is densely enough sampled and can be perfectly reconstructed. The unwanted effects of thresholding can be reduced by a change of the signal basis. We show that by using additional local structure information can significantly reduce the degree of sampling artifacts. Evaluation of the method on a clinical case is presented, the segmentation of a human skull from a CT volume. The method shows that many of the thin bone structures which disappear in a pure thresholding can be recovered.
CVPR	Physically based fluid flow recovery from image sequences.	Richard P. Wildes,Michael J. Amabile,Ann-Marie Lanzillotto,Tzong-Shyng Leu	1997	This paper presents an approach to measuring fluid flow from image sequences. The approach centers around a motion recovery algorithm that is based on principles from fluid mechanics: The algorithm is constrained so that recovered flows observe conservation of mass as well as physically motivated boundary conditions. Empirical results are presented from application of the algorithm to fluid flows captured via transmittance imagery (i.e., radiographs). In these experiments, fluids seeded with tracers were driven through simple physical systems. The significance of this work is twofold. First, from a theoretical point of view it is shown how information derived from the physical behavior of fluids can be used to motivate a flow recovery algorithm. Second, from an applications point of view the developed algorithm can be used to augment the tools that are available for the measurement of fluid dynamics; other imaged flows that observe compatible constraints might benefit in a similar fashion.
CVPR	Hyper-patches for 3D model acquisition and tracking.	Charles Wiles,Atsuto Maki,Natsuko Matsuda,Mutsumi Watanabe	1997	Automatic 3D model acquisition and 3D tracking of simple objects under motion using a single camera is often difficult due to the sparsity of information from which to establish the model. We have developed an automatic scheme that first computes a simple pointalistic Euclidean model of the object and then enriches this model using hyper-patches. These hyper-patches contain information on both the orientation and intensity pattern variation of roughly planar patches on an object. This information allows both the spatial and intensity distortions of the projected patch to be modelled accurately under 3D object motion. We show that hyper-patches not only can be computed automatically during model acquisition from a monocular image sequence, but that they are also extremely appropriate for the task of visual tracking.
CVPR	Rational Discrete Generalized Cylinders and their Application to Shape Recovery in Medical Images.	James P. Williams,Lawrence B. Wolff,John K. Johnstone	1997	Generalized cylinders (GCs) are a popular representational tool in computer vision. In medical imaging, the curved axis GC is particularly applicable to a number of elongated physical structures such as vasculature, bone and bronchi. In many of these instances, it is necessary to recover curved-axis GCs with arbitrary cross-sections. It is also vital that these structures, once recovered, can be analyzed and visualized with off-the-shelf algorithms and software packages. Such tools are usually designed to operate on the domain of polynomial or rational surfaces. Unfortunately most extant, suitably versatile GC representations do not admit rational parameterizations. We develop an entirely rational B-spline representation for generalized cylinders with curved axes and arbitrary cross-section functions. We demonstrate how our representation can be used as a deformable model by extracting a rational GC from pre-segmented spinal data using a discrete, dynamic surface fit.
CVPR	Temporal Classification of Natural Gesture and Application to Video Coding.	Andrew D. Wilson,Aaron F. Bobick,Justine Cassell	1997	A method for the temporal classification of natural gesture from video imagery is presented. The work is motivated by recent developments in the theory of natural gesture which have identified several key temporal aspects of gesture important to communication. In particular, gesticulation during conversation can be coarsely characterized as periods of bi-phasic or tri-phasic gesture separated by a rest state. We first present an automatic procedure for hypothesizing plausible rest state configurations of a speaker. Second, we develop a state-based parsing algorithm used to both select among candidate rest states and to parse an incoming video stream into bi-phasic and tri-phasic gestures. Finally, we demonstrate the use of the bi-phasic/tri-phasic labeling to select semantically significant static images for low bandwidth coding of video of story-telling speakers.
CVPR	A Minimum-Variance Adaptive Surface Mesh.	Richard C. Wilson,Edwin R. Hancock	1997	Our main contribution in this paper is to describe a new class of the adaptive mesh. The mesh uses both split and merge operations to adapt itself to the structure of volumetric data-points. The adaptive behavior is controlled by the variance of the data-point positions about maximum-likelihood quadric patches. We show that the density of control points on the mesh is regulated by the curvature of the underlying surface. Finally, we illustrate the effectiveness of the method on both real-world and simulated data-sets.
CVPR	Error Analysis of a Real-Time Stereo Syste.	Yalin Xiong,Larry Matthies	1997	Correlation-based real-time stereo systems have been proven to be effective in applications such as robot navigation, elevation map building etc. This paper provides an in-depth analysis of the major error sources for such a real-time stereo system in the context of cross-country navigation of an autonomous vehicle. Three major types of errors: foreshortening error, misalignment error and systematic error, are identified. The combined disparity errors can easily exceed three-tenths of a pixel, which translates to significant range errors. Upon understanding these error sources, we demonstrate different approaches to either correct them or model their magnitudes without excessive additional computations. By correcting those errors, we show that the precision of the stereo algorithm can be improved by 50%.
CVPR	Creating Image-Based VR Using a Self-Calibrating Fisheye Lens.	Yalin Xiong,Kenneth Turkowski	1997	Image-based virtual reality is emerging as a major alternative to the more traditional 3D-based VR. The main advantages of the image-based VR are its photo-quality realism and 3D illusion without any 3D information. Unfortunately, creating content for image-based VR is usually a very tedious process. This paper proposes to use a non-perspective fisheye lens to capture the spherical panorama with very few images. Unlike most of camera calibration in computer vision, self-calibration of the fisheye lens poses new questions regarding the parameterization of the distortion and wrap-around effects. Because of its unique projection model and large field of view (near 180 degrees), most of the ambiguity problems in self-calibrating a traditional lens can be solved trivially. We demonstrate that with four fisheye lens images, we can seamlessly register them to create the spherical panorama, while self-calibrating its distortion and field of view.
CVPR	Determining a Polyhedral Shape Using Interreflections.	Jun Yang,Dili Zhang,Noboru Ohnishi,Noboru Sugie	1997	We discuss the uniqueness of 3-D shape recovery of a polyhedron from a single shading image. First, we analytically show that multiple convex (and concave) shape solutions usually exist for a simple polyhedron if interreflections are not considered. Then we propose a new approach to uniquely determine the concave shape solution using interreflections as a constraint. A numerical example, in which two convex shapes and two concave shapes exist for a trihedral corner, has been shown by Horn. However, it is difficult to prove the uniqueness using constraint equations. We analytically show that multiple convex (and concave) shape solutions usually exist for a pyramid using a reflectance map, if interreflection distribution is not considered. However, if interreflection distribution is used as a constraint that limits the shape solution (for a concave polyhedron), the polyhedral shape can be uniquely determined. Interreflections are used as a constraint to determine the shape solution in our approach.
CVPR	Gradient Vector Flow: A New External Force for Snakes.	Chenyang Xu,Jerry L. Prince	1997	Snakes, or active contours, are used extensively in computer vision and image processing applications, particularly to locate object boundaries. Problems associated with initialization and poor convergence to concave boundaries, however, have limited their utility. This paper develops a new external force for active contours, largely solving both problems. This external force, which we call gradient vector flow (GVF), is computed as a diffusion of the gradient vectors of a gray-level or binary edge map derived from the image. The resultant field has a large capture range and forces active contours into concave regions. Examples on simulated images and one real image are presented.
CVPR	Temporal Multi-scale Models for Flow and Acceleration.	Yaser Yacoob,Larry S. Davis	1997	A model for computing image flow in image sequences containing a very wide range of instantaneous flows is proposed. This model integrates the spatio-temporal image derivatives from multiple temporal scales to provide both reliable and accurate instantaneous flow estimates. The integration employs robust regression and automatic scale weighting in a generalized brightness constancy framework. In addition to instantaneous flow estimation the model supports recovery of dense estimates of image acceleration and can be readily combined with parameterized flow and acceleration models. A demonstration of performance on image sequences of typical human actions taken with a high frame-rate camera is given.
CVPR	Identification of salient contours in cluttered images.	Shih-Cheng Yen,Leif H. Finkel	1997	We present a model of contour extraction in which the perceptual salience of contours arises from long-range interactions between orientation-selective filters. It has been previously shown that salient contours may be extracted from noisy images by using a number heuristic features. Our algorithm is based on cortical mechanisms, and simulations show close agreement with results from recent anatomical, physiological and psychophysical studies including recent results of D.J. Field et al. (1993), I. Kovacs et al., and M.K. Kapadia et al. (1995). The performance of the algorithm is demonstrated on a range of psychophysical stimuli and real images.
CVPR	Robust Occluding Contour Detection Using the Hausdorff Distance.	Xilin Yi,Octavia I. Camps	1997	In this paper, a correlational approach for distinguishing occluding contours from object markings for 3D object modeling is presented. The proposed method is valid under weak perspective projection, does not require to search for correspondences between frames, can handle scaling between consecutive images, thus can estimate the full Euclidean surface structure, and does not require camera calibration or camera motion measurement. Extensive experimental results show that the method is robust to the occlusion of feature points and image noise unlike previous affine-based approaches. Qualitative and quantitative results for the relation between the required minimum viewing angle change for the detection and the surface curvature are also presented.
CVPR	Surface Shape from Warping.	Alan L. Yuille,Mario Ferraro,Tony Zhang	1997	We propose a variant of shape from shading which we call shape from image warping. The idea is that the three-dimensional shape of an object is estimated by determining how much the image of the object is warped with respect to the image of a known prototype shape. We demonstrate that, for a class of reflectance functions, there is a direct relationship between these image warps and geometric warps of the underlying three-dimensional shapes. Therefore detecting the image warp relative to a prototype of known shape allows us to reconstruct the shape of the imaged object. We derive properties of these shape warps and illustrate the results by recovering the shapes of faces. This relationship between image and shape warps helps us understand the relationship between image based models of object recognition and approaches based on three dimensional object models.
CVPR	Shape and albedo from multiple images using integrability.	Alan L. Yuille,Daniel Snow	1997	Previous work has developed an approach for estimating shape and albedo from multiple images assuming Lambertian reflectance with single light sources. The main contributions of this paper are: (i) to show how the approach can be generalized to include ambient background illumination, (ii) to demonstrate the use of the integrability constraint for solving this problem, and (iii) an iterative algorithm which is able to improve the analysis by finding shadows and rejecting them.
CVPR	Learning Generic Prior Models for Visual Computation.	Song Chun Zhu,David Mumford	1997	This paper presents a novel theory for learning generic prior models from a set of observed natural images based on a minimax entropy theory that the authors studied in texture modeling. We start by studying the statistics of natural images including the scale invariant properties, then generic prior models were learnt to duplicate the observed statistics. The learned Gibbs distributions confirm and improve the forms of existing prior models, and more interestingly inverted potentials are found to be necessary, and such potentials produce patterns and enhance preferred image features. The learned model is compared with existing prior models in experiments of image restoration.
CVPR	Fully 3D Active Surface Models with Self--Inflation and Self--Deflation Forces.	Zixin Zhang,Michael Braun	1997	In this paper, we propose fully 3D active surface models for image segmentation. Our models are capable of fitting a diverse range of region shapes. They have low sensitivity to initial shape and position. We design self--inflation/deflation forces, which cooperate naturally with gradient forces. They permit the active surface to travel a long distance without the aid of any external forces. They are easily controlled in both their directions and magnitude. The models produce accurate segmentation when tested with synthetic and real images. They manifest robustness to image noise and imperfect image data. Importantly, they are capable of converging to the correct boundary even if the initial estimate is not close.
CVPR	Multi-Scale Classification of 3-D Objects.	Dongmei Zhang,Martial Hebert	1997	We describe an approach to the classification of 3-D objects using a multi-scale representation. This approach starts with a smoothing algorithm for representing objects at different scales. Smoothing is applied in curvature space directly, thus avoiding the usual shrinkage problems and allowing for efficient implementations. A 3-D similarity measure that integrates the representations of the objects at multiple scales is introduced. Given a library of models, objects that are similar based on this multi-scale measure are grouped together into classes. The objects that are in the same class are combined into a single prototype object. Finally, the prototypes are used for hierarchical recognition by first comparing the scene representation to the prototypes and then matching it only to the objects in the most likely class rather than to the entire library of models. Beyond its application to object recognition, this approach provides an attractive implementation of the intuitive notions of scale and approximate similarity for 3-D shapes.
CVPR	Using geometric corners to build a 2D mosaic from a set of image.	Imad Zoghlami,Olivier D. Faugeras,Rachid Deriche	1997	The main problem for building a mosaic is the computation of the warping functions (homographies). In fact two cases are to be distinguished. The first is when the homography is mainly a translation (i.e. the rotation around the optical axis and the zooming factor are small). The second is the general case (when the rotation around the optical axis and zooming are arbitrary). Some efficient methods have been developed to solve the first case. But the second case is more difficult, in particular, when the rotation around the optical axis is very large (90 degrees or more). Often in this case human interaction is needed to provide a first approximation of the transformation that will bring us back to the first case. In this article we present a method to solve this problem without human interaction for any rotation around the optical axis and fairly large zooming factors.
CVPR	Self-Maintaining Camera Calibration Over Time.	Zhengyou Zhang,Veit Schenk	1997	The success of an intelligent robotic system depends on the performance of its vision-system which in turn depends to a great extend upon the quality of its calibration. During the execution of a task the vision-system is subject to external influences such as vibrations, thermal expansion etc. which affect and possibly render invalid the initial calibration. Moreover, it is possible that the parameters of the vision-system like e.g. the zoom or the focus are altered intentionally in order to perform specific vision-tasks. This paper describes a technique for automatically maintaining calibration of stereovision systems over time without using again any particular calibration apparatus. It uses all available information, i.e. both spatial and temporal data. Uncertainty is systematically manipulated and maintained. Synthetical and real data are used to validate the proposed technique, and the results compare very favourably with those given by classical calibration methods.
CVPR	"1997 Conference on Computer Vision and Pattern Recognition (CVPR '97), June 17-19, 1997, San Juan, Puerto Rico"		1997	"1997 Conference on Computer Vision and Pattern Recognition (CVPR '97), June 17-19, 1997, San Juan, Puerto Rico"
IEEE Visualization	Wildfire visualization (case study).	James P. Ahrens,Patrick S. McCormick,James Bossert,Jon Reisner,Judith Winterkamp	1997	Wildfire visualization (case study).
IEEE Visualization	Efficient subdivision of finite-element datasets into consistent tetrahedra.	Guy Albertelli,Roger Crawfis	1997	Efficient subdivision of finite-element datasets into consistent tetrahedra.
IEEE Visualization	Architectural walkthroughs using portal textures.	Daniel G. Aliaga,Anselmo Lastra	1997	Architectural walkthroughs using portal textures.
IEEE Visualization	Two-phase perspective ray casting for interactive volume navigation.	Martin L. Brady,Kenneth K. Jung,H. T. Nguyen,Thinh P. Q. Nguyen	1997	Two-phase perspective ray casting for interactive volume navigation.
IEEE Visualization	Displaying data in multidimensional relevance space with 2D visualization maps.	Jackie Assa,Daniel Cohen-Or,Tova Milo	1997	Displaying data in multidimensional relevance space with 2D visualization maps.
IEEE Visualization	Interactive visualization of aircraft and power generation engines.	Lisa Sobierajski Avila,William J. Schroeder	1997	Interactive visualization of aircraft and power generation engines.
IEEE Visualization	The contour spectrum.	Chandrajit L. Bajaj,Valerio Pascucci,Daniel Schikore	1997	The contour spectrum.
IEEE Visualization	Instructional software for visualizing optical phenomena.	David C. Banks,John T. Foley,Kiril Vidimce,Ming-Hoe Kiu	1997	Instructional software for visualizing optical phenomena.
IEEE Visualization	Principal stream surfaces.	Wenli Cai,Pheng-Ann Heng	1997	Principal stream surfaces.
IEEE Visualization	Repairing CAD models.	Gill Barequet,Subodh Kumar	1997	Repairing CAD models.
IEEE Visualization	Application-controlled demand paging for out-of-core visualization.	Michael Cox,David Ellsworth	1997	Application-controlled demand paging for out-of-core visualization.
IEEE Visualization	Image synthesis from a sparse set of views.	Qian Chen,Gérard G. Medioni	1997	Image synthesis from a sparse set of views.
IEEE Visualization	I/O optimal isosurface extraction (extended abstract).	Yi-Jen Chiang,Cláudio T. Silva	1997	I/O optimal isosurface extraction (extended abstract).
IEEE Visualization	Integrated volume compression and visualization.	Tzi-cker Chiueh,Chuan-Kai Yang,Taosong He,Hanspeter Pfister,Arie E. Kaufman	1997	Integrated volume compression and visualization.
IEEE Visualization	Optimized geometry compression for real-time rendering.	Mike M. Chow	1997	Optimized geometry compression for real-time rendering.
IEEE Visualization	Visualization of height field data with physical models and texture photomapping.	Dru Clark,Michael J. Bailey	1997	Visualization of height field data with physical models and texture photomapping.
IEEE Visualization	Simplifying polygonal models using successive mappings.	Jonathan D. Cohen,Dinesh Manocha,Marc Olano	1997	Simplifying polygonal models using successive mappings.
IEEE Visualization	Controlled simplification of genus for polygonal models.	Jihad El-Sana,Amitabh Varshney	1997	Controlled simplification of genus for polygonal models.
IEEE Visualization	Isosurface extraction using particle systems.	Patricia Crossno,Edward Angel	1997	Isosurface extraction using particle systems.
IEEE Visualization	Towards efficient visualization support for single-block and multi-block datasets.	Jean-Marie Favre	1997	Towards efficient visualization support for single-block and multi-block datasets.
IEEE Visualization	Global tele-immersion.	Thomas A. DeFanti	1997	Global tele-immersion.
IEEE Visualization	Collision detection for volumetric objects.	Taosong He,Arie E. Kaufman	1997	Collision detection for volumetric objects.
IEEE Visualization	Building and traversing a surface at variable resolution.	Leila De Floriani,Paola Magillo,Enrico Puppo	1997	Building and traversing a surface at variable resolution.
IEEE Visualization	Determination of unknown particle charges in a thunder cloud based upon detected electric field vectors.	Dan Drake,Thomas Simpson,Larry Smithmier,Penny Rheingans	1997	Determination of unknown particle charges in a thunder cloud based upon detected electric field vectors.
IEEE Visualization	ROAMing terrain: real-time optimally adapting meshes.	Mark A. Duchaineau,Murray Wolinsky,David E. Sigeti,Mark C. Miller,Charles Aldrich,Mark B. Mineev-Weinstein	1997	ROAMing terrain: real-time optimally adapting meshes.
IEEE Visualization	Accelerated volume rendering using homogeneous region encoding.	Jason Freund,Kenneth R. Sloan	1997	Accelerated volume rendering using homogeneous region encoding.
IEEE Visualization	Collaborative augmented reality: exploring dynamical systems.	Anton L. Fuhrmann,Helwig Löffelmann,Dieter Schmalstieg	1997	Collaborative augmented reality: exploring dynamical systems.
IEEE Visualization	GADGET: goal-oriented application design guidance for modular visualization environments.	Issei Fujishiro,Yuriko Takeshima,Yoshihiko Ichikawa,Kyoko Nakamura	1997	GADGET: goal-oriented application design guidance for modular visualization environments.
IEEE Visualization	Smooth hierarchical surface triangulations.	Tran S. Gieng,Bernd Hamann,Kenneth I. Joy,Gregory L. Schussman,Issac J. Trotts	1997	Smooth hierarchical surface triangulations.
IEEE Visualization	Information exploration shootout project and benchmark data sets: evaluating how visualization does in analyzing real-world data analysis problems (panel).	Georges G. Grinstein,Sharon J. Laskowski,Graham J. Wills,Bernice E. Rogowitz	1997	Information exploration shootout project and benchmark data sets: evaluating how visualization does in analyzing real-world data analysis problems (panel).
IEEE Visualization	The multilevel finite element method for adaptive mesh optimization and visualization of volume data.	Roberto Grosso,Christoph Lürig,Thomas Ertl	1997	The multilevel finite element method for adaptive mesh optimization and visualization of volume data.
IEEE Visualization	Constrained 3D navigation with 2D controllers.	Andrew J. Hanson,Eric A. Wernert	1997	Constrained 3D navigation with 2D controllers.
IEEE Visualization	DNA visual and analytic data mining.	Patrick Hoffman,Georges G. Grinstein,Kenneth A. Marx,Ivo Grosse,Eugene Stanley	1997	DNA visual and analytic data mining.
IEEE Visualization	Terascale visualization: approaches, pitfalls and issues (panel).	Carol L. Hunter,Roger Crawfis,Michael Cox,Bernd Hamann,Charles D. Hansen,Mark C. Miller	1997	Terascale visualization: approaches, pitfalls and issues (panel).
IEEE Visualization	Strategies for effectively visualizing 3D flow with volume LIC.	Victoria Interrante,Chester Grosch	1997	Strategies for effectively visualizing 3D flow with volume LIC.
IEEE Visualization	CAVEvis: distributed real-time visualization of time-varying scalar and vector fields using the CAVE virtual reality theater.	Vijendra Jaswal	1997	CAVEvis: distributed real-time visualization of time-varying scalar and vector fields using the CAVE virtual reality theater.
IEEE Visualization	The motion map: efficient computation of steady flow animations.	Bruno Jobard,Wilfrid Lefer	1997	The motion map: efficient computation of steady flow animations.
IEEE Visualization	Efficient visualization of physical and structural properties in crash-worthiness simulations (case study).	Sven Kuschfeldt,Thomas Ertl,Michael Holzner	1997	Efficient visualization of physical and structural properties in crash-worthiness simulations (case study).
IEEE Visualization	Vortex identification - applications in aerodynamics: a case study.	David N. Kenwright,Robert Haimes	1997	Vortex identification - applications in aerodynamics: a case study.
IEEE Visualization	Singularities in nonuniform tensor fields.	Yingmei Lavin,Yuval Levy,Lambertus Hesselink	1997	Singularities in nonuniform tensor fields.
IEEE Visualization	Extracting feature lines from 3D unstructured grids.	Kwan-Liu Ma,Victoria Interrante	1997	Extracting feature lines from 3D unstructured grids.
IEEE Visualization	Visualization of rotation fields.	Mark A. Livingston	1997	Visualization of rotation fields.
IEEE Visualization	Visualization of plant growth.	Jeremy J. Loomis,Xiuwen Liu,Zhaohua Ding,Kikuo Fujimura,Michael L. Evans,Hideo Ishikawa	1997	Visualization of plant growth.
IEEE Visualization	A comparison of normal estimation schemes.	Torsten Möller,Raghu Machiraju,Klaus Mueller,Roni Yagel	1997	A comparison of normal estimation schemes.
IEEE Visualization	Dynamic smooth subdivision surfaces for data visualization.	Chhandomay Mandal,Hong Qin,Baba C. Vemuri	1997	Dynamic smooth subdivision surfaces for data visualization.
IEEE Visualization	Viewing IGES files through VRML.	Jed Marti	1997	Viewing IGES files through VRML.
IEEE Visualization	Computing the separating surface for segmented data.	Gregory M. Nielson,Richard Franke	1997	Computing the separating surface for segmented data.
IEEE Visualization	Haar wavelets over triangular domains with applications to multiresolution models for flow over a sphere.	Gregory M. Nielson,Il-Hong Jung,Junwon Sung	1997	Haar wavelets over triangular domains with applications to multiresolution models for flow over a sphere.
IEEE Visualization	Interval volume tetrahedrization.	Gregory M. Nielson,Junwon Sung	1997	Interval volume tetrahedrization.
IEEE Visualization	VizWiz: a Java applet for interactive 3D scientific visualization on the web.	Cherilyn Michaels,Michael J. Bailey	1997	VizWiz: a Java applet for interactive 3D scientific visualization on the web.
IEEE Visualization	Dissolving Descartes: perception and the construction of reality (address).	Mark Pesce	1997	Dissolving Descartes: perception and the construction of reality (address).
IEEE Visualization	An interactive cerebral blood vessel exploration system.	Anna Puig,Dani Tost,Isabel Navazo	1997	An interactive cerebral blood vessel exploration system.
IEEE Visualization	Visualization of large terrains in resource-limited computing environments.	Boris Rabinovich,Craig Gotsman	1997	Visualization of large terrains in resource-limited computing environments.
IEEE Visualization	Virtualized reality: constructing time-varying virtual worlds from real world events.	Peter Rander,P. J. Narayanan,Takeo Kanade	1997	Virtualized reality: constructing time-varying virtual worlds from real world events.
IEEE Visualization	Dynamic color mapping of bivariate qualitative data.	Penny Rheingans	1997	Dynamic color mapping of bivariate qualitative data.
IEEE Visualization	Perceptual measures for effective visualizations.	Holly E. Rushmeier,Harrison H. Barrett,Penny Rheingans,Samuel P. Uselton,Andrew Watson	1997	Perceptual measures for effective visualizations.
IEEE Visualization	Visualizing customer segmentations produce by self organizing maps (case study).	Holly E. Rushmeier,Richard D. Lawrence,George S. Almasi	1997	Visualizing customer segmentations produce by self organizing maps (case study).
IEEE Visualization	Visualization of higher order singularities in vector fields.	Gerik Scheuermann,Hans Hagen,Heinz Krüger,Martin Menzel,Alyn P. Rockwood	1997	Visualization of higher order singularities in vector fields.
IEEE Visualization	A topology modifying progressive decimation algorithm.	William J. Schroeder	1997	A topology modifying progressive decimation algorithm.
IEEE Visualization	UFLIC: a line integral convolution algorithm for visualizing unsteady flows.	Han-Wei Shen,David L. Kao	1997	UFLIC: a line integral convolution algorithm for visualizing unsteady flows.
IEEE Visualization	Visualization of geometric algorithms in an electronic classroom.	Maria Shneerson,Ayellet Tal	1997	Visualization of geometric algorithms in an electronic classroom.
IEEE Visualization	A visualization of music.	Sean M. Smith,Glen N. Williams	1997	A visualization of music.
IEEE Visualization	Multiresolution compression and reconstruction.	Oliver G. Staadt,Markus H. Gross,Roger Weber	1997	Multiresolution compression and reconstruction.
IEEE Visualization	An anti-aliasing technique for splatting.	J. Edward Swan II,Klaus Mueller,Torsten Möller,Naeem Shareef,Roger Crawfis,Roni Yagel	1997	An anti-aliasing technique for splatting.
IEEE Visualization	Volume rendering of abdominal aortic aneurysms.	Roger C. Tam,Christopher G. Healey,Borys Flak,Peter Cahoon	1997	Volume rendering of abdominal aortic aneurysms.
IEEE Visualization	Pearls found on the way to the ideal interface for scanned-probe microscopes.	Russell M. Taylor II,Jun Chen,Shoji Okimoto,Noel Llopis-Artime,Vernon L. Chi,Frederick P. Brooks Jr.,Mike Falvo,Scott Paulson,Pichet Thiansathaporn,David Glick,Sean Washburn,Richard Superfine	1997	Pearls found on the way to the ideal interface for scanned-probe microscopes.
IEEE Visualization	exVis: developing a wind tunnel data visualization tool.	Samuel P. Uselton	1997	exVis: developing a wind tunnel data visualization tool.
IEEE Visualization	Auralization of streamline vorticity in computational fluid dynamics data.	Christopher R. Volpe,Ephraim P. Glinert	1997	Auralization of streamline vorticity in computational fluid dynamics data.
IEEE Visualization	Fast oriented line integral convolution for vector field visualization via the Internet.	Rainer Wegenkittl,Eduard Gröller	1997	Fast oriented line integral convolution for vector field visualization via the Internet.
IEEE Visualization	Visualizing the behavior of higher dimensional dynamical systems.	Rainer Wegenkittl,Helwig Löffelmann,Eduard Gröller	1997	Visualizing the behavior of higher dimensional dynamical systems.
IEEE Visualization	The VSBUFFER: visibility ordering of unstructured volume primitives by polygon drawing.	Rüdiger Westermann,Thomas Ertl	1997	The VSBUFFER: visibility ordering of unstructured volume primitives by polygon drawing.
IEEE Visualization	Multivariate visualization using metric scaling.	Pak Chung Wong,R. Daniel Bergeron	1997	Multivariate visualization using metric scaling.
IEEE Visualization	Brushing techniques for exploring volume datasets.	Pak Chung Wong,R. Daniel Bergeron	1997	Brushing techniques for exploring volume datasets.
IEEE Visualization	Collaborative visualization.	Jason Wood,Helen Wright,Ken Brodlie	1997	Collaborative visualization.
IEEE Visualization	Interactive volume rendering for virtual colonoscopy.	Suya You,Lichan Hong,Ming Wan,Kittiboon Junyaprasert,Arie E. Kaufman,Shigeru Muraki,Yong Zhou,Mark Wax,Zhengrong Liang	1997	Interactive volume rendering for virtual colonoscopy.
IEEE Visualization	Multiresolution tetrahedral framework for visualizing regular volume data.	Yong Zhou,Baoquan Chen,Arie E. Kaufman	1997	Multiresolution tetrahedral framework for visualizing regular volume data.
IEEE Visualization	Wavelet-based multiresolutional representation of computational field simulation datasets.	Zhifan Zhu,Raghu Machiraju,Bryan Fry,Robert J. Moorhead	1997	Wavelet-based multiresolutional representation of computational field simulation datasets.
IEEE Visualization	"IEEE Visualization '97, Proceedings, Phoenix, AZ, USA, October 19-24, 1997."		1997	"IEEE Visualization '97, Proceedings, Phoenix, AZ, USA, October 19-24, 1997."
IEEE Trans. Vis. Comput. Graph.	The Discrete Analytical Hyperspheres.	Eric Andres,Marie-Andrée Jacob	1997	"An analytical definition of a discrete hypersphere with arbitrary center, radius, and thickness in dimension n is introduced. The new discrete hypersphere is called discrete analytical hypersphere. The hypersphere has important original properties including exact point localization, space tiling, k-separation, etc. These properties are almost obvious with this new discrete analytical definition contrary to the classical approaches based on digitization schemes. The analytically defined circle is compared to Pham's classically defined circle. Efficient incremental circle and hypersphere generation algorithms are provided."
IEEE Trans. Vis. Comput. Graph.	Multiresolution Representation and Visualization of Volume Data.	Paolo Cignoni,Claudio Montani,Enrico Puppo,Roberto Scopigno	1997	A system to represent and visualize scalar volume data at multiple resolution is presented. The system is built on a multiresolution model based on tetrahedral meshes with scattered vertices that can be obtained from any initial dataset. The model is built off-line through data simplification techniques, and stored in a compact data structure that supports fast on-line access. The system supports interactive visualization of a representation at an arbitrary level of resolution through isosurface and projective methods. The user can interactively adapt the quality of visualization to requirements of a specific application task and to the performance of a specific hardware platform. Representations at different resolutions can be used together to further enhance interaction and performance through progressive and multiresolution rendering.
IEEE Trans. Vis. Comput. Graph.	On Compatible Star Decompositions of Simple Polygons.	Michal Etzion,Ari Rappoport	1997	We introduce the notion of compatible star decompositions of simple polygons. In general, given two polygons with a correspondence between their vertices, two polygonal decompositions of the two polygons are said to be compatible if there exists a one-to-one mapping between them such that the corresponding pieces are defined by corresponding vertices. For compatible star decompositions, we also require correspondence between star points of the star pieces. Compatible star decompositions have applications in computer animation and shape representation and analysis.We present two algorithms for constructing compatible star decompositions of two simple polygons. The first algorithm is optimal in the number of pieces in the decomposition, providing that such a decomposition exists without adding Steiner vertices. The second algorithm constructs compatible star decompositions with Steiner vertices, which are not minimal in the number of pieces but are asymptotically worst case optimal in this number and in the number of added Steiner vertices. We prove that some pairs of polygons require ¿(n2) pieces, and that the decompositions computed by the second algorithm possess no more than O(n2) pieces.In addition to the contributions regarding compatible star decompositions, the paper also corrects an error in the only previously published polynomial algorithm for constructing a minimal star decomposition of a simple polygon, an error which might lead to a nonminimal decomposition.
IEEE Trans. Vis. Comput. Graph.	Speeding Up Isosurface Extraction Using Interval Trees.	Paolo Cignoni,Paola Marino,Claudio Montani,Enrico Puppo,Roberto Scopigno	1997	The interval tree is an optimally efficient search structure proposed by Edelsbrunner [5] to retrieve intervals on the real line that contain a given query value. We propose the application of such a data structure to the fast location of cells intersected by an isosurface in a volume dataset. The resulting search method can be applied to both structured and unstructured volume datasets, and it can be applied incrementally to exploit coherence between isosurfaces. We also address issues about storage requirements, and operations other than the location of cells, whose impact is relevant in the whole isosurface extraction task.In the case of unstructured grids, the overhead, due to the search structure, is compatible with the storage cost of the dataset, and local coherence in the computation of isosurface patches is exploited through a hash table. In the case of a structured dataset, a new conceptual organization is adopted, called the chess-board approach, which exploits the regular structure of the dataset to reduce memory usage and to exploit local coherence. In both cases, efficiency in the computation of surface normals on the isosurface is obtained by a precomputation of the gradients at the vertices of the mesh.Experiments on different kinds of input show that the practical performance of the method reflects its theoretical optimality.
IEEE Trans. Vis. Comput. Graph.	Dynamic Free-Form Deformations for Animation Synthesis.	Petros Faloutsos,Michiel van de Panne,Demetri Terzopoulos	1997	Free-form deformations (FFDs) are a popular tool for modeling and keyframe animation. This paper extends the use of FFDs to a dynamic setting. Our goal is to enable normally inanimate graphics objects, such as teapots and tables, to become animated, and learn to move about in a charming, cartoon-like manner. To achieve this goal, we implement a system that can transform a wide class of objects into dynamic characters. Our formulation is based on parameterized hierarchical FFDs augmented with Lagrangian dynamics, and provides an efficient way to animate and control the simulated characters. Objects are assigned mass distributions and elastic deformation properties, which allow them to translate, rotate, and deform according to internal and external forces. In addition, we implement an automated optimization process that searches for suitable control strategies. The primary contributions of the work are threefold. First, we formulate a dynamic generalization of conventional, geometric FFDs. The formulation employs deformation modes which are tailored by the user and are expressed in terms of FFDs. Second, the formulation accommodates a hierarchy of dynamic FFDs that can be used to model local as well as global deformations. Third, the deformation modes can be active, thereby producing locomotion.
IEEE Trans. Vis. Comput. Graph.	Complex Character Positioning Based on a Compatible Flow Model of Multiple Supports.	Ronan Boulic,Ramon Mas-Sanso,Daniel Thalmann	1997	We present a new posture design paradigm for the positioning of complex characters. It is illustrated here on human figures. We exploit the Inverse Kinetics technique which allows the center of mass position control for postures with either single or multiple supports. For the multiple support case, we introduce a Compatible Flow model of the supporting influence. With this approach, we are able to handle continuous modification of the support distribution. By construction, Inverse Kinetics presents the same control architecture as Inverse Kinematics, and, thus, it shows equivalent computing cost and similar intuitive concepts. Furthermore, Inverse Kinetics for the center of mass and Inverse Kinematics for fixed end effectors can be combined to generate a posture displaying static balance, goal-oriented features, and an additional gravity optimization.
IEEE Trans. Vis. Comput. Graph.	Animation of Deformable Models Using Implicit Surfaces.	Marie-Paule Cani,Mathieu Desbrun	1997	Animation of Deformable Models Using Implicit Surfaces.
IEEE Trans. Vis. Comput. Graph.	Physically-Based Stochastic Simplification of Mathematical Knots.	Robert P. Grzeszczuk,Milana Huang,Louis H. Kauffman	1997	"The article describes a tool for simplification and analysis of tangled configurations of mathematical knots. The proposed method addresses optimization issues common in energy-based approaches to knot classification. In this class of methods, an initially tangled elastic rope is ""charged"" with an electrostatic-like field which causes it to self-repel, prompting it to evolve into a mechanically stable configuration. This configuration is believed to be characteristic for its knot type. We propose a physically-based model to implicitly guard against isotopy violation during such evolution and suggest that a robust stochastic optimization procedure, simulated annealing, be used for the purpose of identifying the globally optimal solution. Because neither of these techniques depends on the properties of the energy function being optimized, our method is of general applicability, even though we applied it to a specific potential here. The method has successfully analyzed several complex tangles and is applicable to simplifying a large class of knots and links. Our work also shows that energy-based techniques will not necessarily terminate in a unique configuration, thus we empirically refute a prior conjecture that one of the commonly used energy functions (Simon's) is unimodal. Based on these results we also compare techniques that rely on geometric energy optimization to conventional algebraic methods with regards to their classification power."
IEEE Trans. Vis. Comput. Graph.	On Approximating Contours of the Piecewise Trilinear Interpolant Using Triangular Rational-Quadratic Bézier Patches.	Bernd Hamann,Issac J. Trotts,Gerald E. Farin	1997	On Approximating Contours of the Piecewise Trilinear Interpolant Using Triangular Rational-Quadratic Bézier Patches.
IEEE Trans. Vis. Comput. Graph.	Multipole Expansion of the Light Vector.	Alejo Hausner	1997	"Computing the light field due to an area light source remains an interesting problem in computer graphics. This paper presents a series approximation of the light field due to an unoccluded area source, by expanding the light field in spherical harmonics. The source can be nonuniform and need not be a planar polygon. The resulting formulas give expressions whose cost and accuracy can be chosen between the exact and expensive Lambertian solution for a diffuse polygon, and the fast but inexact method of replacing the area source by a point source of equal power. The formulas break the computation of the light vector into two phases: The first phase represents the light source's shape and brightness with numerical coefficients, and the second uses these coefficients to compute the light field at arbitrary locations. We examine the accuracy of the formulas for spherical and rectangular Lambertian sources, and apply them to obtaining light gradients. We also show how to use the formulas to estimate light from uniform polygonal sources, sources with polynomially varying radiosity, and luminous textures."
IEEE Trans. Vis. Comput. Graph.	A Wavelet Representation of Reflectance Functions.	Paul Lalonde,Alain Fournier	1997	Analytical models of light reflection are in common use in computer graphics. However, models based on measured reflectance data promise increased realism by making it possible to simulate many more types of surfaces to a greater level of accuracy than with analytical models. They also require less expert knowledge about the illumination models and their parameters. There are a number of hurdles to using measured reflectance functions, however. The data sets are very large. A reflectance distribution function sampled at five degrees angular resolution, arguably sparse enough to miss highlights and other high frequency effects, can easily require over a million samples, which in turn amount to over four megabytes of data. These data then also require some form of interpolation and filtering to be used effectively.In this paper, we examine issues of representation of measured reflectance distribution functions. In particular, we examine a wavelet basis representation of reflectance functions, and the algorithms required for efficient point-wise reconstruction of the BRDF. We show that the nonstandard wavelet decomposition leads to considerably more efficient algorithms than the standard wavelet decomposition. We also show that thresholding allows considerable improvement in running times, without unduly sacrificing image quality.
IEEE Trans. Vis. Comput. Graph.	A Visibility Matching Tone Reproduction Operator for High Dynamic Range Scenes.	Gregory Ward Larson,Holly E. Rushmeier,Christine D. Piatko	1997	We present a tone reproduction operator that preserves visibility in high dynamic range scenes. Our method introduces a new histogram adjustment technique, based on the population of local adaptation luminances in a scene. To match subjective viewing experience, the method incorporates models for human contrast sensitivity, glare, spatial acuity, and color sensitivity. We compare our results to previous work and present examples of our techniques applied to lighting simulation and electronic photography.
IEEE Trans. Vis. Comput. Graph.	The Topology of Symmetric, Second-Order 3D Tensor Fields.	Lambertus Hesselink,Yuval Levy,Yingmei Lavin	1997	"We study the topology of symmetric, second-order tensor fields. The results of this study can be readily extended to include general tensor fields through linear combination of symmetric tensor fields and vector fields. The goal is to represent their complex structure by a simple set of carefully chosen points, lines, and surfaces analogous to approaches in vector field topology. We extract topological skeletons of the eigenvector fields and use them for a compact, comprehensive description of the tensor field. Our approach is based on the premise: ""Analyze, then visualize.""The basic constituents of tensor topology are the degenerate points, or points where eigenvalues are equal to each other. Degenerate points play a similar role as critical points in vector fields. In tensor fields we identify two kinds of elementary degenerate points, which we call wedge points and trisector points. They can combine to form more familiar singularities¿such as saddles, nodes, centers, or foci. However, these are generally unstable structures in tensor fields. Based on the notions developed for 2D tensor fields, we extend the theory to include 3D degenerate points. Examples are given on the use of tensor field topology for the interpretation of physical systems."
IEEE Trans. Vis. Comput. Graph.	Scattered Data Interpolation with Multilevel B-Splines.	Seungyong Lee,George Wolberg,Sung Yong Shin	1997	This paper describes a fast algorithm for scattered data interpolation and approximation. Multilevel B-splines are introduced to compute a $C^2$-continuous surface through a set of irregularly spaced points. The algorithm makes use of a coarse-to-fine hierarchy of control lattices to generate a sequence of bicubic B-spline functions whose sum approaches the desired interpolation function. Large performance gains are realized by using B-spline refinement to reduce the sum of these functions into one equivalent B-spline function. Experimental results demonstrate that high-fidelity reconstruction is possible from a selected set of sparse and irregular samples.
IEEE Trans. Vis. Comput. Graph.	Conveying the 3D Shape of Smoothly Curving Transparent Surfaces via Texture.	Victoria Interrante,Henry Fuchs,Stephen M. Pizer	1997	"Transparency can be a useful device for depicting multiple overlapping surfaces in a single image. The challenge is to render the transparent surfaces in such a way that their three-dimensional shape can be readily understood and their depth distance from underlying structures clearly perceived.This paper describes our investigations into the use of sparsely-distributed discrete, opaque texture as an ""artistic device"" for more explicitly indicating the relative depth of a transparent surface and for communicating the essential features of its 3D shape in an intuitively meaningful and minimally occluding way. The driving application for this work is the visualization of layered surfaces in radiation therapy treatment planning data, and the technique is illustrated on transparent isointensity surfaces of radiation dose.We describe the perceptual motivation and artistic inspiration for defining a stroke texture that is locally oriented in the direction of greatest normal curvature (and in which individual strokes are of a length proportional to the magnitude of the curvature in the direction they indicate), and discuss two alternative methods for applying this texture to isointensity surfaces defined in a volume.We propose an experimental paradigm for objectively measuring observers' ability to judge the shape and depth of a layered transparent surface, in the course of a task relevant to the needs of radiotherapy treatment planning, and use this paradigm to evaluate the practical effectiveness of our approach through a controlled observer experiment based on images generated from actual clinical data."
IEEE Trans. Vis. Comput. Graph.	Editorial.		1997	Editorial.
IEEE Trans. Vis. Comput. Graph.	Evaluation and Design of Filters Using a Taylor Series Expansion.	Torsten Möller,Raghu Machiraju,Klaus Mueller,Roni Yagel	1997	We describe a new method for analyzing, classifying, and evaluating filters that can be applied to interpolation filters as well as to arbitrary derivative filters of any order. Our analysis is based on the Taylor series expansion of the convolution sum. Our analysis shows the need and derives the method for the normalization of derivative filter weights. Under certain minimal restrictions of the underlying function, we are able to compute tight absolute error bounds of the reconstruction process. We demonstrate the utilization of our methods to the analysis of the class of cubic BC-spline filters. As our technique is not restricted to interpolation filters, we are able to show that the Catmull-Rom spline filter and its derivative are the most accurate reconstruction and derivative filters, respectively, among the class of BC-spline filters. We also present a new derivative filter which features better spatial accuracy than any derivative BC-spline filter, and is optimal within our framework. We conclude by demonstrating the use of these optimal filters for accurate interpolation and gradient estimation in volume rendering.
IEEE Trans. Vis. Comput. Graph.	Modeling and Visualization for a Pearl-Quality Evaluation Simulator.	Noriko Nagata,Toshimasa Dobashi,Yoshitsugu Manabe,Teruo Usami,Seiji Inokuchi	1997	"Visual simulation using CG and VR has attracted wide attention in the machine vision field. This paper proposes a method of modeling and visualizing pearls that will be the central technique of a pearl-quality evaluation simulator. Pearls manifest a very specific optical phenomenon that is not dependent on the direction of the light source. To investigate this feature, we propose a physical model, called an ""illuminant model,"" for multilayer film interference considering the multiple reflection in spherical bodies. The rendering algorithm has been configured from such representations of physical characteristics as interference, mirroring, and texture, which correspond, respectively, to the sense of depth, brightness, and grain that are the main evaluation factors obtained from psychological experiments. Further, portions of photos of real pearls and the images generated by the present method were evaluated based on a scale of psychological evaluations of ""pearl-like quality,"" demonstrating, thereby, that not merely the generated images as a whole, but the respective parts of images can present such a pearl-like quality."
IEEE Trans. Vis. Comput. Graph.	Breadth-First Ray Tracing Utilizing Uniform Spatial Subdivision.	Koji Nakamaru,Yoshio Ohno	1997	Breadth-first ray tracing is based on the idea of exchanging the roles of rays and objects. For scenes with a large number of objects, it may be profitable to form a set of rays and compare each object in turn against this set. By doing so, thrashing, due to disk access, can be minimized. In this paper, we present ways to combine breadth-first methods with traditional efficient algorithms, along with new schemes to minimize accessing objects stored on disk. Experimental analysis, including comparisons with depth-first ray tracing, shows that large databases can be handled efficiently with this approach.
IEEE Trans. Vis. Comput. Graph.	Incremental Algorithms for Collision Detection Between Polygonal Models.	Madhav K. Ponamgi,Dinesh Manocha,Ming C. Lin	1997	"Fast and accurate collision detection between general polygonal models is a fundamental problem in physically based and geometric modeling, robotics, animation, and computer-simulated environments. Most earlier collision detection algorithms are either restricted to a class of models (such as convex polytopes) or are not fast enough for practical applications. We present an incremental algorithm for collision detection between general polygonal models in dynamic environments. The algorithm combines a hierarchical representation with incremental computation to rapidly detect collisions. It makes use of coherence between successive instances to efficiently determine the number of object features interacting. For each pair of objects, it tracks the closest features between them on their respective convex hulls. It detects the objects' penetration using pseudo internal Voronoi cells and constructs the penetration region, thus identifying the regions of contact on the convex hulls. The features associated with these regions are represented in a precomputed hierarchy. The algorithm uses a coherence based approach to quickly traverse the precomputed hierarchy and check for possible collisions between the features. We highlight its performance on different applications."
IEEE Trans. Vis. Comput. Graph.	Error and Complexity of Random Walk Monte Carlo Radiosity.	Mateu Sbert	1997	In this paper, we study the error and complexity of the discrete random walk Monte Carlo technique for Radiosity, both the shooting and gathering methods. We show that the shooting method exhibits a lower complexity than the gathering one, and under some constraints, it has a linear complexity. This is an improvement over a previous result that pointed to an O(n log n) complexity. We give and compare three unbiased estimators for each method, and obtain closed forms and bounds for their variances. We also bound the expected value of the Mean Square Error (MSE). Some of the results obtained are also shown to be valid for the nondiscrete gathering case. We also give bounds for the variances and MSE for the infinite path length estimators; these bounds might be useful in the study of the biased estimators resulting of cutting off the infinite path.
IEEE Trans. Vis. Comput. Graph.	The Lazy Sweep Ray Casting Algorithm for Rendering Irregular Grids.	Cláudio T. Silva,Joseph S. B. Mitchell	1997	"Lazy Sweep Ray Casting is a fast algorithm for rendering general irregular grids. It is based on the sweep-plane paradigm, and it is able to accelerate ray casting for rendering irregular grids, including disconnected and nonconvex (even with holes) unstructured irregular grids with a rendering cost that decreases as the ""disconnectedness"" decreases. The algorithm is carefully tailored to exploit spatial coherence even if the image resolution differs substantially from the object space resolution.Lazy Sweep Ray Casting has several desirable properties, including its generality, (depth-sorting) accuracy, low memory consumption, speed, simplicity of implementation, and portability (e.g., no hardware dependencies).We establish the practicality of our method through experimental results based on our implementation, which is shown to be substantially faster (by up to two orders of magnitude) than other algorithms implemented in software.We also provide theoretical results, both lower and upper bounds, on the complexity of ray casting of irregular grids."
IEEE Trans. Vis. Comput. Graph.	Tracking and Visualizing Turbulent 3D Features.	Deborah Silver,Xin Wang	1997	"Visualizing 3D time-varying fluid datasets is difficult because of the immense amount of data to be processed and understood. These datasets contain many evolving amorphous regions, and it is difficult to observe patterns and visually follow regions of interest. In this paper, we present a technique which isolates and tracks full volume representations of regions of interest from 3D regular and curvilinear Computational Fluid Dynamics datasets. Connected voxel regions, ""features,"" are extracted from each time step and matched to features in subsequent time steps. Spatial overlap is used to determine matching. The features from each time step are stored in octree forests to speed the matching process. Once features are identified and tracked, properties of the features and their evolutionary history can be computed. This information can be used to enhance isosurface visualization and volume rendering by color coding individual regions. We demonstrate the algorithm on four 3D time-varying simulations from ongoing research in Computational Fluid Dynamics and show how tracking can significantly improve and facilitate the processing of massive datasets."
IEEE Trans. Vis. Comput. Graph.	Fast Display of Illuminated Field Lines.	Detlev Stalling,Malte Zöckler,Hans-Christian Hege	1997	A new technique for interactive vector field visualization using large numbers of properly illuminated field lines is presented. Taking into account ambient, diffuse, and specular reflection terms, as well as transparency and depth cueing, we employ a realistic shading model which significantly increases quality and realism of the resulting images. While many graphics workstations offer hardware support for illuminating surface primitives, usually no means for an accurate shading of line primitives are provided. However, we show that proper illumination of lines can be implemented by exploiting the texture mapping capabilities of modern graphics hardware. In this way, high rendering performance with interactive frame rates can be achieved. We apply the technique to render large numbers of integral curves of a vector field. The impression of the resulting images can be further improved by a number of visual enhancements, like color coding or particle animation. We also describe methods for controlling the distribution of field lines in space. These methods enable us to use illuminated field lines for interactive exploration of vector fields.
IEEE Trans. Vis. Comput. Graph.	Converting Discrete Images to Partitioning Trees.	Kalpathi R. Subramanian,Bruce F. Naylor	1997	The discrete space representation of most scientific datasets (pixels, voxels, etc.), generated through instruments or by sampling continuously defined fields, while being simple, is also verbose and structureless. We propose the use of a particular spatial structure, the binary space partitioning tree, or, simply, partitioning tree, as a new representation to perform efficient geometric computation in discretely defined domains. The ease of performing affine transformations, set operations between objects, and correct implementation of transparency (exploiting the visibility ordering inherent to the representation) makes the partitioning tree a good candidate for probing and analyzing medical reconstructions, in such applications as surgery planning and prostheses design. The multiresolution characteristics of the representation can be exploited to perform such operations at interactive rates by smooth variation of the amount of geometry. Application to ultrasound data segmentation and visualization is proposed.$$\sum\limits^{d \log h}_{i=1} \tilde O\!\!\left({{n}\over{k^3_i}}\right) = \sum\limits^{d \log h}_{i=1}\tilde O\!\!\left({{n}\over{{\left(k/2^i\right)}^3\right)}}=\tilde\!\!{O} (n).$$The paper describes methods for constructing partitioning trees from a discrete image/volume data set. Discrete space operators developed for edge detection are used to locate discontinuities in the image from which lines/planes containing the discontinuities are fitted by using either the Hough transform or a hyperplane sort. A multiresolution representation can be generated by ordering the choice of hyperplanes by the magnitude of the discontinuities. Various approximations can be obtained by pruning the tree according to an error metric. The segmentation of the image into edgeless regions can yield significant data compression. A hierarchical encoding schema for both lossless and lossy encodings is described.
IEEE Trans. Vis. Comput. Graph.	Virtual Data Visualizer.	Ron van Teylingen,William Ribarsky,Charles van der Mast	1997	We present the Virtual Data Visualizer, a highly interactive, immersive environment for visualizing and analyzing data. VDV is a set of tools for exploratory data visualization that does not focus on just one type of application. It employs a data organization with data arranged hierarchically in classes that can be modified by the user within the virtual environment. The class structure is the basis for bindings or mappings between data variables and glyph elements, which the user can make, change, or remove. The binding operation also has a set of defaults so that the user can quickly display the data. The VDV requires a user interface that is fairly complicated for a virtual environment. We have taken the approach that a combination of more-or-less traditional menus and more direct means of icon manipulation will do the job. This work shows that a useful interface and set of tools can be built. Controls in VDV include a panel for controlling animation of the data and zooming in and out. Tools include a workbench for changing the glyphs and setting glyph/variable ranges and a boundary tool for defining new classes spatially.
IEEE Trans. Vis. Comput. Graph.	Out-of-Core Streamline Visualization on Large Unstructured Meshes.	Shyh-Kuang Ueng,Christopher A. Sikorski,Kwan-Liu Ma	1997	"This paper presents an out-of-core approach for interactive streamline construction on large unstructured tetrahedral meshes containing millions of elements. The out-of-core algorithm uses an octree to partition and restructure the raw data into subsets stored into disk files for fast data retrieval. A memory management policy tailored to the streamline calculations is used such that, during the streamline construction, only a very small amount of data are brought into the main memory on demand. By carefully scheduling computation and data fetching, the overhead of reading data from the disk is significantly reduced and good memory performance results. This out-of-core algorithm makes possible interactive streamline visualization of large unstructured-grid data sets on a single mid-range workstation with relatively low main-memory capacity: 5-15 megabytes. We also demonstrate that this approach is much more efficient than relying on virtual memory and operating system's paging algorithms."
IEEE Trans. Vis. Comput. Graph.	A Fast Gibbs Sampler for Synthesizing Constrained Fractals.	Baba C. Vemuri,Chhandomay Mandal,Shang-Hong Lai	1997	It is well known that the spatial frequency spectrum of membrane and thin plate splines exhibit self-affine characteristics and, hence, behave as fractals. This behavior was exploited in generating the constrained fractal surfaces, which were generated by using a Gibbs sampler algorithm in the work of Szeliski and Terzopoulos. The algorithm involves locally perturbing a constrained spline surface with white noise until the spline surface reaches an equilibrium state.In this paper, we introduce a fast generalized Gibbs sampler that combines two novel techniques, namely, a preconditioning technique in a wavelet basis for constraining the splines and a perturbation scheme in which, unlike the traditional Gibbs sampler, all sites (surface nodes) that do not share a common neighbor are updated simultaneously. In addition, we demonstrate the capability to generate arbitrary order fractal surfaces without resorting to blending techniques. Using this fast Gibbs sampler algorithm, we demonstrate the synthesis of realistic terrain models from sparse elevation data.
IEEE Trans. Vis. Comput. Graph.	Editorial.		1997	Editorial.
IEEE Trans. Vis. Comput. Graph.	Adaptive Real-Time Level-of-Detail-Based Rendering for Polygonal Models.	Julie C. Xia,Jihad El-Sana,Amitabh Varshney	1997	We present an algorithm for performing adaptive real-time level-of-detail-based rendering for triangulated polygonal models. The simplifications are dependent on viewing direction, lighting, and visibility and are performed by taking advantage of image-space, object-space, and frame-to-frame coherences. In contrast to the traditional approaches of precomputing a fixed number of level-of-detail representations for a given object, our approach involves statically generating a continuous level-of-detail representation for the object. This representation is then used at run time to guide the selection of appropriate triangles for display. The list of displayed triangles is updated incrementally from one frame to the next. Our approach is more effective than the current level-of-detail-based rendering approaches for most scientific visualization applications, where there are a limited number of highly complex objects that stay relatively close to the viewer. Our approach is applicable for scalar (such as distance from the viewer) as well as vector (such as normal direction) attributes.
CVPR	Real-Time 2-D Feature Detection on a Reconfigurable Computer.	Arrigo Benedetti,Pietro Perona	1998	Real-Time 2-D Feature Detection on a Reconfigurable Computer.
CVPR	Reliable Tracking of Human Arm Dynamics by Multiple Cue Integration and Constraint Fusion.	Yusuf Azoz,Lalitha Devi,Rajeev Sharma	1998	Reliable Tracking of Human Arm Dynamics by Multiple Cue Integration and Constraint Fusion.
CVPR	A Layered Approach to Stereo Reconstruction.	Simon Baker,Richard Szeliski,P. Anandan	1998	A Layered Approach to Stereo Reconstruction.
CVPR	Extraction and Tracking of the Tongue Surface from Ultrasound Image Sequences.	Yusuf Sinan Akgul,Chandra Kambhamettu,Maureen C. Stone	1998	Extraction and Tracking of the Tongue Surface from Ultrasound Image Sequences.
CVPR	Clustering Appearances of 3D Objects.	Ronen Basri,Dan Roth,David W. Jacobs	1998	Clustering Appearances of 3D Objects.
CVPR	Ground from Figure Discrimination.	Arnon Amir,Michael Lindenbaum	1998	Ground from Figure Discrimination.
CVPR	Elliptical Head Tracking Using Intensity Gradients and Color Histograms.	Stan Birchfield	1998	Elliptical Head Tracking Using Intensity Gradients and Color Histograms.
CVPR	Robust Recognition of Scaled Eigenimages through a Hierarchical Approach.	Horst Bischof,Ales Leonardis	1998	Robust Recognition of Scaled Eigenimages through a Hierarchical Approach.
CVPR	Using Hierarchical Shape Models to Spot Keywords in Cursive Handwriting Data.	Michael C. Burl,Pietro Perona	1998	Using Hierarchical Shape Models to Spot Keywords in Cursive Handwriting Data.
CVPR	Action Recognition Using Probabilistic Parsing.	Aaron F. Bobick,Yuri A. Ivanov	1998	Action Recognition Using Probabilistic Parsing.
CVPR	Texture Recognition Using a Non-Parametric Multi-Scale Statistical Model.	Jeremy S. De Bonet,Paul A. Viola	1998	Texture Recognition Using a Non-Parametric Multi-Scale Statistical Model.
CVPR	Hierarchical Organization of Appearance-Based Parts and Relations for Object Recognition.	Octavia I. Camps,Chien-Yuan Huang,Tapas Kanungo	1998	Hierarchical Organization of Appearance-Based Parts and Relations for Object Recognition.
CVPR	Automatic Mosaicing with Super-Resolution Zoom.	David P. Capel,Andrew Zisserman	1998	Automatic Mosaicing with Super-Resolution Zoom.
CVPR	A Methodology for Deriving Probabilistic Correctness Measures from Recognizers.	Djamel Bouchaffra,Venu Govindaraju,Sargur N. Srihari	1998	A Methodology for Deriving Probabilistic Correctness Measures from Recognizers.
CVPR	Remote Reality Demonstration.	Terrance E. Boult	1998	Remote Reality Demonstration.
CVPR	Head Tracking via Robust Registration in Texture Map Images.	Marco La Cascia,John Isidoro,Stan Sclaroff	1998	Abstract A novel method for 3D head tracking in the presence of large head rotations and facial expression changes is described. Tracking is formulated in terms of color image registration in the texture map of a 3D surface model. Model appearance is recursively updated via image mosaicking in the texture map as the head orientation varies. The resulting dynamic texture map provides a stabilized view of the face that can be used as input to many existing 2D techniques for face recognition, facial expressions analysis, lip reading, and eye tracking. Parameters are estimated via a robust minimization procedure; this provides robustness to occlusions, wrinkles, shadows, and specular highlights. The system was tested on a variety of sequences taken with low quality, uncalibrated video cameras. Experimental results are reported.
CVPR	Markov Random Fields with Efficient Approximations.	Yuri Boykov,Olga Veksler,Ramin Zabih	1998	"Markov Random Fields (MRF''s) can be used for a wide variety of vision problems. In this paper we address the estimation of first-order MRF''s with a particular clique potential that resembles a well. We show that the maximum {\em a posteriori} estimate of such an MRF can be obtained by solving a multiway cut problem on a graph. This allows the application of near linear-time algorithms for computing provably good approximations. We formulate the visual correspondence problem as an MRF in our framework, and show that this yields quite promising results on real data with ground truth."
CVPR	A Statistical Framework for Long-Range Feature Matching in Uncalibrated Image Mosaicing.	Tat-Jen Cham,Roberto Cipolla	1998	A Statistical Framework for Long-Range Feature Matching in Uncalibrated Image Mosaicing.
CVPR	Tracking People with Twists and Exponential Maps.	Christoph Bregler,Jitendra Malik	1998	(November 1997)
CVPR	Anomaly Detection through Registration.	Mei Chen,Takeo Kanade,Henry A. Rowley,Dean Pomerleau	1998	Anomaly Detection through Registration.
CVPR	Orientation Space Filtering for Multiple Orientation Line Segmentation.	Jian Chen,Yoshinobu Sato,Shinichi Tamura	1998	The goal of this paper is to present an appropriate method for the segmentation of lines at intersections (X-junctions) and branches (T-junctions), which can be regarded as local regions where lines occur at multiple orientations. A novel representation called ¿orientation space¿ is proposed, which is derived by adding the orientation axis to the abscissa and the ordinate of the image. The orientation space representation is constructed by treating the orientation parameter, to which Gabor filters can be tuned, as a continuous variable. The problem of segmenting lines at multiple orientations is dealt with by thresholding 3D imagesin the orientation space and then detecting the connected components therein. In this way, X-junctions and T-junctions can be separated effectively. Curve grouping can also be accomplished. The segmentation of mathematically modeled X-, T-, and L-junctions is demonstrated and analyzed. The sensitivity limits of the method are also discussed. Experimental results using both synthesized and real images show the method to be effective for junction segmentation and curve grouping.
CVPR	Curves Matching Using Geodesic Paths.	Isaac Cohen,Isabelle Herlin	1998	Curves Matching Using Geodesic Paths.
CVPR	Retrieval of Commercials by Video Semantics.	Carlo Colombo,Alberto Del Bimbo,Pietro Pala	1998	Retrieval of Commercials by Video Semantics.
CVPR	Shape from Equal Thickness Contours.	Ge Cong,Bahram Parvin	1998	Shape from Equal Thickness Contours.
CVPR	Acquisition of a Large Pose-Mosaic Dataset.	Satyan R. Coorg,Neel Master,Seth J. Teller	1998	Acquisition of a Large Pose-Mosaic Dataset.
CVPR	Efficient Optimization of a Deformable Template Using Dynamic Programming.	James M. Coughlan,Alan L. Yuille,Camper English,Daniel Snow	1998	Efficient Optimization of a Deformable Template Using Dynamic Programming.
CVPR	An Optimized Interaction Strategy for Bayesian Relevance Feedback.	Ingemar J. Cox,Matthew L. Miller,Thomas P. Minka,Peter N. Yianilos	1998	An Optimized Interaction Strategy for Bayesian Relevance Feedback.
CVPR	Histogram Model for 3D Textures.	Kristin J. Dana,Shree K. Nayar	1998	Histogram Model for 3D Textures.
CVPR	A Radial Cumulative Similarity Transform for Robust Image Correspondence.	Trevor Darrell	1998	A Radial Cumulative Similarity Transform for Robust Image Correspondence.
CVPR	Integrated Person Tracking Using Stereo, Color, and Pattern Detection.	Trevor Darrell,Gaile G. Gordon,Michael Harville,John Woodfill	1998	We present an approach to real-time person tracking in crowded and/or unknown environments using integration of multiple visual modalities. We combine stereo, color, and face detection modules into a single robust system, and show an initial application in an interactive, face-responsive display. Dense, real-time stereo processing is used to isolate users from other objects and people in the background. Skin-hue classification identifies and tracks likely body parts within the silhouette of a user. Face pattern detection discriminates and localizes the face within the identified body parts. Faces and bodies of users are tracked over several temporal scales: short-term (user stays within the field of view), medium-term (user exits/reenters within minutes), and long term (user returns after hours or days). Short-term tracking is performed using simple region position and size correspondences, while medium and long-term tracking are based on statistics of user appearance. We discuss the failure modes of each individual module, describe our integration method, and report results with the complete system in trials with thousands of users.
CVPR	"Magic Morphin' Mirror: Person Detection and Tracking."	Trevor Darrell,Gaile G. Gordon,Michael Harville,John Woodfill	1998	"Magic Morphin' Mirror: Person Detection and Tracking."
CVPR	On 3-D Surface Reconstruction Using Shape from Shadows.	Michael Daum,Gregory Dudek	1998	On 3-D Surface Reconstruction Using Shape from Shadows.
CVPR	Mosaics of Scenes with Moving Objects.	James Davis	1998	Mosaics of Scenes with Moving Objects.
CVPR	Fitting Undeformed Superquadrics to Range Data: Improving Model Recovery and Classification.	Erik R. van Dop,Paul P. L. Regtien	1998	Fitting Undeformed Superquadrics to Range Data: Improving Model Recovery and Classification.
CVPR	Bagging in Computer Vision.	Bruce A. Draper,Kyungim Baek	1998	Bagging in Computer Vision.
CVPR	Motion Feature Detection Using Steerable Flow Fields.	David J. Fleet,Michael J. Black,Allan D. Jepson	1998	Motion Feature Detection Using Steerable Flow Fields.
CVPR	Image Editing in the Contour Domain.	James H. Elder,Richard M. Goldberg	1998	Image editing systems are essentially pixel-based. In this paper, we propose a novel method for image editing in which the primitive working unit is not a pixel but an edge. The feasibility of this proposal is suggested by recent work showing that a gray-scale image can be accurately represented by its edge map if a suitable edge model and scale selection method are employed [1]. In particular, an efficient algorithm has been reported to invert such an edge representation to yield a high-fidelity reconstruction of the original image [2], [3]. We have combined these algorithms together with an efficient method for contour grouping and an intuitive user interface to allow users to perform image editing operations (crop, paste, delete) directly in the contour domain. Experimental results suggest that this novel combination of vision algorithms may increase the efficiency of certain classes of image editing operations.
CVPR	Background Modeling for Segmentation of Video-Rate Stereo Sequences.	Christopher K. Eveland,Kurt Konolige,Robert C. Bolles	1998	Background Modeling for Segmentation of Video-Rate Stereo Sequences.
CVPR	Mixtures of Local Linear Subspaces for Face Recognition.	Brendan J. Frey,Antonio Colmenarez,Thomas S. Huang	1998	Mixtures of Local Linear Subspaces for Face Recognition.
CVPR	A Theory of Selection for Gamut Mapping Color Constancy.	Graham D. Finlayson,Steven D. Hordley	1998	A Theory of Selection for Gamut Mapping Color Constancy.
CVPR	Image Segmentation Using Local Variation.	Pedro F. Felzenszwalb,Daniel P. Huttenlocher	1998	Image Segmentation Using Local Variation.
CVPR	Scene Segmentation from 3D Motion.	Xiaolin Feng,Pietro Perona	1998	Scene Segmentation from 3D Motion.
CVPR	VADIS: A Video Analysis, Display and Indexing System.	Ullas Gargi,Sameer Antani,Rangachar Kasturi	1998	VADIS: A Video Analysis, Display and Indexing System.
CVPR	Performance Characterization and Comparison of Video Indexing Algorithms.	Ullas Gargi,Rangachar Kasturi,Sameer Antani	1998	Performance Characterization and Comparison of Video Indexing Algorithms.
CVPR	Automatic Hierarchical Classification of Silhouettes of 3D Objects.	Yoram Gdalyahu,Daphna Weinshall	1998	Automatic Hierarchical Classification of Silhouettes of 3D Objects.
CVPR	Salient and Multiple Illusory Surfaces.	Davi Geiger,Hsing-Kuo Pao,Nava Rubin	1998	Salient and Multiple Illusory Surfaces.
CVPR	Illumination Cones for Recognition under Variable Lighting: Faces.	Athinodoros S. Georghiades,David J. Kriegman,Peter N. Belhumeur	1998	Illumination Cones for Recognition under Variable Lighting: Faces.
CVPR	Image Indexing and Retrieval Based on Human Perceptual Color Clustering.	Yihong Gong,Guido Proietti,Christos Faloutsos	1998	Image Indexing and Retrieval Based on Human Perceptual Color Clustering.
CVPR	Face Recognition Based on Nearest Linear Combinations.	Stan Z. Li	1998	Face Recognition Based on Nearest Linear Combinations.
CVPR	W4: A Real Time System for Detecting and Tracking People.	Ismail Haritaoglu,David Harwood,Larry S. Davis	1998	W4: A Real Time System for Detecting and Tracking People.
CVPR	The Sample Tree: A Sequential Hypothesis Testing Approach to 3D Object Recognition.	Michael A. Greenspan	1998	The Sample Tree: A Sequential Hypothesis Testing Approach to 3D Object Recognition.
CVPR	Using Adaptive Tracking to Classify and Monitor Activities in a Site.	W. Eric L. Grimson,Chris Stauffer,R. Romano,L. Lee	1998	Using Adaptive Tracking to Classify and Monitor Activities in a Site.
CVPR	Segmenting 3-D Surfaces Using Multicolored Illumination.	Glenn Healey,Lizhi Wang	1998	Segmenting 3-D Surfaces Using Multicolored Illumination.
CVPR	Segmentation by Grouping Junctions.	Hiroshi Ishikawa,Davi Geiger	1998	Segmentation by Grouping Junctions.
CVPR	Generation of the Euclidean Skeleton from the Vector Distance Map by a Bisector Decision Rule.	Hong Li,Albert M. Vossepoel	1998	Generation of the Euclidean Skeleton from the Vector Distance Map by a Bisector Decision Rule.
CVPR	Depth Measurement by the Multi-Focus Camera.	Shinsaku Hiura,Takashi Matsuyama	1998	Depth Measurement by the Multi-Focus Camera.
CVPR	Comparing Images under Variable Illumination.	David W. Jacobs,Peter N. Belhumeur,Ronen Basri	1998	Comparing Images under Variable Illumination.
CVPR	From Parametric Warping to the Cooperation of Local Features and Global Models.	Thomas Jelonek,Frank P. Ferrie	1998	From Parametric Warping to the Cooperation of Local Features and Global Models.
CVPR	Fuzzy Convergence.	Adam Hoover,Michael H. Goldbaum	1998	Fuzzy Convergence.
CVPR	The Acquisition and Use of Interaction Behavior Models.	Neil Johnson,Aphrodite Galata,David Hogg	1998	The Acquisition and Use of Interaction Behavior Models.
CVPR	Efficient Multiple Model Recognition in Cluttered 3-D Scenes.	Andrew Edie Johnson,Martial Hebert	1998	Efficient Multiple Model Recognition in Cluttered 3-D Scenes.
CVPR	Computer Modeling, Analysis and Synthesis of Dressed Humans.	Nebojsa Jojic,Jin Gu,Ivan Mak,Helen C. Shen,Thomas S. Huang	1998	Computer Modeling, Analysis and Synthesis of Dressed Humans.
CVPR	Image Segmentation Based on the Integration of Pixel Affinity and Deformable Models.	Timothy N. Jones,Dimitris N. Metaxas	1998	Image Segmentation Based on the Integration of Pixel Affinity and Deformable Models.
CVPR	Hierarchical Morphable Models.	Michael J. Jones,Tomaso Poggio	1998	Hierarchical Morphable Models.
CVPR	Fuzzy Relational Distance for Large-Scale Object Recognition.	Benoit Huet,Edwin R. Hancock	1998	Fuzzy Relational Distance for Large-Scale Object Recognition.
CVPR	Fingerprint Preselection Using Eigenfeatures.	Toshio Kamei,Masanori Mizoguchi	1998	Fingerprint Preselection Using Eigenfeatures.
CVPR	Video Scene Segmentation via Continuous Video Coherence.	John R. Kender,Boon-Lock Yeo	1998	Video Scene Segmentation via Continuous Video Coherence.
CVPR	Correspondence between Different View Breast X-Rays Using a Simulation of Breast Deformation.	Yasuyo Kita,Ralph Highnam,Michael Brady	1998	Correspondence between Different View Breast X-Rays Using a Simulation of Breast Deformation.
CVPR	A Weighted Combination of Classifiers Employing Shared and Distinct Representations.	Josef Kittler,S. A. Hojjatoleslami	1998	A Weighted Combination of Classifiers Employing Shared and Distinct Representations.
CVPR	Model-Based Multiscale Detection of 3D Vessels.	Karl Krissian,Grégoire Malandain,Nicholas Ayache,Régis Vaillant,Yves Trousset	1998	Model-Based Multiscale Detection of 3D Vessels.
CVPR	Direct Shape from Texture Using a Parametric Surface Model and an Adaptive Filtering Technique.	Kyoung Mu Lee,C. C. Jay Kuo	1998	Direct Shape from Texture Using a Parametric Surface Model and an Adaptive Filtering Technique.
CVPR	Inferring Segmented Surface Description from Stereo Data.	Mi-Suen Lee,Gérard G. Medioni	1998	Inferring Segmented Surface Description from Stereo Data.
CVPR	Probablistic Affine Invariants for Recognition.	Thomas K. Leung,Michael C. Burl,Pietro Perona	1998	Probablistic Affine Invariants for Recognition.
CVPR	Creaseness Measures for CT and MR Image Registration.	Antonio M. López,Joan Serrat	1998	Creaseness Measures for CT and MR Image Registration.
CVPR	Metric Rectification for Perspective Images of Planes.	David Liebowitz,Andrew Zisserman	1998	Metric Rectification for Perspective Images of Planes.
CVPR	Subtly Different Facial Expression Recognition and Expression Intensity Estimation.	James Jenn-Jier Lien,Takeo Kanade,Jeffrey F. Cohn,Ching-Chung Li	1998	Subtly Different Facial Expression Recognition and Expression Intensity Estimation.
CVPR	Dense Shape and Motion from Region Correspondences by Factorization.	Jianbo Ma,Narendra Ahuja	1998	Dense Shape and Motion from Region Correspondences by Factorization.
CVPR	A Classification Based Similarity Metric for 3D Image Retrieval .	Yanxi Liu,Frank Dellaert	1998	A Classification Based Similarity Metric for 3D Image Retrieval .
CVPR	Towards the Computational Perception of Action.	Richard Mann,Allan D. Jepson	1998	Towards the Computational Perception of Action.
CVPR	Probabilistic Reasoning Models for Face Recognition.	Chengjun Liu,Harry Wechsler	1998	Probabilistic Reasoning Models for Face Recognition.
CVPR	Soccer Image Sequence Computed by a Virtual Camera.	Kentaro Matsui,Masaki Iwase,Masato Agata,Toshimitsu Tanaka,Noboru Ohnishi	1998	Soccer Image Sequence Computed by a Virtual Camera.
CVPR	2D Fluid Motion Analysis from a Single Image.	Mariette Maurizot,Patrick Bouthemy,Bernard Delyon	1998	2D Fluid Motion Analysis from a Single Image.
CVPR	Exhaustive Detection of Manufacturing Flaws as Abnormalities.	Van-Duc Nguyen,J. Alison Noble,Joseph L. Mundy,John Janning,Joseph Ross	1998	Exhaustive Detection of Manufacturing Flaws as Abnormalities.
CVPR	Singularity Analysis for Articulated Object Tracking.	Daniel D. Morris,James M. Rehg	1998	Singularity Analysis for Articulated Object Tracking.
CVPR	Edge/Region-Based Segmentation and Reconstruction of Underwater Acoustic Images by Markov Random Fields.	Vittorio Murino,Andrea Trucco	1998	Edge/Region-Based Segmentation and Reconstruction of Underwater Acoustic Images by Markov Random Fields.
CVPR	Weak Orthogonalization of Face and Perturbation for Recognition.	Kenji Nagao,Masaki Sohma	1998	Weak Orthogonalization of Face and Perturbation for Recognition.
CVPR	Efficient Query Refinement for Image Retrieval.	Chahab Nastar,Matthias Mitschke,Christophe Meilhac	1998	Efficient Query Refinement for Image Retrieval.
CVPR	Computing the Camera Heading from Multiple Frames.	John Oliensis	1998	Computing the Camera Heading from Multiple Frames.
CVPR	Variable-Scale Smoothing and Edge Detection Guided by Stereoscopy.	Clark F. Olson	1998	Variable-Scale Smoothing and Edge Detection Guided by Stereoscopy.
CVPR	A Probabilistic Formulation for Hausdorff Matching.	Clark F. Olson	1998	A Probabilistic Formulation for Hausdorff Matching.
CVPR	Recognizing Abruptly Changing Facial Expressions from Time-Sequential Face Images.	Takahiro Otsuka,Jun Ohya	1998	Recognizing Abruptly Changing Facial Expressions from Time-Sequential Face Images.
CVPR	Camera Calibration and Euclidean Reconstruction from Known Observer Translations.	Tomás Pajdla,Václav Hlavác	1998	Camera Calibration and Euclidean Reconstruction from Known Observer Translations.
CVPR	A Content-Based Search Engine for VRML Databases.	Eric Paquet,Marc Rioux	1998	A Content-Based Search Engine for VRML Databases.
CVPR	A New Linear Method for Euclidean Motion/Structure from Three Calibrated Affine Views.	Long Quan,Yuichi Ohta	1998	A New Linear Method for Euclidean Motion/Structure from Three Calibrated Affine Views.
CVPR	Stereo and Color Analysis for Dynamic Obstacle Avoidance.	Tamer F. Rabie,Demetri Terzopoulos	1998	Stereo and Color Analysis for Dynamic Obstacle Avoidance.
CVPR	Real-Time Tracking for Enhanced Tennis Broadcasts.	Gopal Sarma Pingali,Yves Jean,Ingrid Carlbom	1998	Real-Time Tracking for Enhanced Tennis Broadcasts.
CVPR	Human Action Detection Using PNF Propagation of Temporal Constraints.	Claudio S. Pinhanez,Aaron F. Bobick	1998	Human Action Detection Using PNF Propagation of Temporal Constraints.
CVPR	Joint Probabilistic Techniques for Tracking Multi-Part Objects.	Christopher Rasmussen,Gregory D. Hager	1998	Joint Probabilistic Techniques for Tracking Multi-Part Objects.
CVPR	Object Detection and Localization by Dynamic Template Warping.	Aparna Lakshmi Ratan,W. Eric L. Grimson,William M. Wells III	1998	"A simple method is presented for detecting, localizing and recognizing instances of classes of objects, while accommodating a wide variation in an object's pose. The method utilizes a small two-dimensional template that is warped into an image, and converts localization to a one-dimensional sub-problem, with the search for a match between image and template executed by dynamic programming. For roughly cylindrical objects (like heads), the method recovers three of the six degrees of freedom of motion (2 translation, 1 rotation), and accommodates two more degrees of freedom in the search process (1 rotation, 1 translation). Experiments demonstrate that the method provides an efficient search strategy that outperforms normalized correlation. This is demonstrated in the example domain of face detection and localization, and can extended to more general detection tasks. An additional technique recovers rough object pose from the match results, and is used in a two stage recognition experiment in conjunction with maximization of mutual information."
CVPR	Well-Behaved, Tunable 3D-Affine Invariants.	Isidore Rigoutsos	1998	Well-Behaved, Tunable 3D-Affine Invariants.
CVPR	A Verification Protocol and Statistical Performance Analysis for Face Recognition Algorithms.	Syed A. Rizvi,P. Jonathon Phillips,Hyeonjoon Moon	1998	A Verification Protocol and Statistical Performance Analysis for Face Recognition Algorithms.
CVPR	Exploitation of Natural Image Statistics by Biological Vision Systems: 1/f2 Power Spectra and Self-Similar Bandpass Decompositions.	Florian Röhrbein,Christoph Zetzsche	1998	Exploitation of Natural Image Statistics by Biological Vision Systems: 1/f2 Power Spectra and Self-Similar Bandpass Decompositions.
CVPR	Varying Focal Length Self-Calibration and Pose Estimation from Two Images .	Benny Rousso,Erez Shilat	1998	Varying Focal Length Self-Calibration and Pose Estimation from Two Images .
CVPR	Rotation Invariant Neural Network-Based Face Detection.	Henry A. Rowley,Shumeet Baluja,Takeo Kanade	1998	Rotation Invariant Neural Network-Based Face Detection.
CVPR	Rotation Invariant Neural Network-Based Face Detection.	Henry A. Rowley,Shumeet Baluja,Takeo Kanade	1998	Rotation Invariant Neural Network-Based Face Detection.
CVPR	Projective Translations and Affine Stereo Calibration.	Andreas Ruf,Gabriela Csurka,Radu Horaud	1998	Projective Translations and Affine Stereo Calibration.
CVPR	Moment Invariants and Quantization Effects.	Gouda I. Salama,A. Lynn Abbott	1998	Moment Invariants and Quantization Effects.
CVPR	Incorporating Illumination Constraints in Deformable Models.	Dimitris Samaras,Dimitris N. Metaxas	1998	Incorporating Illumination Constraints in Deformable Models.
CVPR	Robust Defect Segmentation in Woven Fabrics.	Hamed Sari-Sarraf,James S. Goddard Jr.	1998	Robust Defect Segmentation in Woven Fabrics.
CVPR	Learning to Form Large Groups of Salient Image Features.	Sudeep Sarkar	1998	Learning to Form Large Groups of Salient Image Features.
CVPR	Probabilistic Modeling of Local Appearance and Spatial Relationships for Object Recognition.	Henry Schneiderman,Takeo Kanade	1998	Probabilistic Modeling of Local Appearance and Spatial Relationships for Object Recognition.
CVPR	Minimax Entropy and Learning by Diffusion.	Jayant Shah	1998	Minimax Entropy and Learning by Diffusion.
CVPR	Multiple Feature Integration for Robust Object Localization.	Shishir Shah,J. K. Aggarwal	1998	Multiple Feature Integration for Robust Object Localization.
CVPR	An Objective Comparison Methodology of Edge Detection Algorithms Using a Structure from Motion Task.	Min C. Shin,Dmitry B. Goldgof,Kevin W. Bowyer	1998	An Objective Comparison Methodology of Edge Detection Algorithms Using a Structure from Motion Task.
CVPR	Interactive Construction of 3D Models from Panoramic Mosaics.	Heung-Yeung Shum,Mei Han,Richard Szeliski	1998	Interactive Construction of 3D Models from Panoramic Mosaics.
CVPR	What is the Spectral Dimensionality of Illumination Functions in Outdoor Scenes?	David Slater,Glenn Healey	1998	What is the Spectral Dimensionality of Illumination Functions in Outdoor Scenes?
CVPR	Optimal Structure from Motion: Local Ambiguities and Global Estimates.	Stefano Soatto,Roger W. Brockett	1998	Optimal Structure from Motion: Local Ambiguities and Global Estimates.
CVPR	Interactive Sensor Planning.	Ioannis Stamos,Peter K. Allen	1998	Interactive Sensor Planning.
CVPR	Direct Estimation of Motion and Extended Scene Structure from a Moving Stereo Rig.	Gideon P. Stein,Amnon Shashua	1998	We describe a new method for motion estimation and 3D reconstruction from stereo image sequences obtained by a stereo rig moving through a rigid world. We show that given two stereo pairs one can compute the motion of the stereo rig directly from the image derivatives (spatial and temporal). Correspondences are not required. One can then use the images from both pairs combined to compute a dense depth map. The motion estimates between stereo pairs enable us to combine depth maps from all the pairs in the sequence to form an extended scene reconstruction and we show results from a real image sequence. The motion computation is a linear least squares computation using all the pixels in the image. Areas with little or no contrast are implicitly weighted less so one does not have to explicitly apply a confidence measure.
CVPR	Analyzing the Bidirectional Texture Function.	Pei-hsiu Suen,Glenn Healey	1998	Analyzing the Bidirectional Texture Function.
CVPR	Connected Vibrations: A Modal Analysis Approach for Non-Rigid Motion Tracking.	Hai Tao,Thomas S. Huang	1998	Connected Vibrations: A Modal Analysis Approach for Non-Rigid Motion Tracking.
CVPR	A New Complex Basis for Implicit Polynomial Curves and its Simple Exploitation for Pose Estimation and Invariant Recognition.	Jean-Philippe Tarel,David B. Cooper	1998	A New Complex Basis for Implicit Polynomial Curves and its Simple Exploitation for Pose Estimation and Invariant Recognition.
CVPR	Variants of Dynamic Link Architecture Based on Mathematical Morphology for Frontal Face Authentication.	Anastasios Tefas,Constantine Kotropoulos,Ioannis Pitas	1998	Variants of Dynamic Link Architecture Based on Mathematical Morphology for Frontal Face Authentication.
CVPR	Making Good Features Track Better.	Tiziano Tommasini,Andrea Fusiello,Emanuele Trucco,Vito Roberto	1998	Making Good Features Track Better.
CVPR	Nonrigid Motion Analysis Based on Dynamic Refinement of Finite Element Models.	Leonid V. Tsap,Dmitry B. Goldgof,Sudeep Sarkar	1998	"In this paper, we propose new algorithms for accurate nonrigid motion tracking. Given an initial model representing general knowledge of the object, a set of sparse correspondences, and incomplete or missing information about geometry or material properties, we can recover dense motion vectors using finite element models. The method is based on the iterative analysis of the differences between the actual and predicted behavior. Large differences indicate that an object's properties are not captured properly by the model describing it. Feedback from the images during the motion allows the refinement of the model by minimizing the error between the expected and true position of the object's points. These errors are due to flaws in the model parameter estimation such as geometry and material properties. Unknown parameters are recovered using an iterative descent search for the best nonlinear finite element model that approximates nonrigid motion of the given object. During this search process, we not only estimate material properties, but also infer dense point correspondences from our initial set of sparse correspondences. Thus, during tracking, the model is refined which, in turn, improves tracking quality. As a result, we obtain a more precise description of nonrigid motion. Experimental results demonstrate the success of the proposed algorithm. The method was applied to man-made elastic materials and human skin to recover unknown elasticity, to complex 3D objects to find details of their geometry, and to a hand motion analysis application. Our work demonstrates the possibility of accurate quantitative analysis of nonrigid motion in range image sequences with objects consisting of multiple materials and 3D volumes."
CVPR	Image Segmentation for Human Tracking Using Sequential-Image-Based Hierarchical Adaptation.	Akira Utsumi,Jun Ohya	1998	Image Segmentation for Human Tracking Using Sequential-Image-Based Hierarchical Adaptation.
CVPR	A Spatiotemporal Motion Model for Video Summarization.	Nuno Vasconcelos,Andrew Lippman	1998	A Spatiotemporal Motion Model for Video Summarization.
CVPR	A Bayesian Framework for Semantic Content Characterization.	Nuno Vasconcelos,Andrew Lippman	1998	A Bayesian Framework for Semantic Content Characterization.
CVPR	A Coupled PDE Model of Nonlinear Diffusion for Image Smoothing and Segmentation.	Baba C. Vemuri,Li Wang,Yunmei Chen	1998	A Coupled PDE Model of Nonlinear Diffusion for Image Smoothing and Segmentation.
CVPR	Appearance Based Behavior Recognition by Event Driven Selective Attention.	Toshikazu Wada,Takashi Matsuyama	1998	Appearance Based Behavior Recognition by Event Driven Selective Attention.
CVPR	Boundary Finding with Correspondence Using Statistical Shape Models.	Yongmei Wang,Lawrence H. Staib	1998	Boundary Finding with Correspondence Using Statistical Shape Models.
CVPR	A Specialized Multibaseline Stereo Technique for Obstacle Detection.	Todd Williamson,Charles E. Thorpe	1998	A Specialized Multibaseline Stereo Technique for Obstacle Detection.
CVPR	Nonlinear PHMMs for the Interpretation of Parameterized Gesture.	Andrew D. Wilson,Aaron F. Bobick	1998	Nonlinear PHMMs for the Interpretation of Parameterized Gesture.
CVPR	Image Alignment for Precise Camera Fixation and Aim.	Lambert E. Wixson,Jayakrishnan Eledath,Michael W. Hansen,Robert Mandelbaum,Deepam Mishra	1998	Image Alignment for Precise Camera Fixation and Aim.
CVPR	Image Understanding from Thermal Emission Polarization.	Lawrence B. Wolff,Andrew Lundberg,Renjie Tang	1998	Image Understanding from Thermal Emission Polarization.
CVPR	Incremental Tracking of Human Actions from Multiple Views.	Masanobu Yamamoto,Akitsugu Sato,Satoshi Kawada,Takuya Kondo,Yoshihiko Osaki	1998	Incremental Tracking of Human Actions from Multiple Views.
CVPR	3D Object Depth Recovery from Highlights Using Active Sensor and Illumination Control.	Xilin Yi,Octavia I. Camps	1998	3D Object Depth Recovery from Highlights Using Active Sensor and Illumination Control.
CVPR	Extraction and Classification of Visual Motion Patterns for Hand Gesture Recognition.	Ming-Hsuan Yang,Narendra Ahuja	1998	Extraction and Classification of Visual Motion Patterns for Hand Gesture Recognition.
CVPR	Stochastic Computation of Medial Axis in Markov Random Fields.	Song Chun Zhu	1998	Stochastic Computation of Medial Axis in Markov Random Fields.
CVPR	Rotated Wedge Averaging Method for Junction Classification.	Weichuan Yu,Konstantinos Daniilidis,Gerald Sommer	1998	Rotated Wedge Averaging Method for Junction Classification.
CVPR	Model-Based Target Recognition in Pulsed Ladar Imagery.	Qinfen Zheng,Sandor Z. Der,Rama Chellappa	1998	Model-Based Target Recognition in Pulsed Ladar Imagery.
CVPR	Qualitative and Quantitative Car Tracking from a Range Image Sequence.	Liang Zhao,Chuck Thorpe	1998	Qualitative and Quantitative Car Tracking from a Range Image Sequence.
CVPR	Volumetric Layer Segmentation Using Coupled Surfaces Propagation.	Xiaolan Zeng,Lawrence H. Staib,Robert T. Schultz,James S. Duncan	1998	Volumetric Layer Segmentation Using Coupled Surfaces Propagation.
CVPR	Empirical Performance Analysis of Linear Discriminant Classifiers.	Wen-Yi Zhao,Rama Chellappa,N. Nandhakumar	1998	Empirical Performance Analysis of Linear Discriminant Classifiers.
CVPR	"1998 Conference on Computer Vision and Pattern Recognition (CVPR '98), June 23-25, 1998, Santa Barbara, CA, USA"		1998	"1998 Conference on Computer Vision and Pattern Recognition (CVPR '98), June 23-25, 1998, Santa Barbara, CA, USA"
ICCV	Self-calibrating a Stereo Head: An Error Analysis in the Neighbourhood of Degenerate Configurations.	Lourdes de Agapito,D. Q. Huynh,Michael J. Brooks	1998	We show that the self-calibration of a stereo head from corresponding points in an image pair is in certain circumstances prone to considerable error. A novel error analysis reveals that the automated determination of relative orientation and focal length is adversely affected when the cameras verge inwards a similar amount, and when the principal point locations have a horizontal error. This analysis is facilitated by the adoption of closed-form solutions for self-calibration from previous work of the authors. It is also shown that estimation of the fundamental matrix associated with a stereo head image pair is improved when a domain-specific parameterization and associated computational techniques are adopted. Experiments conducted with such image pairs suggest that, given cognisance of sensitive configurations and adoption of the revised method of fundamental matrix estimation, robust reconstructions are attainable. This is demonstrated on the problem of metrically reconstructing a scene from two pairs of images obtained by an uncalibrated stereo head undergoing unknown ground-plane motion.
ICCV	Depth Discontinuities by Pixel-to-Pixel Stereo.	Stan Birchfield,Carlo Tomasi	1998	An algorithm to detect depth discontinuities from a stereo pair of images is presented. The algorithm matches individual pixels in corresponding scanline pairs while allowing occluded pixels to remain unmatched, then propagates the information between scanlines by means of a fast postprocessor. The algorithm handles large untextured regions, uses a measure of pixel dissimilarity that is insensitive to image sampling, and prunes bad search nodes to increase the speed of dynamic programming. The computation is relatively fast, taking about 1.5 microseconds per pixel per disparityon a workstation. Approximate disparity maps and precise depth discontinuities (along both horizontal and vertical boundaries) are shown for five stereo images containing textured, untextured, fronto-parallel, and slanted objects.
ICCV	A Framework for Modeling Appearance Change in Image Sequences.	Michael J. Black,David J. Fleet,Yaser Yacoob	1998	"Image ""appearance"" may change over time due to a variety of causes such as 1) object or camera motion; 2) generic photometric events including variations in illumination (e.g. shadows) and specular reflections; and 3) ""iconic changes"" which are specific to the objects being viewed and include complex occlusion events and changes in the material properties ofthe objects. We propose a general framework for representing and recovering these ""appearance changes"" in an image sequence as a ""mixture"" of different causes. The approach generalizes previous work on optical flow to provide a richer description of image events and more reliable estimates of image motion."
ICCV	Color- and Texture-based Image Segmentation Using the Expectation-Maximization Algorithm and its Application to Content-Based Image Retrieval.	Serge Belongie,Chad Carson,Hayit Greenspan,Jitendra Malik	1998	Color- and Texture-based Image Segmentation Using the Expectation-Maximization Algorithm and its Application to Content-Based Image Retrieval.
ICCV	From Projective to Euclidean Space Under any Practical Situation, a Criticism of Self-Calibration.	Sylvain Bougnoux	1998	For many practical applications it is important to relax the self-calibration conditions to allow for changing internal camera parameters (e.g. zooming/focusing ¿). Classical techniques failed for such conditions. We present the available constraints that allow us to right a projective calibration to a Euclidean one. Meanwhile, we found that the estimations of the internal parameters were rather inaccurate. We discuss theoretically this difficulty and above all the resulting effect on the 3D reconstruction. In fact, we show that the uncertainty on the focal length estimation leads to a Euclidean calibration up to a quasi anisotropic homothety whereas the error on the principal point can often be interpreted as a translation. Hopefully, the calibration we come up with, is quite acceptable for reconstruction of models.
ICCV	Segmentation of Range Data into Rigid Subsets Using Surface Patches.	Anthony Ashbrook,Robert B. Fisher	1998	Segmentation of Range Data into Rigid Subsets Using Surface Patches.
ICCV	3D Photography on Your Desk.	Jean-Yves Bouguet,Pietro Perona	1998	"A simple and inexpenssive approach for extracting the three-dimentional shape of objects is presented. It is based on 'weak structured lighting'; it differs from other conventional structured lighting approaches in that it requires very little hardware besides the camera: a desk-lamp, a pencil and a checkerboard. The camera faces the object, which is illuminated by the desk-lamp. The user moves a pencil in front of the light source casting a moving shadow on the object. The 3D shape of the object is extracted from the spatial and temporal location of the observed shadow. Experimental results are presented on three different scenes demonstrating that the error in reconstructing the surface is less than 1%."
ICCV	A Theory of Catadioptric Image Formation.	Simon Baker,Shree K. Nayar	1998	Conventional video cameras have limited fields of view which make them restrictive for certain applications in computational vision. A catadioptric sensor uses a combination of lenses and mirrors placed in a carefully arranged configuration to capture a much wider field of view. When designing a catadioptric sensor, the shape of the mirror(s) should ideally be selected to ensure that the complete catadioptric system has a single effective viewpoint. In this paper, we derive the complete class of single-lens single-mirror catadioptric sensors which have a single viewpoint and an expression for the spatial resolution of a catadioptric sensor in terms of the resolution of the camera used to construct it. We also include a preliminary analysis of thedefocus blur caused by the use of a curved mirror.
ICCV	Resolution-Appropriate Shape Representation.	Bernard Baldwin,Davi Geiger,Robert A. Hummel	1998	"We present a new type of ""snake"" in which the dimensionality of the shapes is scaled appropriately for the resolution of the images in which the shapes are embedded. We define shapes as an ordered list of control points and compute the principal components of the shapes in a prior training set. Our energy function is based upon the Mahalanobis distance of a given shape from the mean shape and on the Mahalanobis distance of the image attributes from image attribute values extracted from the training set. We show that the derivative of this energy function with respect to the modal weights is reduced as the image resolution is reduced, and that the derivative of the energy scales with the variance associated with each mode. We exploit this property to determine the subset of the modes which are relevant at a particular level of image resolution, thereby reducing the dimensionality of the shapes. We implement a coarse-to-fine search procedure in the image and shape domains simultaneously, and demonstrate this procedure on the identification of anatomic structures in Computed Tomography images."
ICCV	Contagion-Based Image Segmentation and Labeling.	Amit Banerjee,Philippe Burlina,Fady Alajaji	1998	Contagion-Based Image Segmentation and Labeling.
ICCV	Self-Calibration from Image Derivatives.	Tomás Brodský,Cornelia Fermüller,Yiannis Aloimonos	1998	This study investigates the problem of estimating the calibration parameters from image motion fields induced by a rigidly moving camera with unknown calibration parameters, where the image formation is modeled with a linear pinhole-camera model. The equations obtained show the flow to be clearly separated into a component due to the rotation and the calibration parameters. A set of parameters encoding the latter component are linearly related to the flow, and from these parameters the calibration can be determined. However, as for discrete motion, in the general case it is not possible, to decouple image measurements from two frames only into their translational and rotational component.Geometrically, the ambiguity takes the form of a part of the rotational component being parallel to the translational component, and thus the scene can be reconstructed only up to a projective transformation. In general, for a full calibration at least four successiveimage frames are necessary with the 3D-rotation changing between the measurements.The geometric analysis gives rise to a direct self-calibration method that avoids computation of optical flow or point correspondences and uses only normal flow measurements. In this technique the direction of translation is estimated employing in a novel way smoothness constraints. Then the calibration parameters are estimated from the rotational components of several flow fields using levenberg-Marquardt parameter estimation, iterative in the calibration parameters only. The technique proposed does not require calibration objects in the scene or special camera motions and it also avoids the computation of exact correspondence. This makes it suitable for the calibration of active vision systems which have to acquire knowledge about their intrinsic parameters while they perform other tasks, or as a tool for analyzing image sequences in large video databases.
ICCV	Model Selection and Surface Merging in Reconstruction Algorithms.	Kishore Bubna,Charles V. Stewart	1998	The problem of model selection is relevant to many areas of computer vision. Model selection criteria have been used in the vision literature and many more have been proposed in statistics, but the relative strengths of these criteria have not been analyzed in vision. More importantly, suitable extensions to these criteria must be made to solve problems unique to computer vision. Using the problem of surface reconstruction as our context, we analyze existing criteria using simulations and sensor data, introduce new criteria from statistics, develop novel criteria capable of handling unknown error distributions and outliers, and extendmodel selection criteria to apply to the surface merging problem. The new surface merging rules improve upon previous results, and work well even at small step heights (h = 3¿) and crease discontinuities. Our results show that a Bayesian criteria and its bootstrapped variant perform the best, although for time-sensitive applications, a variant of the Akaike criterion may be a better choice. Unfortunately, none of the criteria work reliably for small region sizes, implying that model selection and surface merging should be avoided unless the region size is sufficiently large.
ICCV	Separability of Pose and Expression in Facial Tracing and Animation.	Benedicte Bascle,Andrew Blake	1998	"We explore the application of facial tracking to automated re-animation. To this end, it is necessary to recover both head-pose and facial expression from the facial movement of a performer. However, both effects are coupled. This is a serious problem, which previous studies haven't fully considered. The solution to this interaction problem proposed here is to solve explicitly, at each timestep, for pose and expression variables. In principle this is a nonlinear inverse problem. However, appropriate parameterisation of pose in terms of affine transformations with parallax, and of expression in terms of key-frames, reduces the problem to a bilinear one. This can then be solved directly by Singular Value Decomposition. Thus actor-driven animation has b een implemented in real-time, at video field-rate, using two Indy desktop workstations."
ICCV	Color Recognition in Outdoor Images.	Shashi D. Buluswar,Bruce A. Draper	1998	"The color associated with an object in machine vision images is not constant; under varying illuminating and viewing conditions (such as in outdoor images), the perceived color of an object can vary significantly, thus making color-based recognition difficult. Existing methods in color-based recognition have been applied mostly to indoor and/or constrained imagery, but not to realistic outdoor data.This work analyzes the variation of object color in outdoor images with respect to existing models of day-light illumination and surface reflectance. Two approaches for color recognition are then proposed: the first develops context-based models of daylight illumination and hybrid surface reflectance, and predicts the color of objects based on scene context. The secondmethod shows that object color can be nonparametrically ""learned"" through classification methods such as Neural Networks and Multivariate Decision Trees. The methods have been successfully tested in domains such as road/highway scenes, off-road navigation and military target detection."
ICCV	Automatic Tracking of Human Motion in Indoor Scenes Across Multiple Synchronized Video Streams.	Quin Cai,Jake K. Aggarwal	1998	This pap er presents a comprehensive framework for tracking moving humans in an indoor environment from sequences of synchronized monocular grayscale images captured from multiple fixed cameras. The proposed framework consists of three main modules: Single View Tracking (SVT), Multiple View Transition Tracking (MVTT), and Automatic Camera Switching (ACS). Bayesian classification schemes based on motion analysis of human features are used to track (spatially and temporally) a subject image of interest between consecutive frames. The automatic camera switching module predicts the position of the subject along a spatial-temporal domain, and then selects the camera which provides the best view and requires the least switching to continue tracking. Limited degrees of occlusion are tolerated within the system. Tracking is based up on the images of upper human bodies captured from various viewing angles, and non-human moving objects are excluded using Principal Component Analysis (PCA). Experimental results are presented to evaluate the performance of the tracking system.
ICCV	When is it Possible to Identify 3D Objects from Single Images Using Class Constraints?	Ronen Basri,Yael Moses	1998	One approach to recognizing objects seen from arbitrary viewpoint is by extracting invariant properties of the objects from single images. Such properties are found in images of 3D objects only when the objects are constrained to belong to certain classes (e.g., bi;aterally symmetric objects). Existing studies that follow this approach propose how to compute invariant representations for a handful of classes of objects. A fundmental question regarding the invariance approach is whether it can be applied to a wide range of classes. To answer this question it is essential to study the set of classes for which invariance exists. This paper introduces a new method for determining the existence of invariance for classes of objects together with the set of images from which these invariance can be computed. We developed algebraic tests that, given a class of objects undergoing affine projection, determine whether the objects in the class can be identified from single images. In addition, these tests allow us to determine the set of views of the objects which are degenerate. We apply these tests to several classes of objects and determine which of them is identifiable and which of their views are degenerate.
ICCV	Visual Homing: Surfing on the Epipoles.	Ronen Basri,Ehud Rivlin,Ilan Shimshoni	1998	We introduce a novel method for visual homing. Using this method a robot can be sent to desired positions and orientations in 3-D space specified by single images taken from these positions. Our method is based on recovering the epipolar geometry relating the current image taken by the robot and the target image. Using the epipolar geometry, most of the parameters which specify the differences in position and orientation of the camera between the two images are recovered. However, since not all of the parameters can be recovered from two images, we have developed specific methods to bypass these missing parameters and resolve the ambiguities that exist. We present two homing algorithms for two standard projection models, weak and full perspective. Our method determines the path of the robot on-line, the starting position of the robot is relatively not constrained, and a 3-D model of the environment is not required. The method is almost entirely memoryless, in the sense that at every step the path to the target position is determined independently of the previous path taken by the robot. Because of this property the robot may be able, while moving toward the target, to perform auxiliary tasks or to avoid obstacles, without this impairing its ability to eventually reach the target position. We have performed simulations and real experiments which demonstrate the robustness of the method and that the algorithms always converge to the target pose.
ICCV	3D Modeling of Human Lip Motion.	Sumit Basu,Nuria Oliver,Alex Pentland	1998	3D Modeling of Human Lip Motion.
ICCV	Iterative Multi-Step Explicit Camera Calibration.	Jorge Batista,Helder Araújo,Anibal T. de Almeida	1998	Perspective camera calibration has been in the last decades a research subject for a large group of researchers and as a result several camera calibration methodologies can be foundin the literature. However only a small number of those methods base their approaches on the use of monoplane calibration points to realize an explicit 3D camera calibration. To avoid the singularity obtained with the calibration equations when monoplane calibration points are used, this method computes the calibration parameters in a multi-step procedure and requires a first-guess solution for the intrinsic parameters. These parameters are updated and their accuracy increased through an iterative procedure. A stability analysis as a function of the pose of the camera is presented. Camera pose view strategies for accurate camera orientation computation can be extracted from the pose view stability analysis.
ICCV	Using Algebraic Functions of Views for Indexing-Based Object Recognition.	George Bebis,Michael Georgiopoulos,Mubarak Shah,Niels da Vitoria Lobo	1998	Current indexing-based approaches build the hash table using either a large number of reference views or 3D models. In this paper, we propose building the hash table using algebraic functions of views. During preprocessing, we consider groups of model points and we represent all the views (i.e., images) that they can produce in a hash table. These views are computed using algebraic functions of a small number of reference views which contain the group. Fundamental to this procedure is a methodology based on Singular Value Decomposition and Interval Arithmetic for estimating the ranges of values that the parameters of algebraic functions can assume. During recognition, scene groups are used to retrieve from the hash table the model groups that might have produced them. Using algebraic functions of views for indexing-based recognition offers a number of advantages. First of all, the hash table can be built easier, without requiring 3D models or a large number of reference views. Second, recognition does not rely on the similarities between new and reference views. Third, verification becomes simpler. Finally, the approach is more general and extendible.
ICCV	3D Shape Reconstruction Using Volume Intersection Techniques.	Jonathan C. Carr,W. Richard Fright,Andrew H. Gee,Richard W. Prager,Kevin J. Dalton	1998	"Volume intersection algorithms are used to reconstruct incomplete objects from their silhouettes. An imagined light source is moved about the data and the cumulative amount of ""light"" seen at each point in space is interpreted as indicating the likelihood that the point is inside the object. The object data need not be uniformly distributed nor exclusively surface data. Explicit distinction between noise, surface and interior data is avoided. The novel concept of a localised viewing region is introduced to overcome the inherent inability of volume intersection algorithms to reconstruct concave surfaces. Algorithms for 2D pixel and3D voxel data are described and applied to 3D ultra-sound data."
ICCV	A Chromaticity Space for Specularity-, Illumination Color- and Illumination Pose-Invariant 3-D Object Recognition.	Daniel Berwick,Sang Wook Lee	1998	Most of the recent color recognition/indexing approaches concentr ate on establishing invariance to illumination color to improve the utility of color recognition. However, other effects caused by illumination pose and specularity on three-dimensional object surfaces have not received notable attention. We present a chromaticity recognition method that discounts the effects of illumination pose, illumination color and specularity. It utilizes a chromaticity space based on log-ratio of sensor responses for illumination pose and color invariance. A model-based specularity detection/rejection algorithm can be used to improve the chromaticity recognition and illumination estimation for objects including specular reflections.
ICCV	Information-Conserving Object Recognition.	Margrit Betke,Nicholas C. Makris	1998	"Following the theory of Statistical estimation, the problem of recognizing objects imaged in complex real-world scenes is examined from a parametric perspective. A scalar measure of an objects's complexity, which is invariant under affine transformation and changes in image noise level, is extracted from the object's Fisher information. The volume of Fisher information is shown to provide an overall statistical measure of the object's recognizability in a particular image, while the complexity provides an intrinsically physical measure that characterizes the object in any image. An information-conserving method is then developed for recognizing an object imaged in a complex scene. Here the term information-conserving means that the method uses al the measured data pertinent to the object's recognizability, attains the theoretical lower bound on estimation error for any unbiased estimate, and therefore is statistically optimal. This method is then successfully applied to finding objects imaged in thousands of complex real-world scenes."
ICCV	3D Point Distribution Models of the Cortical Sulci.	Angela Caunce,Christopher J. Taylor	1998	In this paper we present the first steps in the development of a statistical shape model, specifically a point distribution model (PDM), of the cortical surface of the brain. This will ultimately be used to locate, label, and describe the cortex, for visualisation, diagnosis, andquantification. In order to produce the model it was necessary to find and label the sulcal fissures on a series of MR images. Due to the complexity of the surface, an automated method was developed to facilitate development of a full surface model. Automating the marking process introduced the problem of identifying correspondences between examples, the knowledge of which is essential to the development of a PDM. Various methods were investigated to solve this problem including simple point matching and more complex curve matching. Each is outlined and discussed. The models obtained so far provide interesting insights into the shape and cortical pattern variations over a group of normal subjects.
ICCV	A Fast and Robust Approach for Registration of Partially Overlapping Range Images.	Chu-Song Chen,Yi-Ping Hung,Jen-Bo Cheng	1998	A Fast and Robust Approach for Registration of Partially Overlapping Range Images.
ICCV	Tracking Meteorological Structures Through Curve Matching Using Geodesic Paths.	Isaac Cohen,Isabelle Herlin	1998	This paper is concerned with the problem of tracking clouds structures like vortices in meteorological images. For this purpose we characterize the deformation between two successive occurrences, by matching their two boundary curves. Our approach is based on the computation of the set of paths connecting the two curves to be matched. It minimizes a cost function which measures the local similarity of the two curves. These matching paths are obtained as geodesic curves on this cost surface. Moreover our method allows to consider complex curves of arbitrary topologysince these curves are represented through an implicit function rather than through a parameterization. Experimental results are given to illustrate the properties of themethod in processing synthetic and then meteorologic remotely-sensed data.
ICCV	Quadric Surface Reconstruction from Dual-Space Geometry.	Geoffrey Cross,Andrew Zisserman	1998	Quadric Surface Reconstruction from Dual-Space Geometry.
ICCV	Recognition of Plane Projective Symmetry.	Rupert W. Curwen,Charles V. Stewart,Joseph L. Mundy	1998	Recognition of Plane Projective Symmetry.
ICCV	Error-Tolerant Visual Planning of Planar Grasp.	Colin Davidson,Andrew Blake	1998	"This paper describes an effcient method to calculate, from an image of an object,configurations of a two-fingered robot gripp er that form a ""cage"" to contain that object. Closing the fingers on the object from these configurations is guaranteed to reach a given desired grasp. This builds on the visual grasping theory of Blake, Taylor and Cox [1], which describes how to find optimal grasps. It extends the results of Rimon and Blake [10], which show how to construct such cages, in two ways. First, a more effcient algorithm for computing the cage is described. Second, a further development deals with occlusion by solving thecaging problem within a restricted image window. The new methods greatly reduce the complexity of the visual caging problem, making it feasible in a real time computer vision system."
ICCV	Deformable Model-Based Shape and Motion Analysis from Images using Motion Residual Error.	Douglas DeCarlo,Dimitris N. Metaxas	1998	We present a novel method for the shape and motion estimation of a deformable model using error residuals from model-based motion analysis. The motion of the model is first estimated using a model-based least squares method. Using the residuals from the least squares solution, the non-rigid structure of the model can be better estimated by computing how changes in the shape of the model affect its motion parameterization. This method is implemented as a component in a deformable model-based framework that uses optical flow information and edges. This general model-based framework is applied to human face shape and motion estimation. We present experiments that demonstrate that this framework is a considerable improvement over a framework that uses only optical flow information and edges.
ICCV	Initialization of Deformable Models from 3D Data.	Herve Delingette	1998	The robustness of shape recovery based on deformable models dep ends in general, on the relative difference of position and topology of the initial model with respect to the data: a close initialization with correct topology guaranties a proper recovery of the object. Furthermore, the closeness of the initial model greatly influences the time of computation needed for the recovery. In this paper, we propose a method for initializing deformable models from range data or volumetric images. The prop osed method solves two distinct problems. First, we use the topological segmentation of volumetric images in order to recover the approximate topology of the object. Second, we use an efficient mesh sampling algorithm to control the number of vertices of the initial model. The method takes into account missing data and outliers.
ICCV	Illumination-Invariant Color Object Recognition via Compressed Chromaticity Histograms of Color-Channel-Normalized Images.	Mark S. Drew,Jie Wei,Ze-Nian Li	1998	Several color object recognition methods that are based on image retrieval algorithms attempt to discount changes of illumination in order to increase performance when test image illumination conditions differ from those that obtained when the image database was created. Here we extend the seminal method of Swain and Ballard to discount changing illumination.The new method is based on the first stage of the simplest color indexing method, which uses angular invariants between color image and edge image channels. That method first normalizes image channels, and then effectively discards much of the remaining information. Here we adopt the color-normalization stage as an adequate color constancy step. Further,we replace 3D color histograms by 2D chromaticity histograms. Treating these as images, we implement the method in a compressed histogram-image domain using a combination of wavelet compression and Discrete Cosine Transform (DCT) to fully exploit the technique of low-pass filtering for efficiency. Results are very encouraging, with substantially better performance than other methods tested. The method is also fast, in that the indexing process is entirely carried out in the compressed domain and uses a feature vector of only 36 or 72 values.
ICCV	Optimal Polyline Tracking for Artery Motion Compensation in Coronary Angiography.	Marie-Pierre Dubuisson-Jolly,Cheng-Chung Liang,Alok Gupta	1998	We propose a novel solution to the problem of motion compensation of coronary angiographs. As the heart is beating, it is difficult for the physician to observe closely a particular point (e.g. stenosis) on the artery tree. We propose to rigidly compensate the sequence so that the area around the point of interest appears stable. This is a difficult problem because the arteries deform in a nonrigid manner and only their 2D X-ray projection is observed. Also, the lack of features around the selected point makes the matching subject to the aperture problem. The algorithm automatically extracts a section of the artery of interest, models it as a polyline, andtracks it. The problem is formulated as an energy minimization problem which is solved using a shortest path in a graph algorithm. The motion compensated sequence can be obtained by translating every pixel so that the point of interest remains stable. We have applied this algorithm to many examples in two sets of angiography data and have obtained excellent results.
ICCV	Understanding Object Motion of Tools and Vehicles.	Zoran Duric,Ehud Rivlin,Azriel Rosenfeld	1998	Understanding Object Motion of Tools and Vehicles.
ICCV	Learning to Identify and Track Faces in Image Sequences.	Gareth J. Edwards,Christopher J. Taylor,Timothy F. Cootes	1998	We address the problem of robust face identification in the presence of pose, lighting, and expression variation. Previous approaches to the problem have assumed similar models of variation for each individual, estimated from pooled training data. We describe a method of updating a first order global estimate of identity by learning the class-specific correlation between the estimate and the residual variation during a sequence. This is integrated with an optimal tracking scheme, in which identity variation is decoupled from pose, lighting and expression variation. The method results in robust tracking and a more stable estimate of facial identity under changing conditions.
ICCV	A Nonlinear Method for Estimating the Projective Geometry of Three Views.	Olivier D. Faugeras,Théodore Papadopoulo	1998	A Nonlinear Method for Estimating the Projective Geometry of Three Views.
ICCV	What Can Projections of Flow Fields Tell Us About Visual Motion.	Sándor Fejes,Larry S. Davis	1998	What Can Projections of Flow Fields Tell Us About Visual Motion.
ICCV	Sectored Snakes: Evaluating Learned-Energy Segmentations.	Samuel D. Fenster,John R. Kender	1998	"We describe how to teach deformable models to maximize image segmentation correctness based on userspecified criteria, and we present a method for evaluating which criteria work best. We present sectored snakes, a formulation that demonstrably improves upon regular snakes.A traditional deformable model (""snake"" in 2D) fails to find an object's boundary when the strongest nearby image edges are not the ones sought. But models can be trained to respond to other image features instead, by learning their probability distributions. The implementor must then decide on which of many image qualities to teach the model. To this end, we show how to evaluate the efficacy of any resulting deformable model, given a sampling of ground truth, a model of the range of shapes tried during optimization, and a measure of shape closeness.In the domain of abdominal CT images, we demonstrate such evaluation on a simple ""sectoring"" of a snake, in which intensity and perpendicular gradient are observed over equal-length segments. This speci.c set of qualities shows a measured improvement over an objective function that is uniform around the shape, and it follows naturally from examination of the latter's failures due to images variations around the organ boundary."
ICCV	Which Shape from Motion?.	Cornelia Fermüller,Yiannis Aloimonos	1998	In a practical situation, the rigid transformation relating different views is recovered with errors. In such a case, the recovered depth of the scene contains errors, and consequently a distorted version of visual space is computed. What then are meaningful shape representations that can be computed from the images? The result presented in this paper states that if the rigid transformation between different views is estimated in a way that gives rise to a minimum number of negative depth values, then at the center of the image affine shape can be correctly computed. This result is obtained by exploiting properties of the distortion function, developed in [1]. The distortion model turns out to be a very powerful tool in the analysis and design of 3D motion and shape estimation algorithms, and as a byproduct of our analysis we present a computational explanation of psychophysical results demonstrating human visual space distortion from motion information.
ICCV	Building Qualitative Event Models Automatically from Visual Input.	Jonathan H. Fernyhough,Anthony G. Cohn,David Hogg	1998	We describ e an implemented technique for generating event models automatically based on qualitative reasoning and a statistical analysis of video input. Using an existing tracking program which generates labelled contours for objects in every frame, the view from a fixed camera is partitioned into semantically relevant regions based on the paths followed by movingobjects. The paths are indexed with temporal information so objects moving along the same path at different speeds can be distinguished. Using a notion of proximity based on the speed of the moving objects and qualitative spatial reasoning techniques, event models describing the behaviour of pairs of objects can be built, again using statistical methods. The system has been tested on a traffic domain and learns various event models expressed in the qualitative calculus which represent human observable events. The system can then be used to recognise subsequent selected event occurrences or unusual behaviours.
ICCV	Achieving a Fitts Law Relationship for Visual Guided Reaching.	Nicola J. Ferrier	1998	In order to take advantage of the top speed of manipulators, vision can not be tightly integrated into the motion control loop. Past visual servo control systems have performed satisfactorily with this constraint, however it can be shown that the task execution time can be reduced if the vision system is de-coupled from the low level motor control system.For reaching, there is a trade-off between the accuracy of a motion and the time requir ed to execute a motion. In studies of human motor contr ol this trade-off is quantified by Fitts Law, a relationship between the motion time, the target distance, and the target width. These studies suggest that vision is not used tightly within the control loop, i.e. as a sensor that is servo-ed on, but rather vision is used to determine where the reaching target is and whether target has been reached successfully. Through a simple robotic example we demonstrate that a similar trade off exists between motion accuracy and the motion execution time for visual guided robot reaching motions. The ACM Portal is published by the Association for Computing Machinery. Copyright © 2010 ACM, Inc. Terms of Usage Privacy Policy Code of Ethics Contact Us Useful downloads: Adobe Acrobat QuickTime Windows Media Player Real Player
ICCV	3D Reconstruction with Projective Octrees and Epipolar Geometry.	Blanca Garcia,Pere Brunet	1998	"In this paper, the problem of generating a 3D octree-like structure with the help of epipolar geometry within a projective framework is addressed. After a brief introduction on the basics of octrees and epipolar geometry, the new concept called ""projective octree"" is introduced together with an algorithm for building this projective structure. Finally, some results of the implementations are presented in the last section together with the conclusions and future work."
ICCV	Parameterized Image Varieties: A Novel Approach to the Analysis and Synthesis of Image Sequences.	Yakup Genc,Jean Ponce	1998	This paper addresses the problem of characterizing the space formed by all images of a rigid set of n points observed by a weak perspective or paraperspective camera. By taking explicitly into account the Euclidean constraints associated with calibrated cameras, we showthat this space is a six-dimensional variety embedded in IR2n, and parameterize it using the image positions of three reference points. This parameterization is constructed vialinear least squares from point correspondences established across a sequence of images, and it is used to synthesize new pictures without any explicit three-dimensional model. Degenerate scene and camera configurations are analyzed, and experiments with real image sequences are presented.
ICCV	Image Indexing using Composite Color and Shape Invariant Features.	Theo Gevers,Arnold W. M. Smeulders	1998	New sets of color models are proposed for object recognition invariant to a change in view point, object geometry and illumination. Further, computational methods are presented to combine color and shape invariants to produce a high-dimensional invariant feature set for discriminatory object recognition.Experiments on a database of 500 images show that object recognition based on composite color and shape invariant features provides excellent recognition accuracy. Furthermore, object recognition based on color invariants provides very high recognition accuracy whereas object recognition based entirely on shape invariants yields very poor discriminative power.The image database and the performance of the recognition scheme can be experienced within PicToSeek on-line as part of the ZOMAX system at: http://www.wins.uva.nl/research/isis/zomax/.
ICCV	Affine Invariant Medial Axis and Skew Symmetry.	Peter J. Giblin,Guillermo Sapiro	1998	Affine invariant medial axes and symmetry sets of planar shapes are intr oduced and studied in this paper. Two different approaches are presented. The first one is based on affine invariant distances, and defines the symmetry set, a set containing the medial axis, as the closure of the locus of points on (at least) two affine normals and affine-equidistant from the corresponding points on the curve. The second approach is based on a affine bitangent conics. In this case the symmetry set is defined as the closure of the locus of centers of conics with (at least) three-point contact with two or more distinct points on the curve. This is equivalent to conic and curve having, at those points, the same affine tangent, or the same Euclidean tangent and curvature. Although the two analogous definitions for the classical Euclidean symmetry set (medial axis) are equivalent, this is not the case for the affine group. We then show how to use the symmetry set to detect affine skew symmetry, proving that the contact based symmetry set is a straight line if and only if the given shape is the affine transformation of a symmetric object.
ICCV	Ego-Motion and Omnidirectional Cameras.	Joshua Gluckman,Shree K. Nayar	1998	Recent research in image sensors has produced cameras with very large fields of view. An area of computer vision research which will benefit from this technology is the computation of camera motion (ego-motion) from a sequence of images. Traditional cameras suffer from the problem that the direction of translation may lie outside of the field of view, making the computation of camera motion sensitive to noise. In this paper, we present a method for the recovery of ego-motion using omnidirectional cameras. Noting the relationship between spherical projection and wide-angle imaging devices, we propose mapping the image velocity vectors to a sphere, using the Jacobian of the transformation between the projection model of the camera and spherical projection. Once the velocity vectors are mapped to a sphere, we show how existing ego-motion algorithms can be applied and present some experimental results. These results demonstrate the ability to compute ego-motion with omnidirectional cameras.
ICCV	Face Surveillance.	Srinivas Gutta,Jeffrey Huang,Vishal Kakkad,Harry Wechsler	1998	"Most of the research on face recognition addresses the MATCH problem and it assumes a closed universe where there is no need for a REJECT ('false positive') option. The SURVEILLANCE problem is addressed indirectly, if at all, through the MATCH problem, where the size of the gallery rather than that of the probe set is very large. This paper addresses the proper surveillance problem where the size of the probe ('unknown image') set vs gallery ('known image') set is 450 vs 50 frontal images. We developed robust face ID verification ('classification') and retrieval schemes based on hybrid classifiers and showed their feasibility using the FERET face database. The hybrid classifier architecture consists of an ensemble of connectionist networks - Radial Basis Functions (RBF) and inductive decision trees (DT). Experimental results prove the feasibility of our approach and yield 97% accuracy using the probe and gallery sets specified above."
ICCV	Shading Primitives: Finding Folds and Shallow Grooves.	John A. Haddon,David A. Forsyth	1998	"Diffuse interreflections cause effects that make current theories of shape from shading unsatisfactory. We show that distant radiating surfaces produce radiosity effects at low spatial frequencies. This means that, if a shading pattern has a small region of support, unseen surfaces in the environment can only produce effects that vary slowly over the support region. It is therefore relatively easy to construct matching processes for such patterns that are robustto interreflections. We call regions with these patterns ""shading primitives.""Folds and grooves on surfaces provide two examples of shading primitives; the shading pattern is relatively independent of surface shape at a fold or a groove, and the pattern is localised. We show that the pattern of shading can be predicted accurately by a simple model, and derive a matching process from this model. Both groove and fold matchers are shown to work well on images of real scenes."
ICCV	Minimizing Algebraic Error in Geometric Estimation Problems.	Richard I. Hartley	1998	Minimizing Algebraic Error in Geometric Estimation Problems.
ICCV	Wormholes in Shape Space: Tracking Through Discontinuous Changes in Shape.	Tony Heap,David Hogg	1998	"Existing object tracking algorithms generally use some form of local optimisation, assuming that an object's position and shape change smoothly over time. In some situations this assumption is not valid: the trackable shape of an object may change discontinuously, for example if it is the 2D silhouette of a 3D object.In this paper we propose a novel method for modelling temporal shape discontinuities explicitly. Allowable shapes are represented as a union of (learned) bounded regions within a shape space. Discontinuous shape changes are described in terms of transitions between these regions. Transition probabilities are learned from training sequences and stored in a Markov model. In this way we can create ""wormholes"" in shape space. Tracking with such models is via an adaptation of the Condensation algorithm."
ICCV	Fast Stereovision with Subpixel-Precision.	Rolf D. Henkel	1998	A fast stereo algorithm based on aliasing effects of simple disparity estimators within a coherence-detection scheme is presented. The algorithm calculates dense disparity maps with subpixel-precision by performing local spatial filter operations and simple arithmetic transformations. Performance similar to classical area-based approaches is achieved, but without the complicated hierarchical search structure typical for these approaches. The algorithm is completely parallel, the disparity values are calculated independently for each pixel. In addition, local validation counts for the disparity estimates and a fused cyclopean view of the scene are available within the proposed network structure for coherence-based stereo. The ACM Portal is published by the Association for Computing Machinery. Copyright © 2010 ACM, Inc. Terms of Usage Privacy Policy Code of Ethics Contact Us Useful downloads: Adobe Acrobat QuickTime Windows Media Player Real Player
ICCV	Self-Calibration and Euclidean Reconstruction Using Motions of a Stereo Rig.	Radu Horaud,Gabriela Csurka	1998	This paper describes a method to upgrade projective reconstruction to affine and to metric reconstructions using rigid gener almotions of a stereo rig. We make clear the algebraic relationships between projective reconstruction, the plane at infinity (affine reconstruction), camera calibration, and metric reconstruction. We show that all the computations can be carried out using standard linear resolution methods and that these methods compare favorably with nonlinear optimization methods in the presence of Gaussian noise. We carry out a theoretical error analysis which quantify the relative importance of the accuracies of projective-to-affine conversion and affine-to-Euclidean conversion. Experiments with real data are consistent with the theoretical error analysis and with a sensitivity analysis performed with simulated data.
ICCV	Spatial Color Indexing and Applications.	Jing Huang,Ravi Kumar,Mandar Mitra,Wei-Jing Zhu	1998	We suggest the use of the color correlogram as a generic indexing tool to tackle various computer vision problems. Correlograms were shown to be very effective for content-basedimage retrieval [4]. We adapt the correlogram to handle the problems of image subregion querying, object localization, object tracking, and cut detection. Experimental results suggest that the color correlogram is much more effective than the histogram for these applications, with insignificant additional computational, storage, or processing cost. We also provide a technique to cut down the storage requirement of correlograms so that it is the same as that of histograms, with only negligible performance penalty compared to the original correlogram.
ICCV	Detecting Changes in Aerial Views of Man-Made Structures.	Andres Huertas,Ramakant Nevatia	1998	"Many applications require detecting structural changes in a scene over a period of time. Comparing intensity values of successive images is not effective as such changes don't necessarily reflect actual changes at a site but might be caused by changes in the view point, illumination and seasons. We take the approach of comparing a 3-D model of the site, prepared from previous images, with new images to infer significant changes. This task is difficult as the images and the models have very different levels of abstract representations. Our approach consists of several steps: registering a site model to a new image, model validation to confirm the presence of model objects in the image; structural change detection seeks to resolve matching problems and indicate possibly changed structures; and finally updating models to reflect the changes. Our system is able to detect missing (or mis-modeled) buildings, changes in model dimensions, and new buildings under some conditions."
ICCV	Relational Histograms for Shape Indexing.	Benoit Huet,Edwin R. Hancock	1998	This paper is concerned with the retrieval of images from large databases based on their shape similarity to a query image. Our approach is based on two dimensional histograms that encode both the local and global geometric properties of the shapes. The pairwise attributes are the directed segment relative angle and directed relative position The novelty of the proposed approach is to simultaneously use the relational and structural constraints, derived from an adjacency graph, to gate histogram contributions. We investiguate the retrieval cap abilities of the method for various queries. We also investigate the robustness of the method to segmentation errors. We conclude that a relational histo gram of pairwise segment attributespresents a very efficient way of indexing into large databases. The optimal configuration is obtained when the local features are constructed from six neighbouring segments pairs. Moreover, a sensitivity analysis reveals that segmentation errors do not affect the retrieval performances.
ICCV	Robust Multi-Sensor Image Alignment.	Michal Irani,P. Anandan	1998	This paper presents a method for alignment of images acquired by sensors of different modalities (e.g., EO and IR). The paper has two main contributions: (i) It identifies an appropriate image representation for multi-sensor alignment, i.e., a representation which emphasizes the common information between the two multi-sensor images, suppresses the non-common information, and is adequate for coarse-to-fine processing. (ii) It presents a new alignment technique, which applies global estimation to any choice of a local similarity measure. In particular, it is shown that when this registration technique is applied to the chosen image representation with a local-normalized-correlation similarity measure, it providesa new multi-sensor alignment algorithm which is robust to outliers, and applies to a wide variety of globally complex brightness transformations between the two images.Our proposed image representation does not rely on sparse image features (e.g., edge, contour, or point features). It is continuous and does not eliminate the detailed variations within local image regions. Our method naturally extends to coarse-to-fine processing, and applies even in situations when the multi-sensor signals are globally characterized by low statistical correlation.
ICCV	A Mixed-State CONDENSATION Tracker with Automatic Model-Switching.	Michael Isard,Andrew Blake	1998	"There is considerable interest in the computer vision community in representing and modelling motion. Motion models are use d as predictors to increase the robustness and accuracy of visual trackers, and as classifiers for gesture recognition. This paper presents a significant development of random sampling methods to allow automatic switching between multiple motion models as a natural extension of the tracking process. The Bayesian mixed-state framework is described in its generality, and the example of a bouncing ball is used to demonstrate that a mixed-state model can significantly improve tracking performance in heavy clutter. The relevance of the approach to the problem of gesture recognition is then investigated using a tracker which is able to follow the natural drawing action of a hand holding a pen, and switches state according to the hand's motion."
ICCV	Robust Contour Tracking in Echocardiographic Sequences.	Gary Jacob,J. Alison Noble,Andrew Blake	1998	In this paper we present an evaluation of a robust visual image tracker on echocardiographic image sequences. We show how the tracking framework can be customised to define an appropriate shape-space that describes heart shape deformations that can be learnt from a training data set. We also investigate an energy-based temporal boundary enhancement method to improve image feature measurement. Preliminary results are presented demonstrating tracking on real normal heart motion data sequences and synthesised and real abnormal heart motion data sequences. We conclude by discussing some of our current research efforts.
ICCV	Condensing Image Databases when Retrieval is Based on Non-Metric Distances.	David W. Jacobs,Daphna Weinshall,Yoram Gdalyahu	1998	One of the key problems in appearance-based-vision is understanding how to use a set of labeled images to classify new images. Classification systems that can model human performance, or that use robust image matching methods, often make use of similarity judgments that are non-metric, but when the triangle inequality is not obeyed, most existing pattern recognition techniques are not applicable. We note that exemplar-based (or nearest-neighbor) methods can be applied naturally when using a wide class of non-metric similarity functions. The key issue, however, is to find methods for choosing good representatives of a class that accurately characterize it. We show that existing condensing techniques for finding class representatives are ill-suited to deal with non-metric dataspaces.We then focus on developing techniques for solving this problem, emphasizing two points: First, we show that the distance between two images is not a good measure of how well one image can represent another in non-metric spaces. Instead, we use the vector correlation between the distances from each image to other previously seen images. Second, we show that in non-metric spaces, boundary points are less significant for capturing the structure of a class than they are in Euclidean spaces. We suggest that atypical points may be more important in describing classes. We demonstrate the importance of these ideas to learning thatgeneralizes from experience by improving performance using both synthetic and real images.
ICCV	Mixtures of Eigen Features for Real-Time Structure from Texture.	Tony Jebara,Kenneth B. Russell,Alex Pentland	1998	Mixtures of Eigen Features for Real-Time Structure from Texture.
ICCV	Multidimensional Morphable Models.	Michael J. Jones,Tomaso Poggio	1998	We describe a flexible model for representing images of objects of a certain class, known a priori, such as faces, and introduce a new algorithm for matching it to a novel image and thereby performing image analysis. We call this model a multidimensional morphable model or just a morphable model. The morphable model is learned from example images (calledprototypes) of objects of a class. In this paper we introduce an effective stochastic gradient descent algorithm that automatically matches a model to a novel image by finding the parameters that minimize the error between the image generated by the model and the novelimage. Two examples demonstrate the robustness and the broad range of applicability of the matching algorithm and the underlying morphable model. Our approach can provide novel solutions to several vision tasks, including the computation of image correspondence, object verification, image synthesis and image compression.
ICCV	Motion Estimation in Image Sequences Using the Deformation of Apparent Contours.	Fredrik Kahl,Kalle Åström	1998	Motion Estimation in Image Sequences Using the Deformation of Apparent Contours.
ICCV	Using Conic Correspondence in Two Images to Estimate the Epipolar Geometry.	Fredrik Kahl,Anders Heyden	1998	In this paper it is shown how corresponding conics in two images can be used to estimate the epipolar geometry in terms of the fundamental/essential matrix. The corresponding conics can be images of either planar conics or silhouettes of quadrics. It is shown that one conic correspondence gives two independent constraints on the fundamental matrix and a method to estimate the fundamental matrix from at least four corresponding conics is presented. Furthermore, a new type of fundamental matrix for describing conic correspondences is introduced. Finally, it is shown that the problem of estimating the fundamental matrix from 5 point correspondences and 1 conic correspondence in general has 10 different solutions. A method to calculate these solutions is also given together with an experimental validation.
ICCV	Accurate, Real-Time, Unadorned Lip Tracking.	Robert Kaucic,Andrew Blake	1998	Human speech is inherently multi-modal, consisting of both audio and visual components. Recently researchers have shown that the incorporation of information about the position of the lips into acoustic speech recognisers enables robust recognition of noisy speech. In the case of Hidden Markov Model-recognition, we show that his happens because the visual signal stabilises the alignment of states. It is also shown that unadorned lips, both the inner and outer contours, can be robustly tracked in real time on general-purpose workstations. To accomplish this, efficient algorithms are employed which contain three key components: shape models, motion models, and focused colour feature detectors ¿ all of which are learnt from examples.
ICCV	Local Scale Controlled Anisotropic Diffusion with Local Noise Estimate for Image Smoothing and Edge Detection.	Ping Liang	1998	Local Scale Controlled Anisotropic Diffusion with Local Noise Estimate for Image Smoothing and Edge Detection.
ICCV	Recovering Epipolar Geometry by Reactive Tabu Search.	Qifa Ke,Gang Xu,Songde Ma	1998	In this paper we propose a new approach to recovering epipolar geometry from a pair of uncalibrated images. We first detect the feature points. By minimizing a proposed cost function, we match the feature points, discard the outliers and recover the epipolar geometry in one step. Experiments on real images show that this approach is effective and fast.
ICCV	Physics-based 3D Position Analysis of a Soccer Ball from Monocular Image Sequences.	Taeone Kim,Yongduek Seo,Ki-Sang Hong	1998	"In this paper, we propose a method for locating 3D position of a soccor ball from monocular image sequence of soccor games. Toward this goal, we adopted ground-model-to-image transformation together with physics-based approach that a ball follows the parabolic trajectory in the air. By using the transformation the heights of a ball can be easily calculated using simple triangular geometric relations given the start and the end position of the ball on the ground. Here the heights of a ball are determined in terms of a player's height. Even if the end position of a ball is not given on the ground due to kicking or heading of a falling ball before it touches the ground, the most probable trajectory can be determined by searching based on the physical fact taht the ball follows a parabolic trajectory in the air. We have tested and experimented with a real image sequence the results of which seem promising."
ICCV	Robotic Control with Partial Visual Information.	Keisuke Kinoshita,Michael Lindenbaum	1998	We consider a class of control tasks, which rely from partial visual information in a robotic setting. These tasks are hard in the sense that at every given moment, the available information is insufficient for the control task. Still, the amount of information collected throughout the control process is large and thus seems sufficient for carrying out the task.Such situations commonly arise when the object is frequently occluded from one of the cameras in a stereo pair or when only one moving camera is available.We propose a generic contr ol rule for such tasks and characterize the conditions required for the success of the task. The analysis is based on the observation that mathematically, the behavior of such systems is related to a class of row-action (and POCS) optimization algorithms.In the second part of the paper we focus on one particular task from this class: position and orientation control with a single rotating camera. We show that this task may be carried out, in principle, for any camera movement, and suggest efficient control strategy and camera moving strategy. Interestingly, it seems that the advisable control law is not consistent withsimple-minded intuition. We substantiate our claims by simulations and experiments on real data.
ICCV	Representation and Self-Similarity of Shapes.	Tyng-Luh Liu,Davi Geiger,Robert Kohn	1998	Representing shapes is a significant problem for vision systems that must recognize or classify objects. We derive a representation for a given shape by investigating its self-similarities, and constructing its shape axis(SA) and shape axis tree (SA-tree).We start with a shape, its boundary contour, and two different parameterizations for the contour. To measure its self-similarity we consider matching pairs of points (and their tangents) along the boundary contour, i.e., matching the two parameterizations. The matching, or self-similarity criteria may vary, e.g., co-circularity, parallelism, distance, region homogeneity. The loci of middle points of the pairing contour points are the shape axis and they can be grouped into a unique tree graph, the SA-tree. The shape axis for the co-circularity criteria is compared to the symmetry axis. An interpretation in terms of object parts is also presented.
ICCV	2D Affine Transformations are Unlikely to Account for Human 3D Object Recognition.	Zili Liu,Daniel Kersten	1998	2D Affine Transformations are Unlikely to Account for Human 3D Object Recognition.
ICCV	Finding Periodicity in Space and Time.	Fang Liu,Rosalind W. Picard	1998	"An algorithm for simultaneous detection, segmentation, and characterization of spatiotemporal periodicity is presented. The use of periodicity templates is proposed to localize and characterize temporal activities. The templates not only indicate the presence and location of a periodic event, but also give an accurate quantitative periodicity measure. Hence, they can be used as a new means of periodicity representation. The proposed algorithm can also be considered as a ""periodicity filter"" a low-level model of periodicity perception. The algorithm is computationally simple, and shown to be more robust than optical flow based techniques in the presence of noise. A variety of real-world examples are used to demonstrate the performance of the algorithm."
ICCV	Independent 3D Motion Detection Using Residual Parallax Normal Flow Fields.	Manolis I. A. Lourakis,Antonis A. Argyros,Stelios C. Orphanoudakis	1998	"This paper considers a specific problem of visual perception of motion, namely the problem of visual detection of independent 3D motion. Most of the existing techniques for solving this problem rely on restrictive assumptions about the environment, the observer's motion, or both. Moreover, They are based on the computation of a dense optical flow field, which amounts to solving the ill-posed correspondence problem. In this work, independent motion detection is formulated as a problem of robust parameter estimation applied to the visual input acquired by a rigidly moving observer. The proposed method automatically selects a planar surface in the scene and the residual planar parallax normal flow field with respect to the motion of this surface is computed at two successive time instants. The two resulting normal flow field are then combined in a linear model. The parameters of this model are related to the parameters of self-motion (ego-motion) and their robust estimation leads to a segmentation of the scene based on 3D motion. The method avoids a complete solution to the correspondence problem by selectively matching subsets of image points and by employing normal flow fields. Experimental results demonstratethe effectiveness of the proposed method in detecting independent motion in scenes with large depth variations and unrestrited observer motion."
ICCV	A Task Driven 3d Object Recognition System using Bayesian Networks.	Björn Krebs,Bernd Korn,M. Burkhardt	1998	"In this paper we propose a general framework to build a task oriented 3d object recognition system for CAD based vision (CBV). Features from 3d space curves representing the object's rims provide sufficient information to allow identification and pose estimation of industrial CAD models. However, features relying on differential surface properties tend to be very vulnerable with respect to noise. To model the statistical behavior of the data we introduce Bayesian netswhich model the relationship between objects and observable features. Furthermore, task oriented selection of the optimal action to reduce the uncertainty of recognition results is incorporated into the Bayesian nets. This enables the integration of intelligent recognition strategies depending on the already acquired evidence into a robust, and effcient, 3d CAD based recognition system."
ICCV	A Probabilistic Contour Discriminant for Object Localisation.	John MacCormick,Andrew Blake	1998	"A method of localising objects in images is proposed. Possible configurations are evaluated using the contour discriminant, a likelihood ratio which is derived from a probabilistic model of the feature detection process. We treat each step in this process probabilistically, including the occurrence of clutter features, and derive the observation densities for both correct ""target"" configurations and incorrect ""clutter"" configurations. The contour discriminant distinguishes target objects from the background even in heavy clutter, making only the most general assumptions about the form that clutter might take. The method generates samples stochastically to avoid the cost of processing an entire image, and promises to be particularly suited to the task of initialising contour trackers based on sampling methods."
ICCV	Morphological Corner Detection.	Robert Laganière	1998	This paper presents a new operater for corner detection. This operator uses a variant of the morphological closing operator, which we have called asymmetrical closing. It consists of the successive application of different morphological transformations using different structuring elements. Each of these structuring elements used to probe the image under study is tuned to affect corners of different orientation and brightness. We found that this kind of approach, based on brightness comparisons, leads to better quality results than others and is achieved at a lower computational cost.
ICCV	Geotensity: Combining Motion and Lighting for 3D Surface Reconstruction.	Atsuto Maki,Mutsumi Watanabe,Charles Wiles	1998	This paper is about automatically reconstructing the full 3D surface of an object observed in motion by a single static camera. We introduce the geotensity constraint that governs the relationship between four images of a moving object under fairly general lighting conditions. We show that it is possible in theory to solve for 3D surface structure for both the case of a single point light source and a pair of point light sources and propose that a solution exists for an arbitrary number point light sources. The surface may or may not be textured. We then give an example of automatic surface reconstruction of a face under a point light source. The geotensity constraint provides the theoretical foundation for the full automatic 3D reconstruction of Lambertian objects using a single fixed camera and arbitrary unknown object motion under arbitrary lighting conditions.
ICCV	A Real-Time Algorithm for Medical Shape Recovery.	Ravi Malladi,James A. Sethian	1998	In this paper, we present a shape recovery technique in 2D and 3D with specific applications in visualizing and measuring anatomical shapes from medical images. This algorithm models extremely corrugated structures like the brain, is topologically adaptable, is robust, and runs in O(N logN) time where N is the total number of points in the domain. Our two-stage technique is based on the level set shape recovery scheme introduced in [11, 12, 4] and the fast march ing method in [19] for computing solutions to static Hamilton-Jacobi equations.
ICCV	Shape Recovery Using Dynamic Subdivision Surfaces.	Chhandomay Mandal,Baba C. Vemuri,Hong Qin	1998	A new dynamic subdivision surface model is proposed for shape recovery from 3D data sets. The model inherits the attractive properties of the Catmull-Clark subdivision scheme and is set in a physics-based modeling paradigm. Unlike other existing methods, our model does not require a parameterized input mesh to recover shapes of arbitrary topology, allows direct manipulation of the limit surface via application of forces and provides a fast, robust, and hierarchical approach to recover complex shapes from 3D data with very few degrees of freedom (contr ol vertic es). We provide an analytic formulation and introduce the physical quantities required to develop the dynamic subdivision surface model which can be deformed by applying forces synthesized from the data. Our experiments demonstrate that this new dynamic model has a promising future in shape recovery from volume and range data sets.
ICCV	Stereo Depth Estimation: A Confidence Interval Approach.	Robert Mandelbaum,G. Kamberova,Max Mintz	1998	We describe an estimation technique which, given a measurement of the depth of a target from a wide-field-of-view (WFOV) stereo camera pair, produces a minimax risk fixed-size confidence interval estimate for the target depth. This work constitutes the first application to the computer vision domain of optimal fixed-size confidence-interval decision theory. The approach is evaluated in terms of theoretical capture probability and empirical capture frequency during actual experiments with a target on an optical bench. The method is compared to several other procedures including the Kalman Filter. The minimax approach is found to dominate all the other methods in performance. In particular, for the minimax approach, a very close agreement is achieved between theoretical capture probability and empirical capture frequency. This allows performance to be accurately predicted, greatly facilitating the system design, and delineating the tasks that may be performed with a given system.
ICCV	Estimation with Bilinear Constraints in Computer Vision.	Yoram Leedan,Peter Meer	1998	Estimation with Bilinear Constraints in Computer Vision.
ICCV	PIMs and Invariant Parts for Shape Recognition.	Zhibin Lei,Tolga Tasdizen,David B. Cooper	1998	"We present completely new very powerful solutions to two fundamental problems central to computer vision. 1. Given data sets representing C objects to be stored in a database, and given a new data set for an object, determine the object in the database that is most like the object measured. We solve this problem through use of PIMs (""Polynomial Interpolated Measures""), which is a new representation integrating implicit polynomial curves and surfaces, explicit polynomials, and discrete data sets which may be sparse. The method provides high accuracy at low computational cost. 2. Given noisy 2D data along a curve (or 3D data along a surface), decompose the data into patches such that new data taken along affine transformations or Euclidean transformations of the curve (or surface) can be decomposed into correponding patches. Then recognition of complex or partially occluded objects can be done in terms of invariantly determined patches. We briefly outline a low computational cost image-database indexing-system based on this representation for objects having complex shape-geometry."
ICCV	A Probabilistic Framework for Edge Detection and Scale Selection.	David H. Marimont,Yossi Rubner	1998	"We devise a statistical framework for edge detection by performing a statistical analysis of zero crossings of the second derivative of an image. This analysis enables us to estimate at each pixel of an image the probability that an edge passes through the pixel. We present a statistical analysis of the Lindeberg operators that we use to compute image derivatives. We also introduce a confidence probability that tells us how reliable the edge probability is, given the image's noise level and the operator's scale. Combining the edge and confidence probabilities leads to a probabilistic scale selection algorithm. We present the results of experiments on natural images."
ICCV	Ambiguity in Reconstruction from Images of Six Points.	Stephen J. Maybank,Amnon Shashua	1998	"Let S be a set of six points in space, let \psi be any hyperboloid of one sheet containing S, and let I be a sequence of images of S taken by an uncalibrated camera moving over \psi. Then reconstruction from I is subject to a three way ambiguity which is unbroken as long as the optical centre of the camera remains on \psi.Let p be an image of S taken from a point on \psi. The images ""near"" p define a tangent space which splits into a direct sum W_p\oplus N_p\oplus F_p, where Wp corresponds to images near p for which the ambiguity is maintained, Np corresponds to images for which the ambiguity is broken and Fp corresponds to images which are physically impossible."
ICCV	A Multigrid Approach for Hierarchical Motion Estimation.	Étienne Mémin,Patrick Pérez	1998	This paper focuses on the estimation of the apparent motion field between two consecutive frames in an image sequence. The approach developed here is a trade-off between methods based on global parameterized flow models and local dense optic flow estimators.The method relies on an adaptive multigrid minimization approach. In addition to accelerated convergence toward good estimates, it allows to mix different parameterizations of the estimate relative to adaptive partitions of the image.The performances of the resulting algorithms are demonstrated in the difficult context of a non-convex energy. Experimental results on real world Meteosat sequences are presented.
ICCV	A General Framework for Object Detection.	Constantine Papageorgiou,Michael Oren,Tomaso Poggio	1998	This paper presents a general trainable framework for object detection in static images of cluttered scenes. The detection technique we develop is based on a wavelet representation of an object class derived from a statistical analysis of the class instances. By learning an object class in terms of a subset of an overcomplete dictionary of wavelet basis functions, we derive a compact representation of an object class which is used as an input to a support vector machine classifier. This representation overcomes both the problem of in-class variability and provides a low false detection rate in unconstrained environments.We demonstr ate the capabilities of the technique in two domains whose inherent information content differs significantly. The first system is face detection and the second is the domain of people which, in contrast to faces, vary greatly in color, texture, and patterns. Unlike previous approaches, this system learns from examples and does not rely on any a priori (hand-crafted) models or motion-based segmentation. The paper also presents a motion-based extension to enhance the performance of the detection algorithm over video sequences. The results presented here suggest that this architecture may well be quite general.
ICCV	Human Face Recognition: A Minimal Evidence Approach.	Ali Reza Mirhosseini,Catherine Chen,Tuan D. Pham,Hong Yan	1998	Human Face Recognition: A Minimal Evidence Approach.
ICCV	A PDE-Based Level-Set Approach for Detection and Tracking of Moving Objects.	Nikos Paragios,Rachid Deriche	1998	This papers pr esents a framework for detecting and tracking moving objects in a sequence of images. Using a statistical approach, where the inter-frame difference is modeled by a mixture of two Laplacian or Gaussian distributions, and an energy minimization based approach, we reformulate the motion detection and tracking problem as a front propagation problem. The Euler-Lagrange equation of the designed energy functional is first derived and the flow minimizing the energy is then obtained. Following the work by Caselles et al [CKS95],and Malladi et al [MSV95, MSV93], the contours to be detected and tracked are modeled as geodesic active contours evolving toward the minimum of the designed energy, under the influence of internal and external image dependent forces. Using the level set formulation scheme of Osher and Sethian [OS88] complex curves can be detected and tracked and topological changes for the evolving curves are naturally managed. To reduce the computational cost required by a direct implementation of the formulation scheme of Osher and Sethian [OS88], a new approach exploiting aspects from the classical Narrow Band [AS95] and Fast Marching [Set96] methods is proposed and favorably compared to them. In order to further reduce the CPU time, a multi-scale approach has also been considered. Very promising experimental results are provided using real video sequences.
ICCV	A Curvature-Based Approach to Contour Motion Estimation.	Jong Seung Park,Joon Hee Han	1998	We present a novel method of velocity field estimation for points on moving contours in an image sequence. The method determines the corresponding point in the next image frame by considering curvature changes at each point on a contour. In previous methods, there are errors in estimation for the points which have low curvature variations since those methods compute the solutions by approximatingthe normal component of optical flow. The proposed method computes optical flow vectors of contour points by minimizing the curvature changes. As a first step, snakes are used to locate smooth curves in 2D imagery. Then, the extracted curves are tracked continuously. We excluded the rearranging process in snakes and allowed the snaxel distance to vary. Each point on a contour has a unique corresponding point in thenext frame. Experimental results showed that the proposed method computes accurate optical flow vectors for various moving contours.
ICCV	A Unified Factorization Algorithm for Points, Line Segments and Planes with Uncertainty Models.	Daniel D. Morris,Takeo Kanade	1998	In this paper we present a unified factorization algorithm for recovering structure and motion from image sequences by using point features, line segments and planes. This new formulation is based on directional uncertainty model for features. Points and line segments are both described by the same probabilistic models and so can be recovered in the same way. Prior information on the coplanarity of features is shown to fit naturally into the new factorization formulation and provides additional constraints for the shape recovery. This formulation leads to a weighted least squares motion and shape recovery problem which is solved by an efficient quasi-linear algorithm. The statistical uncertainty model also enables us to recover uncertainty estimates for the reconstructed three dimensional feature locations.
ICCV	Robust Tracking with Spatio-Velocity Snakes: Kalman Filtering Approach.	Natan Peterfreund	1998	Using results from robust Kalman filtering, we present a new Kalman filter-based snake model for tracking of nonrigid objects in combined spatio-velocity space. The proposed model is the stochastic version of the velocity snake which is an active contour model for combined tracking of position and velocity of nonrigid boundaries. The proposed model uses image gradient and optical flow measurements along the contour as system measurements. An optical-flow based measurement error is used to detect and reject image measurements which correspond to image clutter or to other objects. The method was applied to object tracking ofboth rigid and nonrigid objects, resulting in good tracking results and robustness to image clutter, occlusions and numerical noise.
ICCV	Bias-Corrected Optical Flow Estimation for Road Vehicle Tracking.	Hans-Hellmut Nagel,Michael Haag	1998	Model-based vehicle tracking in traffic image sequences can be made more robust by matching expected displacementrates of vehicle surface points to optical flow (OF) vectors computed from an image sequence. The capability to track vehicles uninterruptedly in thismanner over extended image sequences results in the ability to investigate even small errors in OF estimation. It turns out that the OF magnitudes are systematically underestimated. The ¿ albeit small ¿ bias can be corrected by analyzing the influence of explicitly modeled grey value noise on the precision of OF values estimated by means of the neighborhood sampling method.
ICCV	Self-Calibration and Metric Reconstruction in Spite of Varying and Unknown Internal Camera Parameters.	Marc Pollefeys,Reinhard Koch,Luc J. Van Gool	1998	In this paper the feasibility of self-calibration in the presence of varying internal camera parameters is under investigation. A self-calibration method is presented which efficiently deals with all kinds of constraints on the internal camera parameters. Within this framework a practical method is proposed which can retrieve metric reconstruction from image sequences obtained with uncalibrated zooming/focusing cameras. The feasibility of the approach is illustrated on real and synthetic examples.
ICCV	Constructing Virtual Worlds Using Dense Stereo.	P. J. Narayanan,Peter Rander,Takeo Kanade	1998	We present Virtualized Reality, a technique to create virtual worlds out of dynamic events using densely distributed stereo views. The intensity image and depth map for each camera view at each time instant are combined to form a Visible Surface Model. Immersive interaction with the virtualized event is possible using a dense collection of such models. Additionally, aComplete Surface Model of each instant can be built by merging the depth maps from different cameras into a common volumetric space. The corresponding model is compatible with traditional virtual models and can be interacted with immersively using standard tools. Because both VSMs and CSMs are fully three-dimensional, virtualized models can also be combined and modi.ed to build larger, more complex environments, an important capability for many non-trivial applications. We present results from 3D Dome, our facility to create virtualized models.
ICCV	Comparing Curved-Surface Range Image Segmenters.	Mark W. Powell,Kevin W. Bowyer,Xiaoyi Jiang,Horst Bunke	1998	This work focuses on creating a framework for objectively evaluating the performance of range image segmentation algorithms. The algorithms are evaluated in terms of correct segmentation, over- and under- segmentation, missed and noise regions. A set of images with ground truth was created for this work. The images were captured using a structured light scanner. Images used in the evaluation contain planar, spherical, cylindrical, toroidal andconical surface patches. The different surface patches in each image were manually identified to establish ground truth for performance evaluation. Two segmentation algorithms from the literature are compared.
ICCV	A Cubist Approach to Object Recognition.	Randal C. Nelson,Andrea Selinger	1998	We describe an appearance-based object recognition system using a keyed, multi-level context representation reminiscent of certain aspects of cubist art. Specifically, we utilize distinctive intermediate-level features in this case automatically extracted 2-Dboundary fragments, as keys, which are then verified within a local context, and assembled within a loose global context to evoke an overall percept. This system demonstrates good recognition of a variety of 3-D shapes, ranging from sports cars and fighter planes to snakes and lizards with full orthographic invariance. We report the results of large-scale tests, involving over 2000 separate test images, that evaluate performance with increasing number of items inthe database, in the presence of clutter, background change, and occlusion, and also the results of some generic classification experiments where the system is tested on objects never previously seen or modeled. To our knowledge, the results we report are the best in the literature for full-sphere tests of general shapes with occlusion and clutter resistance.
ICCV	Stereo with Mirrors.	Sameer A. Nene,Shree K. Nayar	1998	In this paper, we propose the use of mirrors and a single camera for computational stereo. When compared to conventional stereo systems that use two cameras, our method has a number of significant advantages such as wide field of view, single viewpoint projection, identical camera parameters and ease of calibration. We propose four stereo systems that use a single camera pointed towards planar, ellipsoidal, hyperboloidal, and paraboloidal mirrors. In each case, we present a derivation of the epipolar constraints. Next, we attempt to understand what can be seen by each system and formalize the notion of field of view. We conclude with two experiments to obtain 3-D structure. In the first we use a pair of planar mirrors, and in the second a pair of paraboloidal mirrors. The results of our experiments demonstrate the viability of stereo using mirrors.
ICCV	Wide Baseline Stereo Matching.	Philip Pritchett,Andrew Zisserman	1998	The objective of this work is to enlarge the class of camera motions for which epipolar geometry and image correspondences can be computed automatically. This facilitates matching between quite disparate views ¿ wide baseline stereo.Two extensions are made to the current small baseline algorithms: first, and most importantly, a viewpoint invariant measure is developed for assessing the affinity of corner neighbourhoods over image pairs; second, algorithms are given for generating putative corner matches between image pairs using local homographies. Two novel infrastructure developments are also described, the automatic generation of local homographies, and the combination of possibly conflicting sets of matches prior to RANSAC estimation.The wide baseline matching algorithm is demonstrated on a number of image pairs with varying relative motion, and for different scene types. All processing is automatic.
ICCV	Reading between the lines: A method for extracting dynamic 3D with Texture.	Marc Proesmans,Luc J. Van Gool,Filip Defoort	1998	A method is presented that extracts the 3D shape of Objects, together with the surface texture. Both shape and tecture are obtained from a single image. The underlying principle is based on an active technique. A high resolution pattern is projected onto the object and the deformations as observed by a single camera yield the 3rd dimension. Furthermore, the surface texture is extracted from the same image. Because the whole procedure is based on a single image, a frame-by-frame reconstruction of a video taken with the pattern projected throughout, yields 3Dshape Dynamics. The paper sketches the complete system but focuses on the problem of extraction.
ICCV	Multiscale Annealing for Real-Time Unsupervised Texture Segmentation.	Jan Puzicha,Joachim M. Buhmann	1998	We derive real-time global optimization algorithms for several clustering optimization methods used in unsupervised texture segmentation. Speed is achieved by exploiting the topological relation of features to design a multiscale optimization technique, while accuracy and global optimization properties are provided by a deterministic annealing method. Coarse grained cost functions are derived for both central and sparse pairwise clustering, where the problem of coarsening sparse random graphs is solved by the concept of structured randomization. Annealing schedules and coarse-to-fine optimization are tightly coupled by a statistical convergence criterion derived from computational learning theory. The algorithms are benchmarked on Brodatz-like micro-texture mondrians. Results are presented for an autonomous robotics application.
ICCV	Three Dimensional MR Brain Segmentation.	Wiro J. Niessen,Koen L. Vincken,Joachim Weickert,Max A. Viergever	1998	In MR brain images, segmentation using intensity values is severely limited owing to field inhomogeneities, susceptibility artifacts and partial volume effects. Edge based segmentation methods suffer from spurious edges and gaps in boundaries. A method is presented which combines the advantages of edge based and region based segmentation. First a multiscale image representation is constructed which favors intra-tissue diffusion over inter-tissue diffusion by exploiting local contrast. Subsequently a multiscale linking model (the hyperstack) is used to group voxels into a number of segments. This facilitates segmentation of grey matter, white matter and cerebrospinal fluid with minimal user interaction. Using a supervised segmentation technique and MR simulations of a brain phantom as validation it is shown that the errors are in the order of or smaller than reported in literature.
ICCV	Linear N>=4-Point Pose Determination.	Long Quan,Zhong-Dan Lan	1998	The determination of the position and the orientation of the camera from the known correspondences of the reference points and the image points is known as the problem of pose estimation in computer vision or space resection in photogrammetry. It is well known that using 3 corresponding points has at most 4 solutions. Less appears to be known about the cases of 4 and 5 points. In this paper, we describe linear solutions that always give the unique solution to 4-point and 5-point pose determination for the reference points not lying on the critical confogurations. The same linear method can also be extended to any n ¿ 5 points.The robustness and accuracy of the method are experimented both on simulated and real images.
ICCV	Learning Dynamical Models Using Expectation-Maximisation.	Ben North,Andrew Blake	1998	"Tracking with deformable contours in a filtering frame-work requir esa dynamical model for prediction. For any given application, tracking is improved by having an accurate model, learned from training data. We develop a method for learning dynamical models from training sequences, explicitly taking account of the fact that training data are noisy measurements and not true states. By introducing an ""augmented-state smoothing filter"" , we show how the technique of Expectation-Maximisation can be applied to this problem, and show that the resulting algorithm produces more robust and accurate tracking."
ICCV	A Cooperative Framework for Segmentation Using 2D Active Contours and 3D Hybrid Models as Applied to Branching Cylindrical Structures.	"Thomas O'Donnell,Marie-Pierre Dubuisson-Jolly,Alok Gupta"	1998	A Cooperative Framework for Segmentation Using 2D Active Contours and 3D Hybrid Models as Applied to Branching Cylindrical Structures.
ICCV	Optimal Recovery of Depth from Defocused Images Using an MRF Model.	A. N. Rajagopalan,Subhasis Chaudhuri	1998	A MAP-MRF based scheme is proposed for recovering the depth and the focused image of a scene from two defocused images. The space-variant blur parameter and the focused image of the scene are both modeled as MRFs and their MAP estimates are obtained using simulated annealing. The performance of the proposed scheme is tested on both synthetic as well as real data and the estimates of the depth are found to be better than that of the existing window-based technique.
ICCV	Finding Faces in Photographs.	A. N. Rajagopalan,K. Sunil Kumar,Jayashree Karlekar,R. Manivasakan,M. Milind Patil,Uday B. Desai,P. G. Poonacha,Subhasis Chaudhuri	1998	Two new schemes are presented for finding human faces in a photograph. The first scheme approximates the unknown distributions of the face and the face-like manifolds using higher order statistics (HOS). An HOS-based data clustering algorithm is also proposed. In the second scheme, the face to non-face and non-face to face transitions are learnt using a hidden Markov model (HMM). The HMM parameters are estimated corresponding to a given photograph and the faces are located by examining the optimal state sequence of the HMM. Experimental results are presented on the performance of both the schemes.
ICCV	Retrieving Images by Appearance.	S. Ravela,R. Manmatha	1998	A system to retrieve images using a description of the image intensity surface is presented. Gaussian derivative filters at several scales are applied to the image and low order 2D differential invariants are computed. The resulting multi-scale representation is indexed for rapid retrieval. Queries are designed by the users from an example image by selecting appropriate regions. The invariant vectors corresponding to these regions are matched with the database counter-parts both in feature and coordinate space. This yields a match score per image. Images are sorted by the match score and displayed. Experiments conducted with over 1500 images of objects embedded in arbitrary backgrounds are described. It is observed that images similar in appearance and whose viewpoint is within small view variations of the query can be retrieved with an average precision1 of 56%.
ICCV	Agent Orientated Annotation in Model Based Visual Surveillance.	Paolo Remagnino,Tieniu Tan,Keith D. Baker	1998	"The paper presents an agent based surveillance system for use in monitoring scenes involving both pedestrians and vehicles. The surveillance system supplies textual descriptions for the dynamic activity occurring in the 3D world. These are derived by means of dynamic andprobabilistic inference based on geometric information provided by a vision system that tracks vehicles and pedestrians. The symbolic scene annotation is given at two major levels of description: the object level and the interobject level. At object level, each tracked object(pedestrian or vehicle) is assigned a behaviour agent which uses a Bayesian network to infer the fundamental features of the objects' trajectory, and continuously updates its textual description. The inter-object interaction level is interpreted by a situation agent which is created dynamically when two objects are in close proximity."
ICCV	Two-Dimensional Affine Invariants that Distribute Uniformly and can be Tuned to any Convex Feature Domain.	Isidore Rigoutsos	1998	Two-Dimensional Affine Invariants that Distribute Uniformly and can be Tuned to any Convex Feature Domain.
ICCV	Automatic Registration of 3-D Ultrasound Images.	Robert Rohling,Andrew H. Gee,Laurence H. Berman	1998	One of the most promising applications of 3-D ultrasound lies in the visualisation and volume estimation of internal 3-D structures. Unfortunately, artifacts and speckle make automatic analysis of the data difficult. In this paper we investigate the use of 3-D spatial compounding to improve data quality, and find that accurate registration is the key. A correlation-based registration technique is applied to 3-D ultrasound data acquired from in-vivo examinations of a human gall bladder. We find that the registration technique performs well, and visualisation and segmentation of the compounded data are clearly improved.
ICCV	Thresholding for Change Detection.	Paul L. Rosin	1998	Image differencing is used for many applications involving change detection. Although it is usually followed by a thresholding operation to isolate regions of change there are few methods available in the literature specific to (and appropriate for) change detection. We describe four different methods for selecting thresholds that work on very different principles. Either the noise or the signal is modelled, and the model covers either the spatial or intensity distribution characteristics. The methods are: 1/ a Normal model is used for the noise intensity distribution, 2/ signal intensities are tested by making local intensity distribution comparisons in the two image frames (i.e. the difference map is not used), 3/ the spatial properties of the noise are modelled by a Poisson distribution, and 4/ the spatial properties of the signal are modelled as a stable number of regions (or stable Euler number).
ICCV	Motion Segmentation and Tracking Using Normalized Cuts.	Jianbo Shi,Jitendra Malik	1998	We propose a motion segmentation algorithm that aims to break a scene into its most prominent moving groups. A weighted graph is constructed on the image sequence by connecting pixels that are in the spatiotemporal neighborhood of each other. At each pixel, we define motion profile vectors which capture the probability distrobution of the image velocity. The distance between motion profiles is used to assign a weight on the graph edges. Using normalized cuts we find the most salient partitions of the spaciotemporal graph formed by the image sequence. For segmenting long image sequences, we have developed a recursive update procedure that incorporates knowledge of segmentation in previous frames for efficiently finding the group correspondence in the new frame.
ICCV	Universal Mosaicing using Pipe Projection.	Benny Rousso,Shmuel Peleg,Ilan Finci,Alex Rav-Acha	1998	"Video mosaicing is commonly used to increase the visual field of view by pasting together many video frames. Existing mosaicing methods are effective only in very limited cases where the image motion is almost a uniform translation or the camera performs a pure pan. Forward camera motion or camera zoom are very problematic for traditional mosaicing.A mosaicing methodology to allow image mosaicing in the most general cases is presented, where frames in the video sequence are transformed such that the optical flow becomes parallel. This transformation is an oblique projection of the image into a ""viewing pipe"" whose central axis is the trajectory of the camera.The ""pipe projection"" enables to define high-quality mosaicing even for the most challenging cases of forward motion and of zoom. In addition, view interpolation, generating dense intermediate views, is used to overcome parallax effects."
ICCV	View-Based Object Matching.	Ali Shokoufandeh,Ivan Marsic,Sven J. Dickinson	1998	"We introduce a novel view-based object representation, called the saliency map graph (SMG), which captures the salient regions of an object view at multiple scales using a wavelet transform. This compact representation is highly invariant to translation, rotation (image and depth), and scaling, and offers the locality of representation required for occluded object recognition. To compare two saliency map graphs, we introduce two graph similarity algorithms. The first computes the topological similarity between two SMG's, providing a coarse-level matching of two graphs. The second computes the geometrical similarity betweentwo SMG's, providing a fine-level matching of two graphs. We test and compare these two algorithms on a large database of model object views."
ICCV	A Maximum-Flow Formulation of the N-Camera Stereo Correspondence Problem.	Sébastien Roy,Ingemar J. Cox	1998	This paper describes a new algorithm for solving the N-camera stereo correspondence problem by transforming it into a maximum-flow problem. Once solved, the minimum-cut associated to the maximum-flow yields a disparity surface for the whole image at once. This global approach to stereo analysis provides a more accurate and coherent depth map than the traditional line-by-line stereo. Moreover, the optimality of the depth surface is guaranteed and can be shown to be a generalization of the dynamic programming approach that is widely used in standard stereo. Results show improved depth estimation as well as better handling of depth discontinuities. While the worst case running time is O(n2d2 log(nd)), the observed average running time is O(n1.2d1.3) for an image size of n pixels and depth resolution d.
ICCV	A Metric for Distributions with Applications to Image Databases.	Yossi Rubner,Carlo Tomasi,Leonidas J. Guibas	1998	"We introduce a new distance between two distributions that we call the Earth Mover's Distance (EMD), which reflects the minimal amount of work that must be performed to transform one distribution into the other by moving ""distribution mass"" around. This is a special case of the transportation problem from linear optimization, for which efficient algorithms are available. The EMD also allows for partial matching. When used to compare distributions thathave the same overall mass, the EMD is a true metric, and has easy-to-compute lower bounds. In this paper we focus on applications to image databases, especially color and texture. We use the EMD to exhibit the structure of color-distribution and texture spaces by means of Multi-Dimensional Scaling displays. We also propose a novel approach to the problem of navigating through a collection of color images, which leads to a new paradigm for image database search."
ICCV	Construction and Refinement of Panoramic Mosaics with Global and Local Alignment.	Heung-Yeung Shum,Richard Szeliski	1998	This paper presents techniques for constructing full view panoramic mosaics from sequences of images. Our representation associates a rotation matrix (and optionally a focal length) with each input image, rather than explicitly projecting all of the images onto a common surface (e.g., a cylinder). In order to reduce accumulated registration errors, we apply global alignment (block adjustment) to the whole sequence of images, which results in an optimal image mosaic (in the least-squares sense). To compensate for small amounts of motion parallax introduced by translations of the camera and other unmodeled distortions, we develop a local alignment (deghosting) technique which warps each image based on the results of pairwise local image registrations. By combining both global and local alignment, we significantly improve the quality of our image mosaics, thereby enabling the creation of full view panoramic mosaics with hand-held cameras.
ICCV	Shock Graphs and Shape Matching.	Kaleem Siddiqi,Ali Shokoufandeh,Sven J. Dickinson,Steven W. Zucker	1998	"We have been developing a theory for the generic representation of 2-D shape, where structural descriptions are derived from the shocks ""singularities"" of a curve evolution process, acting on bounding contours. We now apply the theory to the problem of shape matching. The shocks are organized into a directed, acyclic shock graph, and complexity is managed by attending to the most significant ""central"" shape components first. The space of all such graphs is highly structured and can be characterized by the rules of a shock graph grammar. The grammar permits a reduction of a shock graph to a unique rooted shock tree. We introduce a novel tree matching algorithm which finds the best set of corresponding nodes between two shock trees in polynomial time. Using a diverse database of shapes, we demonstrate our system's performance under articulation, occlusion, and changes in viewpoint."
ICCV	"Hyperbolic ""Smoothing"" of Shapes."	Kaleem Siddiqi,Allen Tannenbaum,Steven W. Zucker	1998	We have been developing a theory of generic 2-D shape based on a reaction-diffusion model from mathematical physics. The description of a shape is derived from the singularities of a curve evolution process driven by the reaction (hyperbolic) term. The diffusion (parabolic) term is related to smoothing and shape simplification. However, the unification of the two is problematic, because the slightest amount of diffusion dominates and prevents the formation of generic first-order shocks. The technical issue is whether it is possible to smooth a shape, in any sense, without destroying the shocks. We now report a constructive solution to this problem, by embedding the smoothing term in a global metric against which a purely hyperbolic evolution is performed from the initial curve. This is a new flow for shape, that extends the advantages of the original one. Specific metrics are developed, which lead to a natural hierarchy of shape features, analogous to the simplification one might perceive when viewing an object from increasing distances. We illustrate our new flow with a variety of examples.
ICCV	Egomotion Estimation Using Log-Polar Images.	César Silva,José Santos-Victor	1998	We address the problem of egomotion estimation of a monocular observer moving with arbitrary translation and rotation in an unknown environment, using log-polar images.The method we propose is uniquely based on the spatio-temporal image derivatives, or the normal flow. Thus, we avoid computing the complete optical flow field, which is an ill-posed problem due to the aperture problem. We use a search paradigm based on geometric properties of the normal flow field, and consider a family of search subspaces to estimate the egomotion parameters. These algorithms are particularly well-suited for the log-polar image geometry, as we use a selection of special normal flow vectors with simple representation in logpolar coordinates. This approach highlights the close coupling between algorithmic aspects and the sensor geometry (retina physiology), often found in nature.Finally, we present and discuss a set of experiments, for various kinds of camera motions, which show encouraging results.
ICCV	Visual Routines for Autonomous Driving.	Garbis Salgian,Dana H. Ballard	1998	The paper describes visual routines based on models of color and shape, as well as crucial issues involving the sche duling of such routines. The visual routines are developed in a unique platform. The view from a car driving in a simulated world is fed into a Datacube pipeline video processor. The use of this simulation provides a flexible environment from which to set crucial image processing parameters of the individual routines. In addition to the simulations are also tested in similar images generated by driving in the real world, to assure the generalizability of the simulation.
ICCV	Recognizing Novel 3-D Objects Under New Illumination and Viewing Position Using a Small Number of Examples.	Erez Sali,Shimon Ullman	1998	Recognizing Novel 3-D Objects Under New Illumination and Viewing Position Using a Small Number of Examples.
ICCV	A Two-Stage Robust Statistical Method for Temporal Registration from Features of Various Type.	Gilles Simon,Marie-Odile Berger	1998	A model registration system capable of tracking an object, the model of which is known, in an image sequence is presented. It integrates tracking, pose determination and updating of the visible features. The which handles various features (points, lines and free-form curves) in a very robust way and is able to give a correct estimate of the pose even when tracking errorsoccur. The reliability of the system is shown on an augmented reality project.
ICCV	Bilinear Voting.	Guillermo Sapiro	1998	"A geometric-vision approach to solve bilinear problems in general, and the color constancy and illuminant estimation problem in particular, is presented in this paper. We show a general framework, based on ideas from the generalized (probabilistic) Hough transform, to estimate the unknown variables in the bilinear form. In the case of illuminant and reflectance estimation in natural images, each image pixel ""votes"" for possible illuminants (or reflectance), and the estimation is based on cumulative votes. In the general case, the voting is for the parameters of the bilinear model. The framework is natural for the introduction of physical constraints. For the case of illuminant estimation, we briefly show the relation of this work with previous algorithms for color constancy, and present examples."
ICCV	Fish-Scales: Representing Fuzzy Manifolds.	Radim Sára,Ruzena Bajcsy	1998	Fish-Scales: Representing Fuzzy Manifolds.
ICCV	Affine Reconstruction of Curved Surfaces from Uncalibrated Views of Apparent Contours.	Jun Sato,Roberto Cipolla	1998	In this paper, we show that even if the camera is uncalibrated, and its translational motion is unknown, curved surfaces can be reconstructed from their apparent contours up to a 3D affine ambiguity. Furthermore, we show that even if the reconstruction is non-metric (non-Euclidean), we can still extract useful information for many computer vision applications just from the apparent contours.We first show that if the camera undergoes pure translation (unknown direction and magnitude), the epipolar geometry can be recovered from the apparent contours without using any search or optimisation process. The extracted epipolar geometry is next used for reconstructing curved surfaces from the deformations of the apparent contours viewed from uncalibrated cameras. The result is applied to distinguishing curved surfaces from fixed features in images. It is also shown that the time-to-contact to the curved surfaces can be computed from simple measurements of the apparent contours. The proposed method is implemented and tested on real images of curved surfaces.
ICCV	Separation of Transparent Layers Using Focus.	Yoav Y. Schechner,Nahum Kiryati,Ronen Basri	1998	Consider situations where the depth at each point in the scene is multi-valued, due to the presence of a virtual image semi-reflected by a transparent surface. The semi-reflected image is linearly superimposed on the image of the object that is behind the transparent surface. A novel approach is proposed for the recovery of the superimposed layers. By searching for the images in which either of the objects (layers) is focused, the transparent areas are detected and an estimate of the depth map of each layer is obtained. As a result of the focusing, an initial separation of the layers is achieved. The separation is enhanced via mutual blurring of the perturbing components in the images, based on the depths estimate and the parameters of the imaging system.
ICCV	Transinformation for Active Object Recognition.	Bernt Schiele,James L. Crowley	1998	This article develops an analogy between object recognition and the transmission of information through a channel based on the statistical representation of the appearances of 3D objects. This analogy provides a means to quantitatively evaluate the contribution of individual receptive field vectors, and to predict the performance of the object recognition process. Transinformation also provides a quantitative measure of the discrimination provided by each viewpoint, thus permitting the determination of the most discriminant viewpoints. As an application, the article develops an active object recognition algorithm which is able to resolve ambiguities inherent in a single-view recognition algorithm.
ICCV	Comparing and Evaluating Interest Points.	Cordelia Schmid,Roger Mohr,Christian Bauckhage	1998	Many computer vision tasks rely on feature extraction. Inter est points are such features. This paper shows that interest points are geometrically stable under different transformations and have high information content (distinctiveness). These two properties make interest points very successful in the context of image matching. To measure these two properties quantitatively, we introduce two evaluation criteria: repeatability rate and information content.The quality of the interest points depends on the detector used. In this paper several detectors are compared according to the criteria specified above. We determine which detector gives the best results and show that it satisfies the criteria well.
ICCV	Computing Ritz Approximations of Primary Images.	Haim Schweitzer	1998	"Ritz vectors approximate eigenvectors that are a common choice for primary images in content based indexing. They can be computed efficiently even when the images are accessed through slow communication such as the Internet. We develop an algorithm that computes Ritz vectors in one pass through the images. When iterated, the algorithm can recover the exact eigenvectors. In applications to image indexing and learning it may be necessary to compute primary images for indexing many sub-categories of the image set The proposed algorithm can compute these additional primary images ""offline"", without the image data much more costly even when access to the images is inexpensive."
ICCV	Indexing Images by Trees of Visual Content.	Haim Schweitzer	1998	"An unsupervised algorithm for arranging an image database as a binary tree is described. Tree nodes are associated with image subsets, maintaining the property that the similarity among the images associated with the children of a node is higher than the similarity among the images associated with the parent node. Experiments with datasets of hundreds and thousands of images show that shallow trees can produce clustering into ""meaningful"" classes. Visual-content search trees can be used to automate image retrieval by content, orhelp a human to interactively search for images."
ICCV	Active Blobs.	Stan Sclaroff,John Isidoro	1998	"A new region-based approach to nonrigid motion tracking is described. Shape is defined in terms of a deformable triangular mesh that captures object shape plus a color texture map that captures object appearance. Photometric variations are also modeled. Nonrigid shape registration and motion tracking are achieved by posing the problem as an energy-based, robust minimization procedure. The approach provides robustness to occlusions, wrinkles, shadows, and specular highlights. The formulation is tailored to take advantage of texture mapping hardware available in many workstations, PC's, and game consoles. This enablesnonrigid tracking at speeds approaching video rate."
ICCV	Plenoptic Image Editing.	Steven M. Seitz,Kiriakos N. Kutulakos	1998	"This paper presents a new class of interactive image editing operations designed to maintain consistency between multiple images of a physical 3D scene. The distinguishing feature of these operations is that edits to any one image propagate automatically to all other images asif the (unknown) 3D scene had itself been modified. The modified scene can then be viewed interactively from any other camera viewpoint and under different scene illuminations. The approach is useful first as a power-assist that enables a user to quickly modify many images by editing just a few, and second as a means for constructing and editing image-based scene representations by manipulating a set of photographs. The approach works by extending operations like image painting, scissoring, and morphing so that they alter a scene's generalized plenoptic function in a physically-consistent way, thereby affecting scene appearance from all viewpoints simultaneously. A key element in realizing these operations is a new volumetric decomposition technique for reconstructing an scene's plenoptic function from an incomplete set of camera viewpoints."
ICCV	Automatic Model Construction, Pose Estimation, and Object Recognition from Photographs using Triangular Splines.	Steve Sullivan,Jean Ponce	1998	This paper proposes a method for automatically constructing triangular G1 spline models of complex three-dimensional objects from a few registered photographs. These models are used for pose estimation from monocular silhouette data and they form the basis for a simple recognition strategy. The proposed approach is demonstrated by several experiments.
ICCV	Recognition of 3D Free-form Objects Using Segment-based Stereo Vision.	Yasushi Sumi,Yoshihiro Kawai,Takashi Yoshimi,Fumiaki Tomita	1998	We propose a new method to recognize 3D free-form objects from their apparent contours. It is the extension of our established method to recognize objects with fixed edges. Object models are compared with 3D boundaries which are extracted by segment-based stereo vision. Based on the local shapes of the boundaries, candidate transformations are generated. The candidates are verified and adjusted based on the whole shapes of the boundaries. Themodels are built from all-around range data of the objects. Experimental results show the effectiveness of the method.
ICCV	Stereo Matching with Transparency and Matting.	Richard Szeliski,Polina Golland	1998	This paper formulates and solves a new variant of the stereo correspondence problem: simultaneously recovering the disparities, true colors, and opacities of visible surface elements. This problem arises in newer applications of stereo reconstruction, such as view interpolation and the layering of real imagery with synthetic graphics for special effects and virtual studio applications. While this problem is intrinsically more difficult than traditional stereo correspondence, where only the disparities are being recovered, it provides a principled way of dealing with commonly occuring problems such as occlusions and the handling of mixed (foreground/background) pixels near depth discontinuities. It also provides a novel means for separating foreground and background objects (matting) without the use of a special blue screen. We formulate the problem as the recovery of colors and opacities in a generalized 3-D (x, y, d) disparity space, and solve the problem using a combination of initial evidence aggregation followed by iterative energy minimization.
ICCV	Integrated Surface, Curve and Junction Inference from Sparse 3-D Data Sets.	Chi-Keung Tang,Gérard G. Medioni	1998	We are interested in descriptions of 3-D data sets, as obtained from stereo or a 3-D digitizer. We therefore consider as input a sparse set of points, possibly associated with orientation information. In this paper, we address the problem of inferring integrated high-level descriptionssuch as surfaces, curves, and junctions from a sparse point set. While the method described in [5], [6] provides excellent results for smooth structures, it only detects discontinuities, but does not localize them. For precise localization, we propose a non-iterative cooperative algorithm in which surfaces, curves, and junctions work together: Initial estimates are computed based on [5], [6], where each point in the given sparse and possibly noisy point set is convolved with a predefined vector mask to produce dense saliency maps. These maps serve as input to our novel maximal surface and curve marching algorithms for initial surface and curve extraction. Refinement of initial estimates is achieved by hybrid voting using excitatory and inhibitory fields for inferring reliable and natural extension so that surface/curve and curve/junction discontinuities are preserved. Results on several synthetic as well as real data sets are presented.
ICCV	Local Symmetries of Shapes in Arbitrary Dimension.	Sibel Tari,Jayant Shah	1998	"Motivated by a need to define an object-centered reference system determined by the most salient characteristics of the shape, many methods have been proposed, all of which directly or indirectly involve an axis about which the shape is locally symmetric. Recently, a function v, called ""the edge strength function"", has been successfully used to determine efficiently the axes of local symmetries of 2-d shapes. The level curves of v are interpreted as succesively smoother versions of the initial shape boundary. The local minima of the absolute gradient \left\| {\nabla v} \right\| along the level curves of v are shown to be a robust criterion for determining the shape skeleton. More generally, at an extremal point of \left\| {\nabla v} \right\| along a level curve, the level curve is locally symmetric with respect to the gradient vector \nabla v. That is, at such a point, the level curve is approximately a conic section whose one of the principal axes coincides with the gradient vector. Thus, the locus ofthe extremal points of \left\| {\nabla v} \right\| along the level curves determines the axes of local symmetries of the shape. In this paper, we extend this method to shapes of arbitrary dimension."
ICCV	Design of Multi-Parameter Steerable Functions Using Cascade Basis Reduction.	Patrick C. Teo,Yacov Hel-Or	1998	A new cascade basis reduction method of computing the optimal least-squares set of basis functions to steer a given function is presented. The method combines the Lie group-theoretic and the singular value decomposition approaches such that their respective strengthscomplement each other. Since the Lie group-theoretic approach is used, the set of basis and steering functions computed can be expressed in analytic form. Because the singular value decomposition method is used, this set of basis and steering functions is optimal in the least-squares sense. Most importantly, the computational complexity in designing basis functions for transformation groups with large numbers of parameters is significantly reduced. The efficiency of the cascade basis reduction method is demonstrated by designing a set of basis functions to steer a Gabor function under the four-parameter linear transformation group.
ICCV	Segmentating Cortical Gray Matter for Functional MRI Visualization.	P. C. Teo,Guillermo Sapiro,Brian A. Wandell	1998	We describe a system that is being used to segment gray matter and create connected cortical representations from MRI. The method exploits knowledge of the anatomy of the cortex and incorporates structural constraints into the segmentation. First, the white matterand CSF regions in the MR volume are segmented using some novel techniques of posterior anisotropic diffusion. Then, the user selects the cortical white matter component of interest, and its structure is verified by checking for cavities and handles. After this, a connected representation of the gray matter is created by a constrained growing-out from the white matter boundary. Because the connectivity is computed, the segmentation can be used as input to several methods of visualizing the spatial pattern of cortical activity within gray matter. In our case, the connected representation of gray matter is used to create a representation of theflattened cortex. Then, fMRI measurements are over-laid on the flattened representation, yielding a representation of the volumetric data within a single image.
ICCV	Bilateral Filtering for Gray and Color Images.	Carlo Tomasi,Roberto Manduchi	1998	Bilateral filtering smooths images while preserving edges, by means of a nonlinear combination of nearby image values. The method is noniterative, local, and simple. It combines gray levels or colors based on both their geometric closeness and their photometric similarity, and prefers near values to distant values in both domain and range. In contrast with filters that operate on the three bands of a color image separately, a bilateral filter can enforce the perceptual metric underlying the CIE-Lab color space, and smooth colors and preserve edges in a way that is tuned to human perception. Also, in contrast with standard filtering, bilateral filtering produces no phantom colors along edges in color images, and reduces phantom colors where they appear in the original image.
ICCV	Maintaining Multiple Motion Model Hypotheses Through Many Views to Recover Matching and Structure.	Philip H. S. Torr,Andrew W. Fitzgibbon,Andrew Zisserman	1998	Maintaining Multiple Motion Model Hypotheses Through Many Views to Recover Matching and Structure.
ICCV	Robust Computation and Parametrization of Multiple View Relations.	Philip H. S. Torr,Andrew Zisserman	1998	A new method is presented for robustly estimating multiple view relations from image point correspondences. There are three new contributions, the first is a general purpose method of parametrizing these relations using point correspondences. The second contribution is the formulation of a common Maximum Likelihood Estimate (MLE) for each of the multiple view relations. The parametrization facilitates a constrained optimization to obtain the MLE. The third contribution is a new robust algorithm, MLESAC, for obtaining the point correspondences.The method is general and its use is illustrated for the estimation of fundamental matrices, image to image homographics and quadratic transformations.Results are given for both synthetic and real images. It is demonstrated that the method gives results equal or superior to previous approaches.
ICCV	A Cascaded Hough Transform as an Aid in Aerial Image Interpretation.	Tinne Tuytelaars,Luc J. Van Gool,Marc Proesmans,Theo Moons	1998	A Cascaded Hough Transform as an Aid in Aerial Image Interpretation.
ICCV	State Space Construction for Behavior Acquisition in Multi Agent Environments with Vision and Action.	Eiji Uchibe,Minoru Asada,Koh Hosoda	1998	"This paper proposes a method which estimates the relationships between learner's behaviors and other agents' ones in the environment through interactions (observation and action) using the method of system identification. In order to identify the model of each agent, Akaike's Information Criterion is applied to the results of Canonical Variate Analysis for the relationship between the observed data in terms of action and future observation. Next, reinforcement learning based on the estimated state vectors is performed to obtain the optimal behavior. The proposed method is applied to a soccer playing situation, where a rolling ball and other moving agents are well modeled and the learner's behaviors are successfully acquired by the method. Computer simulations and real experiments are shown and a discussion is given."
ICCV	Grouping Based on Projective Geometry Constraints and Uncertainty.	Sven Utcke	1998	"The process of grouping and subsequently recognising objects in cluttered images is one laden with difficulties, however, results can be greatly enhanced if the inherent uncertainty of image-features is taken into account. This paper shows that starting with the individual edgel's uncertainty it is possible to calculate covariance-information for all derived quantities. This information can be used to choose between competing algorithms, selecting the one that produces the more reliable results, but also as an aid during the recognition process. The consequent application of error-propagation leads to a new formulation for the calculation of the cross-ratio, which is both robust and efficient in dealing with measured lines, and does notrequire knowledge about the vanishing point. Extensive Monte-Carlo simulations as well as the application to images of cluttered street-scenes demonstrate the robustness and suitability of the approach."
ICCV	Snake Pedals: Geometric Models with Physics-Based Control.	Baba C. Vemuri,Yanlin Guo	1998	"In this paper, we introduce a novel geometric shape modeling scheme which allows for representation of global and local shape characteristics of an object. Geometric models are traditionally well suited for representing global shapes but not the local details. However, in this paper we propose a powerful geometric shape modeling scheme which allows for the representation of global shap es with local detail and permits model shaping as well as topological changes via physics-based control.The proposed modeling scheme consists of representing shapes by pedal curves and surfaces ¿ pedal curves/surfaces are the loci of the foot of perpendiculars to the tangents of a fixed curve/surface from a fixed point called the pedal point. By varying the location of the pedal point, one can synthesize a large class of shapes which exhibit both local and glob al deformations. We introduce physics-based control for shaping these geometric models by letting the pedal point vary and use a dynamic spline to represent the position of this varying pedal point. The model dubbed as a ""snake pedal"" allows for interactive manipulation via forces applied to the snake. We demonstrate the applicability of this modeling scheme via examples of shape synthesis and shape estimation from real image data."
ICCV	Finding the Epipole from Uncalibrated Optical Flow.	Alessandro Verri,Emanuele Trucco	1998	This paper presents a novel method for determining the location of the instantaneous epipole in a sequence of images acquired by an uncalibrated camera and containing a single, rigid motion (e.g., the camera moves in a static environment). The method uses the full perspective camera model and requires the estimation of the optical flow at a minimum of six image locations. The key observation is that the optical flow equations can be written in terms of the epipole in a strikingly simple form if the translation and rotational flow components are not separated as done usually. The epipole location can then be obtained as the minimum of a least-square residual function asdsociated to the computed optical flow. We report and discuss initial experiments on both synthetic and real data and illustrate possible developments of this method towards the use of uncalibrated optical flow for 3-D motion and structure reconstruction.
ICCV	ASL Recognition Based on a Coupling Between HMMs and 3D Motion Analysis.	Christian Vogler,Dimitris N. Metaxas	1998	We present a framework for recognizing isolated and continuous American Sign Language (ASL) sentences from three-dimensional data. The data are obtained by using physics-based three-dimensional tracking methods and then presented as input to Hidden Markov Models (HMMs) for recognition. To improve recognition performance, we model context-dependent HMMs and present a novel method of coupling three-dimensional computer vision methods and HMMs by temporally segmenting the data stream with vision methods. We then use the geometric properties of the segments to constrain the HMM framework for recognition. We show in experiments with a 53 sign vocabulary that three-dimensional features outperform two-dimensional features in recognition performance. Furthermore, we demonstrate that context-dependent modeling and the coupling of vision methods and HMMs improve the accuracy of continuous ASL recognition.
ICCV	3D Shape and Motion Analysis from Image Blur and Smear: A Unified Approach.	Yuan-Fang Wang,Ping Liang	1998	"This paper addresses 3D shape recovery and motion estimation using a realistic camera model with an aperture and a shutter. The spatial blur and temporal smear effects induced by the camera's finite aperture and shutter speed are used for inferring both the shape and motion of the imaged objects"
ICCV	Intensity and Feature Based Stereo Matching by Disparity Parametrization.	Guo-Qing Wei,Gerd Hirzinger	1998	In this paper, we propose a new solution to the stereo correspondence problem by including features in intensity based matching. The features we use are intensity gradients in both the x and y directions of the left and the deformed right images. Although a uniform smoothness constraint is still used, it is never-the-less applied only to non-feature regions. To avoid local minima in function minimization, we propose to parameterize the disparity function by hierarchical Gaussians. A simple stochastic gradient method is used to estimate the Gaussian weights. Experiments with various real stereo images show robust performances.
ICCV	Consensus Surfaces for Modeling 3D Objects from Multiple Range Images.	Mark D. Wheeler,Yoichi Sato,Katsushi Ikeuchi	1998	In this paper, we present a robust method for creating a triangulated surface mesh from multiple range images. Our method merges a set of range images into a volumeteric implicit-surface representation which is converted to a surface mesh using a variant of the marching-cubes algorithm. Unlike previous techniques based on implicit-surface representations, our method estimates the signed distance to the object surface by finding a consensus of locally coherent observations of the surface. We call this method the consensus-surface algorithm. This algorithm effectively eliminates many of the troublesome effects of noise and extraneous surface observations without sacrificing the accuracy of the resulting surface. We utilize octrees to represent volumetric implicit surfaces ¿ effectively reducing the computation and memory requirements of the volumetric representation without sacrificing accuracy of the resulting surface. We present results which demonstrate that our consensus-surface algorithm can construct accurate geometric models from rather noisy input range date.
ICCV	Recognition and Interpretation of Parametric Gesture.	Andrew D. Wilson,Aaron F. Bobick	1998	A new method for the representation, recognition, and interpretation of parameterized gesture is presented. By parameterized gesture we mean gestures that exhibit a meaniful variation; one example is a point gesture where the important parameter is the 2-dimnesional direction. Our approach is to extend the standard hidden Markov model method of gesture recognition by including a global parametric variation in the output probabilities of the states of the HMM. Using a linear model to derive the theory, we formulate an expectation-maximization (EM) method for training the parametric HMM. During testing, the parametric HMM simultaneously recognizes the gesture and estimates the quantifying parameters. Using visually-derived and directly measured 3-dimensional hand position measurements as input, we present results on two different movements ¿ a size gesture and a point gesture ¿ and show robustness with respect to noise in the input features.
ICCV	Optical Flow Estimation Using Wavelet Motion Model.	Yu-Te Wu,Takeo Kanade,Jeffrey F. Cohn,Ching-Chung Li	1998	A motion estimation algorithm using wavelet approximation as an optical flow model has been developed to estimate accurate dense optical flow from an image sequence. This wavelet motion model is particularly useful in estimating optical flows with large displacement. Traditional pyramid methods which use the coarse-to-fine image pyramid by image burring in estimating optical flow often produce incorrect results when the coarse-level estimates contain large errors that cannot be corrected at the subsequent finer levels. This happens when regions of low texture become flat or certain patterns result in spatial aliasing due to imageblurring. Our method, in contrast, uses large-to-small full-resolution regions without blurring images, and simultaneously optimizes the coarser and finer parts of optical flow so that the large and small motion can be estimated correctly. We compare results obtained by using our method with those obtained by using one of the leading optical flow methods, the Szeliski pyramid spline-based method. The experiments include cases of small displacement (less than4 pixels under 128 × 128 image size or equivalent displacement under other image sizes), and those of large displacement (10 pixels). While both methods produce comparableresults when the displacements are small, our method out performs pyramid spline-based method when the displacements are large.
ICCV	Parameterized Modeling and Recognition of Activities.	Yaser Yacoob,Michael J. Black	1998	A framework for modeling and recognition of temporal activities is proposed. The modeling of sets of exemplar activities is achieved by parameterizing their representation in the form of principal components. Recognition of spatio-temporal variants of modeled activities is achieved by parameterizing the search in the space of admissible transformations that the activities can undergo. Experiments on recognition of articulated and deformable object motion from image motion parameters are presented.
ICCV	Learned Temporal Models of Image Motion.	Yaser Yacoob,Larry S. Davis	1998	An approach for learning and estimating temporal-flow models from image sequences is proposed. The temporal-flow models are represented as a set of orthogonal temporal-flow bases that are learned using principal component analysis of instantaneous flow measurements. Spatial constraints on the temporal-flow are also developed for modeling the motion of regions in rigid and coordinated motion. The performance of these models is demonstrated on several long image sequences of rigid and articulated bodies in motion.
ICCV	GRADE: Gibbs Reaction and Diffusion Equation.	Song Chun Zhu,David Mumford	1998	Recently there have been increasing interest in using nonlinear PDEs for applications in computer vision and image processing, in this paper, we propose a general statistical framework for designing a new class of PDEs. For a given applications, a Markov random field model p(I) is learned according to the minimax entropy principle studied in [25] [26], so that p(I) should characterize the ensemble of images in our application. p(I) is a Gibbs distribution whose energy terms can be divided into two categories. Subsequently the partial differential equations given by gradient descent on the Gibbs potential are essentially reaction-diffusion equations, where the energy terms in one category produce anisotropic diffusion while the inverted energy terms in the second category produce reaction associated with pattern formation. We call this new class of PDEs the Gibbs Reaction And Diffusion Equations ¿ GRADE and we demonstrate experiments where GRADE are used for texture pattern formation, denoising, image enhancements, and clutter removal.
ICCV	Passive Depth from Defocus Using a Spatial Domain Approach.	D. Ziou	1998	This paper presents an algorithm for a dense computation of the difference in blur between two images. The two images are acquired by varying the intrinsic parameters of the camera.The image formation system is assumed to be passive. Estimation of depth from the blur difference is straigh-forward. The algorithm is based on a local image decomposition technique using the Hermite polynomial basis. We show that any coefficient of the Hermite polynomial computed using the more blurred image is a function of the partial derivatives of the other image and the blur difference. Hence, the blur difference can be computed by resolving a system of equations. All computations required are local and carried out in the spatial domain.An algorithm is presented for estimation of the blur in 1D and 2D cases and its behavior is studied for constant images, step edges, line edges and junctions. The algorithm is tested using synthetic and real images. The results obtained are very encouraging. The ACM Portal is published by the Association for Computing Machinery. Copyright © 2010 ACM, Inc. Terms of Usage Privacy Policy Code of Ethics Contact Us Useful downloads: Adobe Acrobat QuickTime Windows Media Player Real Player
ICCV	Schwarz Representation for Matching and Similarity Analysis.	Qing Yang,Songde Ma	1998	"This paper presents a novel multiscale representation of one-dimensional signal based on complex analysis. We show that a signal and its derivative at different scales can be represented by one analytic function defined on the unit disc in the complex plane, which is called the Schwarz representation of a signal. This representation is applied to the matching problem. Using the theory of analytic functions, we are able to define the inverse of a signal. The matching function between two signals can be defined as the composition of one signal's Schwarz representation and another signal's inverse. The matching function determined bythis method has a group structure and is close-formed."
ICCV	Automatic Generation of Robot Program Code: Learning from Perceptual Data.	Mohammed Yeasin,Subhasis Chaudhuri	1998	We propose a novel appr oach to programa robot by demonstrating the task multiple number of times in front of a vision system. Here we integrate human dexterity with sensory data using computer vision techniques in a single platform. A simultaneous feature detection and tracking framework is used to track various features (finger tips and the wrist joint). A Kalman filter does the tracking by predicting the tentative feature location and a HOS-based data clustering algorithm extracts the feature. Color information of the features are used for establishing correspondences. A fast, efficient and robust algorithm for the vision system thus developed process a binocular video sequence to obtain the trajectories and the orientation information of the end effector. The concept of a trajectory bundle is introduced to avoid singularities and to obtain an optimal path.
ICCV	Utilization of Stereo Disparity and Optical Flow Information for Human Interaction.	Ikushi Yoda,Katsuhiko Sakaue	1998	To attain smooth human interaction, we propose a system which simultaneously utilizes the stereo disparity and optical flow information of real-time stereo gray multiresolution images to recognize objects and gestures. For real-time calculation of the disparity and optical flow information of a stereo image, the system first creates pyramid images by utilizing a Gaussian filter. The system then determines the disparity and optical flow of a low density image and extracts regions in front of a certain depth. The three foremost regions are recognized by higher order local auto-correlation features and a linear discriminant analysis. With this process, the system recognizes the face and hand signs of users, which are displayed foremost, and roughly recognizes novements within the region in real-time. With this framework, the system can discriminate the face of a user, can monitor the basic movements of the user, can smoothly learn a presented object by users, and can communicate with users from hand signs learned in advance.
ICCV	Visual Motion Estimation and Prediction: A Probabilistic Network Model for Temporal Coherence.	Alan L. Yuille,Pierre-Yves Burgi,Norberto M. Grzywacz	1998	We develop a theory for the temporal integration of visual motion motivated by psychophysical experiments. The theory proposes that input data are temporally grouped and used to predict and estimate motion flows in the image sequences. Our theory is expressed in terms of the bayesian generalization [10] of standard Kalman Filtering which allows us to solve temporal grouping in conjunction with prediction and estimation. As demonstrated for tracking isolated contours [12], the Bayesian formulation is superior to approaches which use data association as a first stage followed by conventional Kalman filtering. Our computer simulations demonstrate that our theory qualitatively accounts for several psychophysical experiments on motion occlusion and motion outliers [15], [20].
ICCV	Signfinder: Using Color to Detect, Localize and Identify Informational Signs.	Alan L. Yuille,Daniel Snow,Mark Nitzberg	1998	We describe an approach to detecting, locating and normalizing road signs. The approach will apply provided: (i) the signs have stereotypical boundary shapes (i.e. rectangular, or hexagonal ¿ of course, we allow for these shapes to be distorted by projection to unknown viewpoint), (ii) the writingon the sign has one uniform color and the rest of the sign has a second uniform color (we allow for the color of the illuminant to be unknown). We show that the approach works even under significant illuminant color changes, viewpoint direction, shadowing, and occlusion. This work is part of a project intended to help people who are blind, or whose sight is impaired.
ICCV	Understanding the Relationship Between the Optimization Criteria in Two-View Motion Analysis.	Zhengyou Zhang	1998	The three best known criteria in two-view motion analysis are based, respectively, on the distances between points and their corresponding epipolar lines, on the gradient-weighted epipolar errors, and on the distances between points and the reprojections of their reconstructed points. The last one has a better statistical interpretation, but is, however, much slower than the first two. In this paper, we show that the last two criteria are equivalent when the epipoles are at infinity, and differ from each other only a little even when the epipoles are in the image. The first two criteria are equivalent only when the epipoles are at infinity and when the observed object has the same scale in the two images. This suggests that the second criterion is sufficient in practice because of its computational efficiency. The resultis valid for both calibrated and uncalibrated images.
ICCV	Modeling Geometric Structure and Illumination Variation of a Scene from Real Images.	Zhengyou Zhang	1998	We present in this paper a system which automatically builds, from real images, a scene model containing both 3D geometric information of the scene structure and its photometric information under various illumination conditions. The geometric structure is recovered from images taken from distinct viewpoints. Structure-from-motion and correlation-based stereo techniques are used to match pixels between images of different viewpoints and to reconstruct the scene in 3D space. The photometric property is extracted from images taken under different illumination conditions (orientation, position and intensity of the light sources). This is achieved by computing a low-dimensional linear space of the spatio-illumination volume, and is represented by a set of basis images. The model that has been built can be used to create realistic renderings from different viewpoints and illumination conditions. Applications include object recognition, virtual reality and product advertisement.
ICCV	Euclidean Structure from Uncalibrated Images Using Fuzzy Domain Knowledge: Application to Facial Images Synthesis.	Zhengyou Zhang,Katsunori Isono,Shigeru Akamatsu	1998	Use of uncalibrated images has found many applications such as image synthesis. However, it is not easy to specify the desired position of the new image in projective or af.ne space. This paper proposes to recover Euclidean structure from uncalibrated images using domain knowledge such as distances and angles. The knowledge we have is usually about an object category, but not very precise for the particular object being considered. The variation (fuzziness) is modeled as a Gaussian variable. Six types of common knowledge are formulated. Once we have a Euclidean description, the task to specify the desired position in Euclidean space becomes trivial. The proposed technique is then applied to synthesis of new facial images. A number of dif.culties existing in image synthesis are identified and solved. For example, we propose to use edge points to deal with occlusion.
ICCV	Acquiring 3D Object Models from Specular Motion Using Circular Lights Illumination.	Jiang Yu Zheng,Akio Murata	1998	This work recovers 3D graphics models of objects with specular surfaces. An object is routed and continuous images of it are taken. Circular lights that generate cones of rays are used to illuminate the rotating object. When the lights are properly set, each point on the object can be highlighted during the rotation. The shape for each rotational plane is measured independently using its corresponding epipolar palne image. A 3D graphics model is subsequently reconstructed by combining shapes at different rotational planes. Computing a shape is simple and requires only the motion of the highlight on each rotation plane. Results not obtained before are given in the 3D shape recovery experiments on real objects.
ICCV	Object Tracking Using Deformable Templates.	Yu Zhong,Anil K. Jain,Marie-Pierre Dubuisson-Jolly	1998	We propose a novel method for object tracking using prototype-based deformable template models. To track an object in an image sequence, we use a criterion which combines two terms: the deviation of the object shape from its shape in the previous frame, and the fidelity ofthe detected shape to the input image. Shape and gradient information are used to track the object. We have also used the consistency between corresponding object regions throughout the sequence to help in tracking the object of interest. Inter-frame motion is also used to track the boundary of moving objects. We have applied the algorithm to a number of image sequences from different sources. The inherent structure in the deformable template, together with region, motion, and image gradient cues, make the algorithm relatively insensitive to the adverse effects of weak image features and moderate partial occlusion.
IEEE Visualization	Seabed visualization.	Paul Chapman,Peter Stevens,Derek Wills,Graham Brookes	1998	Seabed visualization.
IEEE Visualization	Level of detail visualization of scalar data sets on irregular surface meshes.	Georges-Pierre Bonneau,Alexandre Gerussi	1998	Level of detail visualization of scalar data sets on irregular surface meshes.
IEEE Visualization	Interactive deformations from tensor fields.	Ed Boring,Alex Pang	1998	Interactive deformations from tensor fields.
IEEE Visualization	Visualization of scalar topology for structural enhancement.	Chandrajit L. Bajaj,Valerio Pascucci,Daniel Schikore	1998	Visualization of scalar topology for structural enhancement.
IEEE Visualization	Contour interpolation and surface reconstruction of smooth terrain models.	Jianyun Chai,Takaharu Miyoshi,Eihachiro Nakamae	1998	Contour interpolation and surface reconstruction of smooth terrain models.
IEEE Visualization	Interactive out-of-core isosurface extraction.	Yi-Jen Chiang,Cláudio T. Silva,William J. Schroeder	1998	Interactive out-of-core isosurface extraction.
IEEE Visualization	A general method for preserving attribute values on simplified meshes.	Paolo Cignoni,Claudio Montani,Roberto Scopigno,Claudio Rocchini	1998	A general method for preserving attribute values on simplified meshes.
IEEE Visualization	Rear-projecting virtual data onto physical terrain: an exercise in two senses being better than one.	Dru Clark,Richard Marciano,Rosemarie McKeon,Michael J. Bailey	1998	Rear-projecting virtual data onto physical terrain: an exercise in two senses being better than one.
IEEE Visualization	Interactive display of very large textures.	David Cline,Parris K. Egbert	1998	Interactive display of very large textures.
IEEE Visualization	Image-based transfer function design for data exploration in volume visualization.	Shiaofen Fang,Tom Biddlecome,Mihran Tuceryan	1998	Image-based transfer function design for data exploration in volume visualization.
IEEE Visualization	Intent, perception, and out-of-core visualization applied to terrain.	Douglass Davis,William Ribarsky,Nickolas Faust,T. Y. Jiang	1998	Intent, perception, and out-of-core visualization applied to terrain.
IEEE Visualization	Efficient implementation of multi-triangulations.	Leila De Floriani,Paola Magillo,Enrico Puppo	1998	Efficient implementation of multi-triangulations.
IEEE Visualization	Interpolation of triangle hierarchies.	Axel Friedrich,Konrad Polthier,Markus Schmies	1998	Interpolation of triangle hierarchies.
IEEE Visualization	Battlefield visualization on the responsive workbench.	Jim Durbin,J. Edward Swan II,Brad Colbert,John Crowe,Rob King,Tony King,Christopher Scannell,Zachary Wartell,Terry Welsh	1998	Battlefield visualization on the responsive workbench.
IEEE Visualization	Real-time techniques for 3D flow visualization.	Anton L. Fuhrmann,Eduard Gröller	1998	Real-time techniques for 3D flow visualization.
IEEE Visualization	Simplifying surfaces with color and texture using quadric error metrics.	Michael Garland,Paul S. Heckbert	1998	Simplifying surfaces with color and texture using quadric error metrics.
IEEE Visualization	Interactive virtual angioscopy.	Enrico Gobbetti,Piero Pili,Antonio Zorcolo,Massimiliano Tuveri	1998	Interactive virtual angioscopy.
IEEE Visualization	A unified approach for simplifying polygonal and spline models.	M. Gopi,Dinesh Manocha	1998	A unified approach for simplifying polygonal and spline models.
IEEE Visualization	Key problems and thorny issues in multidimensional visualization.	Georges G. Grinstein,Sharon J. Laskowski,Alfred Inselberg	1998	Key problems and thorny issues in multidimensional visualization.
IEEE Visualization	Converting sets of polygons to manifold surfaces by cutting and stitching.	André Guéziec,Gabriel Taubin,Francis Lazarus,William Horn	1998	Converting sets of polygons to manifold surfaces by cutting and stitching.
IEEE Visualization	"Modern Trompe l'oeil - Keynote."	Pat Hanrahan	1998	"Modern Trompe l'oeil - Keynote."
IEEE Visualization	Constrained optimal framings of curves and surfaces using quaternion Gauss maps.	Andrew J. Hanson	1998	Constrained optimal framings of curves and surfaces using quaternion Gauss maps.
IEEE Visualization	Image-based rendering with occlusions via cubist images.	Andrew J. Hanson,Eric A. Wernert	1998	Image-based rendering with occlusions via cubist images.
IEEE Visualization	Building perceptual textures to visualize multidimensional datasets.	Christopher G. Healey,James T. Enns	1998	Building perceptual textures to visualize multidimensional datasets.
IEEE Visualization	Production visualization for the ASCI one TeraFLOPS machine.	Philip D. Heermann	1998	Production visualization for the ASCI one TeraFLOPS machine.
IEEE Visualization	Feature detection in linked derived spaces.	Chris Henze	1998	Feature detection in linked derived spaces.
IEEE Visualization	Continuous cartogram construction.	Donald H. House,Christopher J. Kocmoud	1998	Continuous cartogram construction.
IEEE Visualization	Visual presentation of magnetic resonance images.	Johanna E. van der Heyden,M. Sheelagh T. Carpendale,Kevin B. Inkpen,M. Stella Atkins	1998	Visual presentation of magnetic resonance images.
IEEE Visualization	Accelerated ray-casting for curvilinear volumes.	Lichan Hong,Arie E. Kaufman	1998	Accelerated ray-casting for curvilinear volumes.
IEEE Visualization	Smooth view-dependent level-of-detail control and its application to terrain rendering.	Hugues Hoppe	1998	Smooth view-dependent level-of-detail control and its application to terrain rendering.
IEEE Visualization	Why is real-time volume rendering no longer a year away?	Arie E. Kaufman,Marty Brady,William E. Lorensen,Frederick L. Kitson,Hanspeter Pfister	1998	Why is real-time volume rendering no longer a year away?
IEEE Visualization	Size preserving pattern mapping.	Yair Kurzion,Torsten Möller,Roni Yagel	1998	Size preserving pattern mapping.
IEEE Visualization	Development of a multi-source visualization prototype.	Leslie Keely,Samuel P. Uselton	1998	Development of a multi-source visualization prototype.
IEEE Visualization	Visualizing diffusion tensor images of the mouse spinal cord.	David H. Laidlaw,Eric T. Ahrens,David Kremers,Matthew J. Avalos,Russell E. Jacobs,Carol Readhead	1998	Visualizing diffusion tensor images of the mouse spinal cord.
IEEE Visualization	The Gridfit algorithm: an efficient and effective approach to visualizing large amounts of spatial data.	Daniel A. Keim,Annemarie Herrmann	1998	The Gridfit algorithm: an efficient and effective approach to visualizing large amounts of spatial data.
IEEE Visualization	Art and visualization: oil and water?	David H. Laidlaw,David Kremers,Felica Frankel,Victoria Interrante,Thomas Banchoff	1998	Art and visualization: oil and water?
IEEE Visualization	Automatic detection of open and closed separation and attachment lines.	David N. Kenwright	1998	Automatic detection of open and closed separation and attachment lines.
IEEE Visualization	Supporting detail-in-context for the DNA representation, H-curves.	M. L. Lantin,M. Sheelagh T. Carpendale	1998	Supporting detail-in-context for the DNA representation, H-curves.
IEEE Visualization	Three-dimensional visualization of microstructures.	Marco Lanzagorta,Milo V. Kral,J. Edward Swan II,George Spanos,Robert Rosenberg,Eddy Kuo	1998	Three-dimensional visualization of microstructures.
IEEE Visualization	"Volumetric modeling of acoustic fields in CNMAT's sound spatialization theatre."	Sami Khoury,Adrian Freed,David Wessel	1998	"Volumetric modeling of acoustic fields in CNMAT's sound spatialization theatre."
IEEE Visualization	"Feature comparisons of vector fields using earth mover's distance."	Yingmei Lavin,Rajesh Batra,Lambertus Hesselink	1998	"Feature comparisons of vector fields using earth mover's distance."
IEEE Visualization	Comparing LIC and spot noise.	Wim C. de Leeuw,Robert van Liere	1998	Spot noise and line integral convolution (LIC) are two texture synthesis techniques for vector field visualization. In this paper the two techniques are compared. Continuous directional convolution is used as a common basis for comparing the techniques. It is shown that the techniques are based on the same mathematical concept. Comparisons of the visual appearance of the output and performance of the algorithms are made.
IEEE Visualization	Visualizing Hilbert curves.	Nelson L. Max	1998	Visualizing Hilbert curves.
IEEE Visualization	A concept for virtual reality tools for design reviews.	Klaus Kremer	1998	A concept for virtual reality tools for design reviews.
IEEE Visualization	Scientific visualization and data modeling of scattered sediment contaminant data in New York/New Jersey estuaries.	Hong Ma,Keith W. Jones,Eric A. Stern	1998	Scientific visualization and data modeling of scattered sediment contaminant data in New York/New Jersey estuaries.
IEEE Visualization	A distributed blackboard architecture for interactive data visualization.	Robert van Liere,Jan Harkes,Wim C. de Leeuw	1998	In this paper the motivation, design and application of a distributed blackboard architecture for interactive data visualization is discussed. The main advantages of the architecture is twofold. First, it allows visualization tools to be tightly integrated with simulations. Second, it allows qualitative and quantitative analysis to be combined during the visualization process.
IEEE Visualization	Fast and memory efficient polygonal simplification.	Peter Lindstrom,Greg Turk	1998	Fast and memory efficient polygonal simplification.
IEEE Visualization	View dependent isosurface extraction.	Yarden Livnat,Charles D. Hansen	1998	View dependent isosurface extraction.
IEEE Visualization	Hierarchical volume analysis and visualization based on morphological operators.	Christoph Lürig,Thomas Ertl	1998	Hierarchical volume analysis and visualization based on morphological operators.
IEEE Visualization	Eliminating popping artifacts in sheet buffer-based splatting.	Klaus Mueller,Roger Crawfis	1998	Eliminating popping artifacts in sheet buffer-based splatting.
IEEE Visualization	Pixel masks for screen-door transparency.	Jurriaan D. Mulder,Frans C. A. Groen,Jarke J. van Wijk	1998	Rendering objects transparently gives additional insight in complex and overlapping structures. However, traditional techniques for the rendering of transparent objects such as alpha blending are not very well suited for the rendering of multiple transparent objects in dynamic scenes. Screen-door transparency is a technique to render transparent objects in a simple and efficient way: No sorting is required and intersecting polygons can be handled without further preprocessing. With this technique, polygons are rendered through a mask: Only where the mask is present, pixels are set. However, artifacts such as incorrect opacities and distracting patterns can easily occur if the masks are not carefully designed. In this paper, first the requirements on the masks are considered. Next, three algorithms are presented for the generation of pixel masks. One algorithm is designed for the creation of small (e.g.\ 4 $\times$ 4) masks. The other two algorithms can be used for the creation of larger masks (e.g.\ 32 $\times$ 32). For each of these algorithms results are presented and discussed.
IEEE Visualization	Image-guided streamline placement on curvilinear grid surfaces.	Xiaoyang Mao,Yuji Hatanaka,Hidenori Higashida,Atsumi Imamiya	1998	Image-guided streamline placement on curvilinear grid surfaces.
IEEE Visualization	A case study using the virtual environment for reconstructive surgery.	Kevin Montgomery,Michael Stephanides,Stephen Schendel,Muriel Ross	1998	A case study using the virtual environment for reconstructive surgery.
IEEE Visualization	POPTEX: interactive ocean model visualization using texture mapping hardware.	Allen McPherson,Mathew Maltrud	1998	POPTEX: interactive ocean model visualization using texture mapping hardware.
IEEE Visualization	Wavelets over curvilinear grids.	Gregory M. Nielson,Il-Hong Jung,Junwon Sung	1998	Wavelets over curvilinear grids.
IEEE Visualization	TOPIC ISLANDS - a wavelet-based text visualization system.	Nancy E. Miller,Pak Chung Wong,Mary Brewster,Harlan Foote	1998	TOPIC ISLANDS - a wavelet-based text visualization system.
IEEE Visualization	Large scale terrain visualization using the restricted quadtree triangulation.	Renato Pajarola	1998	Large scale terrain visualization using the restricted quadtree triangulation.
IEEE Visualization	Interactive ray tracing for isosurface rendering.	Steven G. Parker,Peter Shirley,Yarden Livnat,Charles D. Hansen,Peter-Pike J. Sloan	1998	Interactive ray tracing for isosurface rendering.
IEEE Visualization	Efficient warping for architectural walkthroughs using layered depth images.	Voicu Popescu,Anselmo Lastra,Daniel G. Aliaga,Manuel Menezes de Oliveira Neto	1998	Efficient warping for architectural walkthroughs using layered depth images.
IEEE Visualization	Visualizing differences in movies of cortical activity.	Kay A. Robbins,David M. Senseman	1998	Visualizing differences in movies of cortical activity.
IEEE Visualization	Acoustic imaging and visualization of plumes discharging from black smoker vents on the deep seafloor.	Peter A. Rona,Karen G. Bemis,Deepak R. Kenchammana-Hosekote,Deborah Silver	1998	Acoustic imaging and visualization of plumes discharging from black smoker vents on the deep seafloor.
IEEE Visualization	A higher-order method for finding vortex core lines.	Martin Roth,Ronald Peikert	1998	A higher-order method for finding vortex core lines.
IEEE Visualization	Configuration space visualization for mechanical design.	Elisha Sacks,Leo Joskowicz	1998	Configuration space visualization for mechanical design.
IEEE Visualization	Selective visualization of vortices in hydrodynamic flows.	I. Ari Sadarjoen,Frits H. Post,Bing Ma,David C. Banks,Hans-Georg Pagendarm	1998	Selective visualization of vortices in hydrodynamic flows.
IEEE Visualization	Visualization for multiparameter aircraft designs.	Clifford A. Shaffer,Duane L. Knill,Layne T. Watson	1998	Visualization for multiparameter aircraft designs.
IEEE Visualization	Isosurface extraction in time-varying fields using a temporal hierarchical index tree.	Han-Wei Shen	1998	Isosurface extraction in time-varying fields using a temporal hierarchical index tree.
IEEE Visualization	Data level comparison of wind tunnel and computational fluid dynamics data.	Qin Shen,Alex Pang,Samuel P. Uselton	1998	Data level comparison of wind tunnel and computational fluid dynamics data.
IEEE Visualization	Tracking scalar features in unstructured datasets.	Deborah Silver,Xin Wang	1998	Tracking scalar features in unstructured datasets.
IEEE Visualization	Progressive tetrahedralizations.	Oliver G. Staadt,Markus H. Gross	1998	Progressive tetrahedralizations.
IEEE Visualization	Extremal feature extraction from 3-D vector and noisy scalar fields.	Chi-Keung Tang,Gérard G. Medioni	1998	Extremal feature extraction from 3-D vector and noisy scalar fields.
IEEE Visualization	Surface reconstruction with anisotropic density-scaled alpha shapes.	Marek Teichmann,Michael V. Capps	1998	Surface reconstruction with anisotropic density-scaled alpha shapes.
IEEE Visualization	High quality rendering of attributed volume data.	Ulf Tiede,Thomas Schiemann,Karl Heinz Höhne	1998	High quality rendering of attributed volume data.
IEEE Visualization	Task-specific visualization design: a case study in operational weather forecasting.	Lloyd Treinish	1998	Task-specific visualization design: a case study in operational weather forecasting.
IEEE Visualization	Simplification of tetrahedral meshes.	Issac J. Trotts,Bernd Hamann,Kenneth I. Joy,David F. Wiley	1998	Simplification of tetrahedral meshes.
IEEE Visualization	Multi-source data analysis challenges.	Samuel P. Uselton,Lloyd Treinish,James P. Ahrens,E. Wes Bethel,Andrei State	1998	Multi-source data analysis challenges.
IEEE Visualization	Visualization in corneal topography.	Frans Vos,Hans J. W. Spoelder	1998	Visualization in corneal topography.
IEEE Visualization	Efficient co-triangulation of large data sets.	Henrik Weimer,Joe D. Warren,Jane Troutner,Wendell Wiggins,John Shrout	1998	Efficient co-triangulation of large data sets.
IEEE Visualization	Draw on the wall - Capstone Address.	Turner Whitted	1998	Draw on the wall - Capstone Address.
IEEE Visualization	"Visualization '98, Proceedings, October 18-23, 1998, Research Triangle Park, North Carolina, USA."	David S. Ebert,Holly E. Rushmeier,Hans Hagen	1998	"Visualization '98, Proceedings, October 18-23, 1998, Research Triangle Park, North Carolina, USA."
IEEE Trans. Vis. Comput. Graph.	Novel View Synthesis by Cascading Trilinear Tensors.	Shai Avidan,Amnon Shashua	1998	We present a new method for synthesizing novel views of a 3D scene from two or three reference images in full correspondence. The core of this work is the use and manipulation of an algebraic entity termed the trilinear tensor that links point correspondences across three images. For a given virtual camera position and orientation, a new trilinear tensor can be computed based on the original tensor of the reference images. The desired view can then be created using this new trilinear tensor and point correspondences across two of the reference images.
IEEE Trans. Vis. Comput. Graph.	RSVP: A Geometric Toolkit for Controlled Repair of Solid Models.	Gill Barequet,Christian A. Duncan,Subodh Kumar	1998	"This paper presents a system and the associated algorithms for repairing the boundary representation of CAD models. Two types of errors are considered: topological errors, i.e., aggregate errors, like zero-volume parts, duplicate or missing parts, inconsistent surface orientation, etc., and geometric errors, i.e., numerical imprecision errors, like cracks or overlaps of geometry. The output of our system describes a set of clean and consistent two-manifolds (possibly with boundaries) with derived adjacencies. Such solid representation enables the application of a variety of rendering and analysis algorithms, e.g., finite-element analysis, radiosity computation, model simplification, and solid free-form fabrication. The algorithms described here were originally designed to correct errors in polygonal B-Reps. We also present an extension for spline surfaces.Central to our system is a procedure for inferring local adjacencies of edges. The geometric representation of topologically-adjacent edges are merged to evolve a set of two-manifolds. Aggregate errors are discovered during the merging step. Unfortunately, there are many ambiguous situations where errors admit more than one valid solution. Our system proposes an object-repairing process based on a set of user-tunable heuristics. The system also allows the user to override the algorithm's decisions in a repair-visualization step. In essence, this visualization step presents an organized and intuitive way for the user to explore the space of valid solutions and to select the correct one."
IEEE Trans. Vis. Comput. Graph.	Topology Simplification for Polygonal Virtual Environments.	Jihad El-Sana,Amitabh Varshney	1998	We present a topology simplifying approach that can be used for genus reductions, removal of protuberances, and repair of cracks in polygonal models in a unified framework. Our work is complementary to the existing work on geometry simplification of polygonal datasets and we demonstrate that using topology and geometry simplifications together yields superior multiresolution hierarchies than is possible by using either of them alone. Our approach can also address the important issue of repair of cracks in polygonal models, as well as for rapid identification and removal of protuberances based on internal accessibility in polygonal models. Our approach is based on identifying holes and cracks by extending the concept of ¿-shapes to polygonal meshes under the L¿ distance metric. We then generate valid triangulations to fill them using the intuitive notion of sweeping an L¿ cube over the identified regions.
IEEE Trans. Vis. Comput. Graph.	Line Art Illustrations of Parametric and Implicit Forms.	Gershon Elber	1998	A technique is presented for line art rendering of scenes composed of freeform surfaces. The line art that is created for parametric surfaces is practically intrinsic and is globally invariant to changes in the surface parameterization. This method is equally applicable for line art rendering of implicit forms, creating a unified line art rendering method for both parametric and implicit forms. This added flexibility exposes a new horizon of special, parameterization independent, line art effects. Moreover, the production of the line art illustrations can be combined with traditional rendering techniques such as transparency and texture mapping. Examples that demonstrate the capabilities of the proposed approach are presented for both the parametric and implicit forms.
IEEE Trans. Vis. Comput. Graph.	MIP-Map Level Selection for Texture Mapping.	Jon P. Ewins,Marcus D. Waller,Martin White,Paul F. Lister	1998	Texture mapping is a fundamental feature of computer graphics image generation. In current PC-based acceleration hardware, MIP-mapping with bilinear and trilinear filtering is a commonly used filtering technique for reducing spatial aliasing artifacts. The effectiveness of this technique in reducing image aliasing at the expense of blurring is dependent upon the MIP-map level selection and the associated calculation of screen-space to texture-space pixel scaling. This paper describes an investigation of practical methods for per-pixel and per-primitive level of detail calculation. This investigation was carried out as part of the design work for a screen-space rasterization ASIC. The implementations of several algorithms of comparable visual quality are discussed and a comparison is provided in terms of per-primitive and per-pixel computational costs.
IEEE Trans. Vis. Comput. Graph.	Multiresolution Analysis on Irregular Surface Meshes.	Georges-Pierre Bonneau	1998	"Wavelet-based methods have proven their efficiency for the visualization at different levels of detail, progressive transmission, and compression of large data sets. The required core of all wavelet-based methods is a hierarchy of meshes that satisfies subdivision-connectivity: This hierarchy has to be the result of a subdivision process starting from a base mesh. Examples include quadtree uniform 2D meshes, octree uniform 3D meshes, or 4-to-1 split triangular meshes. In particular, the necessity of subdivision-connectivity prevents the application of wavelet-based methods on irregular triangular meshes. In this paper, a ""wavelet-like"" decomposition is introduced that works on piecewise constant data sets over irregular triangular surface meshes. The decomposition/reconstruction algorithms are based on an extension of wavelet-theory allowing hierarchical meshes without subdivision-connectivity property. Among others, this approach has the following features: It allows exact reconstruction of the data set, even for nonregular triangulations, and it extends previous results on Haar-wavelets over 4-to-1 split triangulations."
IEEE Trans. Vis. Comput. Graph.	Interactive Volume Navigation.	Martin L. Brady,Kenneth K. Jung,H. T. Nguyen,Thinh P. Q. Nguyen	1998	"Volume navigation is the interactive exploration of volume data sets by ""flying"" the viewpoint through the data, producing a volume rendered view at each frame. We present an inexpensive perspective volume navigation method designed to be run on a PC platform with accelerated 3D graphics hardware. The heart of the method is a two-phase perspective raycasting algorithm that takes advantage of the coherence inherent in adjacent frames during navigation. The algorithm generates a sequence of approximate volume-rendered views in a fraction of the time that would be required to compute them individually. The algorithm handles arbitrarily large volumes by dynamically swapping data within the current view frustum into main memory as the viewpoint moves through the volume. We also describe an interactive volume navigation application based on this algorithm. The application renders gray-scale, RGB, and labeled RGB volumes by volumetric compositing, allows trilinear interpolation of sample points, and implements progressive refinement during pauses in user input."
IEEE Trans. Vis. Comput. Graph.	Constructing Hierarchies for Triangle Meshes.	Tran S. Gieng,Bernd Hamann,Kenneth I. Joy,Gregory L. Schussman,Issac J. Trotts	1998	We present a method to produce a hierarchy of triangle meshes that can be used to blend different levels of detail in a smooth fashion. The algorithm produces a sequence of meshes ${\cal M}_0, {\cal M}_1, {\cal M}_2, ..., {\cal M}_n,$ where each mesh ${\cal M}_i$ can be transformed to mesh ${\cal M}_{i+1}$ through a set of triangle-collapse operations. For each triangle, a function is generated that approximates the underlying surface in the area of the triangle, and this function serves as a basis for assigning a weight to the triangle in the ordering operation and for supplying the points to which the triangles are collapsed. The algorithm produces a limited number of intermediate meshes by selecting, at each step, a number of triangles that can be collapsed simultaneously. This technique allows us to view a triangulated surface model at varying levels of detail while insuring that the simplified mesh approximates the original surface well.
IEEE Trans. Vis. Comput. Graph.	Special Issue on Visualization.	Hans Hagen	1998	Special Issue on Visualization.
IEEE Trans. Vis. Comput. Graph.	Calibration-Free Augmented Reality.	Kiriakos N. Kutulakos,James R. Vallino	1998	"Camera calibration and the acquisition of Euclidean 3D measurements have so far been considered necessary requirements for overlaying three-dimensional graphical objects with live video. In this article, we describe a new approach to video-based augmented reality that avoids both requirements: It does not use any metric information about the calibration parameters of the camera or the 3D locations and dimensions of the environment's objects. The only requirement is the ability to track across frames at least four fiducial points that are specified by the user during system initialization and whose world coordinates are unknown.Our approach is based on the following observation: Given a set of four or more noncoplanar 3D points, the projection of all points in the set can be computed as a linear combination of the projections of just four of the points. We exploit this observation by 1) tracking regions and color fiducial points at frame rate, and 2) representing virtual objects in a non-Euclidean, affine frame of reference that allows their projection to be computed as a linear combination of the projection of the fiducial points. Experimental results on two augmented reality systems, one monitor-based and one head-mounted, demonstrate that the approach is readily implementable, imposes minimal computational and hardware requirements, and generates real-time and accurate video overlays even when the camera parameters vary dynamically."
IEEE Trans. Vis. Comput. Graph.	Memory-Efficient Ray Classification for Visibility Operations.	Bomjun Kwon,Dae Seoung Kim,Kyung-Yong Chwa,Sung Yong Shin	1998	In this paper, we present a new ray classification scheme that considerably reduces memory consumption while preserving its inherent time efficiency. Our key idea is due to the fact that the rays lying on the same line are duplicated over many cells in the ray classification scheme [1]. We are thus able to lower the dimensions of the ray space by classifying lines instead of rays. Our scheme produces much simpler-shaped, compact ray cells that eventually accelerate ray shooting operations.
IEEE Trans. Vis. Comput. Graph.	Perception of Human Motion With Different Geometric Models.	"Jessica K. Hodgins,James F. O'Brien,Jack Tumblin"	1998	"Human figures have been animated using a variety of geometric models, including stick figures, polygonal models, and NURBS-based models with muscles, flexible skin, or clothing. This paper reports on experimental results indicating that a viewer's perception of motion characteristics is affected by the geometric model used for rendering. Subjects were shown a series of paired motion sequences and asked if the two motions in each pair were the same or different The motion sequences in each pair were rendered using the same geometric model. For the three types of motion variation tested, sensitivity scores indicate that subjects were better able to observe changes with the polygonal model than they were with the stick figure model."
IEEE Trans. Vis. Comput. Graph.	The Information Mural: A Technique for Displaying and Navigating Large Information Spaces.	Dean F. Jerding,John T. Stasko	1998	The Information Mural: A Technique for Displaying and Navigating Large Information Spaces.
IEEE Trans. Vis. Comput. Graph.	Editorial.	Arie E. Kaufman	1998	Editorial.
IEEE Trans. Vis. Comput. Graph.	Structure-Significant Representation of Structured Datasets.	Raghu Machiraju,Zhifan Zhu,Bryan Fry,Robert J. Moorhead	1998	Numerical simulation of physical phenomena is now an accepted way of scientific inquiry. However, the field is still evolving, with a profusion of new solution and grid-generation techniques being continuously proposed. Concurrent and retrospective visualization are being used to validate the results, compare them among themselves and with experimental data, and browse through large scientific databases. There exists a need for representation schemes which allow access of structures in an increasing order of smoothness (or decreasing order of significance). We describe our methods on datasets obtained from curvilinear grids. Our target application required visualization of a computational simulation performed on a very remote supercomputer. Since no grid adaptation was performed, it was not deemed necessary to simplify or compress the grid. In essence, we treat the solution as if it were in the computational domain. Inherent to the identification of significant structures is determining the location of the scale coherent structures and assigning saliency values to them [22], [23]. Scale coherent structures are obtained as a result of combining the coefficients of a wavelet transform across scales. The result of this operation is a correlation mask that delineates regions containing significant structures. A spatial subdivision (e.g., octree) is used to delineate regions of interest. The mask values in these subdivided regions are used as a measure of information content. Later, another wavelet transform is conducted within each subdivided region and the coefficients are sorted based on a perceptual function with bandpass characteristics. This allows for ranking of structures based on the order of significance, giving rise to an adaptive and embedded representation scheme. We demonstrate our methods on two datasets from computational field simulations. Essentially, we show how our methods allow the ranked access of significant structures. We also compare our adaptive representation scheme with a fixed blocksize scheme.
IEEE Trans. Vis. Comput. Graph.	Fast Collision Detection Among Multiple Moving Spheres.	Dong Jin Kim,Leonidas J. Guibas,Sung Yong Shin	1998	Fast Collision Detection Among Multiple Moving Spheres.
IEEE Trans. Vis. Comput. Graph.	Efficient Collision Detection Using Bounding Volume Hierarchies of k-DOPs.	James T. Klosowski,Martin Held,Joseph S. B. Mitchell,Henry Sowizral,Karel Zikan	1998	"Collision detection is of paramount importance for many applications in computer graphics and visualization. Typically, the input to a collision detection algorithm is a large number of geometric objects comprising an environment, together with a set of objects moving within the environment. In addition to determining accurately the contacts that occur between pairs of objects, one needs also to do so at real-time rates. Applications such as haptic force-feedback can require over 1,000 collision queries per second.In this paper, we develop and analyze a method, based on bounding-volume hierarchies, for efficient collision detection for objects moving within highly complex environments. Our choice of bounding volume is to use a ""discrete orientation polytope"" (""k-dop""), a convex polytope whose facets are determined by halfspaces whose outward normals come from a small fixed set of k orientations. We compare a variety of methods for constructing hierarchies (""BV-trees"") of bounding k-dops. Further, we propose algorithms for maintaining an effective BV-tree of k-dops for moving objects, as they rotate, and for performing fast collision detection using BV-trees of the moving objects and of the environment.Our algorithms have been implemented and tested. We provide experimental evidence showing that our approach yields substantially faster collision detection than previous methods."
IEEE Trans. Vis. Comput. Graph.	Splatting Errors and Antialiasing.	Klaus Mueller,Torsten Möller,J. Edward Swan II,Roger Crawfis,Naeem Shareef,Roni Yagel	1998	This paper describes three new results for volume rendering algorithms utilizing splatting. First, an antialiasing extension to the basic splatting algorithm is introduced that mitigates the spatial aliasing for high-resolution volumes. Aliasing can be severe for high-resolution volumes or volumes where a high depth of field leads to converging samples along the perspective axis. Next, an analysis of the common approximation errors in the splatting process for perspective viewing is presented. In this context, we give different implementations, distinguished by efficiency and accuracy, for adding the splat contributions to the image plane. We then present new results in controlling the splatting errors and also show their behavior in the framework of our new antialiasing technique. Finally, current work in progress on extensions to splatting for temporal antialiasing is demonstrated. Here, we present a simple but highly effective scheme for adding motion blur to fast moving volumes.
IEEE Trans. Vis. Comput. Graph.	Modeling, Animating, and Rendering Complex Scenes Using Volumetric Textures.	Fabrice Neyret	1998	Complex repetitive scenes containing forests, foliage, grass, hair, or fur, are challenging for common modeling and rendering tools. The amount of data, the tediousness of modeling and animation tasks, and the cost of realistic rendering have caused such kind of scene to see only limited use even in high-end productions. We describe here how the use of volumetric textures is well suited to such scenes. These primitives can greatly simplify modeling and animation tasks. More importantly, they can be very efficiently rendered using ray tracing with few aliasing artifacts. The main idea, initially introduced by Kajiya and Kay [9], is to represent a pattern of 3D geometry in a reference volume, that is tiled over an underlying surface much like a regular 2D texture. In our contribution, the mapping is independent of the mesh subdivision, the pattern can contain any kind of shape, and it is prefiltered at different scales as for MIP-mapping. Although the model encoding is volumetric, the rendering method differs greatly from traditional volume rendering: A volumetric texture only exists in the neighborhood of a surface, and the repeated instances (called texels) of the reference volume are spatially deformed. Furthermore, each voxel of the reference volume contains a key feature which controls the reflectance function that represents aggregate intravoxel geometry. This allows for ray-tracing of highly complex scenes with very few aliasing artifacts, using a single ray per pixel (for the part of the scene using the volumetric texture representation). The major technical considerations of our method lie in the ray-path determination and in the specification of the reflectance function.
IEEE Trans. Vis. Comput. Graph.	Adaptive Projection Operators in Multiresolution Scientific Visualization.	Mario Ohlberger,Martin Rumpf	1998	Recently, multiresolution visualization methods have become an indispensable ingredient of real-time interactive postprocessing. The enormous databases, typically coming along with some hierarchical structure, are locally resolved on different levels of detail to achieve a significant savings of CPU and rendering time. Here, the method of adaptive projection and the corresponding operators on data functions, respectively, are introduced. They are defined and discussed as mathematically rigorous foundations for multiresolution data analysis. Keeping in mind data from efficient numerical multigrid methods, this approach applies to hierarchical nested grids consisting of elements which are any tensor product of simplices, generated recursively by an arbitrary, finite set of refinement rules from some coarse grid. The corresponding visualization algorithms, e.g., color shading on slices or isosurface rendering, are confined to an appropriate depth-first traversal of the grid hierarchy. A continuous projection of the data onto an adaptive, extracted subgrid is thereby calculated recursively. The presented concept covers different methods of local error measurement, time-dependent data which have to be interpolated from a sequence of key frames, and a tool for local data focusing. Furthermore, it allows for a continuous level of detail.
IEEE Trans. Vis. Comput. Graph.	Dynamic Catmull-Clark Subdivision Surfaces.	Hong Qin,Chhandomay Mandal,Baba C. Vemuri	1998	"Recursive subdivision schemes have been extensively used in computer graphics, computer-aided geometric design, and scientific visualization for modeling smooth surfaces of arbitrary topology. Recursive subdivision generates a visually pleasing smooth surface in the limit from an initial user-specified polygonal mesh through the repeated application of a fixed set of subdivision rules. In this paper, we present a new dynamic surface model based on the Catmull-Clark subdivision scheme, a popular technique for modeling complicated objects of arbitrary genus. Our new dynamic surface model inherits the attractive properties of the Catmull-Clark subdivision scheme, as well as those of the physics-based models. This new model provides a direct and intuitive means of manipulating geometric shapes, and an efficient hierarchical approach for recovering complex shapes from large range and volume data sets using very few degrees of freedom (control vertices). We provide an analytic formulation and introduce the ""physical"" quantities required to develop the dynamic subdivision surface model which can be interactively deformed by applying synthesized forces. The governing dynamic differential equation is derived using Lagrangian mechanics and the finite element method. Our experiments demonstrate that this new dynamic model has a promising future in computer graphics, geometric shape design, and scientific visualization."
IEEE Trans. Vis. Comput. Graph.	Visualization of Scientific Video Data Using KL Decomposition.	Kay A. Robbins	1998	Fast methods are developed for visualizing and classifying certain types of scientific video data. These techniques, which are based on KL decomposition, find a best coordinate system for a data set. When the data set represents a temporally ordered collection of images, the best coordinate system leads to approximations that are separable in time and space. Practical methods for computing this best coordinate system are discussed and physically significant visualizations for experimental video data are developed. The visualization techniques are applied to two experimental systems¿one from combustion and the other from neurobiology¿to show how relevant information can be quickly extracted from video data. These techniques can be integrated into the video acquisition process to provide real-time feedback to the experimentalist during the operation of an experiment.
IEEE Trans. Vis. Comput. Graph.	Visualizing Nonlinear Vector Field Topology.	Gerik Scheuermann,Heinz Krüger,Martin Menzel,Alyn P. Rockwood	1998	We present our results on the visualization of nonlinear vector field topology. The underlying mathematics is done in Clifford algebra, a system describing geometry by extending the usual vector space by a multiplication of vectors. We started with the observation that all known algorithms for vector field topology are based on piecewise linear or bilinear approximation, and that these methods destroy the local topology if nonlinear behavior is present. Our algorithm looks for such situations, chooses an appropriate polynomial approximation in these areas, and, finally, visualizes the topology. This overcomes the problem, and the algorithm is still very fast because we are using linear approximation outside these small but important areas. The paper contains a detailed description of the algorithm and a basic introduction to Clifford algebra.
IEEE Trans. Vis. Comput. Graph.	Efficient Polygon Clipping for an SIMD Graphics Pipeline.	Bengt-Olaf Schneider,Jim van Welzen	1998	Recently, SIMD processors have become popular architectures for multimedia. Though most of the 3D graphics pipeline can be implemented on such SIMD platforms in a straightforward manner, polygon clipping tends to cause clumsy and expensive interruptions to the SIMD pipeline. This paper describes a way to increase the efficiency of SIMD clipping without sacrificing the efficient flow of a SIMD graphics pipeline. In order to fully utilize the parallel execution units, we have developed two methods to avoid serialization of the execution stream: Deferred clipping postpones polygon clipping and uses hardware assistance to buffer polygons that need to be clipped. SIMD Clipping partitions the actual polygon clipping procedure between the SIMD engine and a conventional RISC processor. To increase the efficiency of SIMD clipping, we introduce the concepts of clip-plane pairs and edge batching. Clip-plane pairs allow clipping a polygon against two clip planes without introducing corner vertices. Edge batching reduces the communication and control overhead for starting of clipping on the SIMD engine.
IEEE Trans. Vis. Comput. Graph.	A New Line Integral Convolution Algorithm for Visualizing Time-Varying Flow Fields.	Han-Wei Shen,David L. Kao	1998	New challenges on vector field visualization emerge as time-dependent numerical simulations become ubiquitous in the field of computational fluid dynamics (CFD). To visualize data generated from these simulations, traditional techniques, such as displaying particle traces, can only reveal flow phenomena in preselected local regions and, thus, are unable to track the evolution of global flow features over time. This paper presents a new algorithm, called UFLIC (Unsteady Flow LIC), to visualize vector data in unsteady flow fields. Our algorithm extends a texture synthesis technique, called Line Integral Convolution (LIC), by devising a new convolution algorithm that uses a time-accurate value scattering scheme to model the texture advection. In addition, our algorithm maintains the coherence of the flow animation by successively updating the convolution results over time. Furthermore, we propose a parallel UFLIC algorithm that can achieve high load-balancing for multiprocessor computers with shared memory architecture. We demonstrate the effectiveness of our new algorithm by presenting image snapshots from several CFD case studies.
IEEE Trans. Vis. Comput. Graph.	Fast Horizon Computation at All Points of a Terrain With Visibility and Shading Applications.	A. James Stewart	1998	A terrain is most often represented with a digital elevation map consisting of a set of sample points from the terrain surface. This paper presents a fast and practical algorithm to compute the horizon, or skyline, at all sample points of a terrain. The horizons are useful in a number of applications, including the rendering of self-shadowing displacement maps, visibility culling for faster flight simulation, and rendering of cartographic data. Experimental and theoretical results are presented which show that the algorithm is more accurate that previous algorithms and is faster than previous algorithms in terrains of more than 100,000 sample points.
IEEE Trans. Vis. Comput. Graph.	Ray-Tracing Triangular Trimmed Free-Form Surfaces.	Wolfgang Stürzlinger	1998	This paper presents a new approach to rendering triangular algebraic free-form surfaces. A hierarchical subdivision of the surface with associated tight bounding volumes provides for quick identification of the surface regions likely to be hit by a ray. For each leaf of the hierarchy, an approximation to the corresponding surface region is stored. The approximation is used to compute a good starting point for the iteration, which ensures rapid convergence. Trimming curves are described by a tree of trimming primitives, such as squares, circles, polygons, and free-form curves, combined with Boolean operations. For trimmed surfaces, an irregular adaptive subdivision is constructed to quickly eliminate all parts outside the trimming curve from consideration during rendering. Cost heuristics are introduced to optimize the rendering time further.
IEEE Trans. Vis. Comput. Graph.	A High Accuracy Volume Renderer for Unstructured Data.	Peter L. Williams,Nelson L. Max,Clifford M. Stein	1998	This paper describes a volume rendering system for unstructured data, especially finite element data, that creates images with very high accuracy. The system will currently handle meshes whose cells are either linear or quadratic tetrahedra. Compromises or approximations are not introduced for the sake of efficiency. Whenever possible, exact mathematical solutions for the radiance integrals involved and for interpolation are used. The system will also handle meshes with mixed cell types: tetrahedra, bricks, prisms, wedges, and pyramids, but not with high accuracy. Accurate semitransparent shaded isosurfaces may be embedded in the volume rendering. For very small cells, subpixel accumulation by splatting is used to avoid sampling error. A revision to an existing accurate visibility ordering algorithm is described, which includes a correction and a method for dramatically increasing its efficiency. Finally, hardware assisted projection and compositing are extended from tetrahedra to arbitrary convex polyhedra.
CVPR	Trajectory Triangulation of Lines: Reconstruction of a 3D point Moving along a Line from a Monocular Image Sequence.	Shai Avidan,Amnon Shashua	1999	Trajectory Triangulation of Lines: Reconstruction of a 3D point Moving along a Line from a Monocular Image Sequence.
CVPR	Automatic Reconstruction of Piecewise Planar Models from Multiple Views.	C. Baillard,Andrew Zisserman	1999	Automatic Reconstruction of Piecewise Planar Models from Multiple Views.
CVPR	Linear Self-Calibration of a Rotating and Zooming Camera.	Lourdes de Agapito,Eric Hayman,Richard I. Hartley	1999	Linear Self-Calibration of a Rotating and Zooming Camera.
CVPR	Global Measures of Coherence for Edge Detector Evaluation.	Simon Baker,Shree K. Nayar	1999	Global Measures of Coherence for Edge Detector Evaluation.
CVPR	Edge Preserving Orientation Adaptive Filtering.	Peter Bakker,Lucas J. van Vliet,Piet W. Verbeek	1999	Edge Preserving Orientation Adaptive Filtering.
CVPR	Factorization as a Rank 1 Problem.	Pedro M. Q. Aguiar,José M. F. Moura	1999	Factorization as a Rank 1 Problem.
CVPR	Bias Field Estimation and Adaptive Segmentation of MRI Data Using a Modified Fuzzy C-Means Algorithm.	Mohamed N. Ahmed,Sameh M. Yamany,Aly A. Farag,Thomas Moriarty	1999	Bias Field Estimation and Adaptive Segmentation of MRI Data Using a Modified Fuzzy C-Means Algorithm.
CVPR	Space-Variant Dynamic Neural Fields for Visual Attention.	Ingo Ahrns,Heiko Neumann	1999	Space-Variant Dynamic Neural Fields for Visual Attention.
CVPR	A New Multi-Level Framework for Deformable Contour Optimization.	Yusuf Sinan Akgul,Chandra Kambhamettu	1999	A New Multi-Level Framework for Deformable Contour Optimization.
CVPR	Graph-Theoretic Clustering for Image Grouping and Retrieval.	Selim Aksoy,Robert M. Haralick	1999	Graph-Theoretic Clustering for Image Grouping and Retrieval.
CVPR	Advances in Daylight Statistical Colour Modelling.	Daniel C. Alexander	1999	Advances in Daylight Statistical Colour Modelling.
CVPR	Elastic Matching of Diffusion Tensor MRIs.	Daniel C. Alexander,James C. Gee,Ruzena Bajcsy	1999	Elastic Matching of Diffusion Tensor MRIs.
CVPR	Ligature Instabilities in the Perceptual Organization of Shape.	Jonas August,Steven W. Zucker,Kaleem Siddiqi	1999	Ligature Instabilities in the Perceptual Organization of Shape.
CVPR	Combining Central and Peripheral Vision for Reactive Robot Navigation.	Antonis A. Argyros,Fredrik Bergholm	1999	Combining Central and Peripheral Vision for Reactive Robot Navigation.
CVPR	Audio-Visual Person Verification.	S. Ben-Yacoub,Juergen Luettin,Kenneth Jonsson,Jiri Matas,Josef Kittler	1999	Audio-Visual Person Verification.
CVPR	Explaining Optical Flow Events with Parameterized Spatio-Temporal Models.	Michael J. Black	1999	Explaining Optical Flow Events with Parameterized Spatio-Temporal Models.
CVPR	Eigenshapes for 3D Object Recognition in Range Data.	Richard J. Campbell,Patrick J. Flynn	1999	Eigenshapes for 3D Object Recognition in Range Data.
CVPR	Performance Prediction and Validation for Object Recognition.	Michael Boshra,Bir Bhanu	1999	Performance Prediction and Validation for Object Recognition.
CVPR	Toward Recovering Shape and Motion of 3D Curves from Multi-View Image Sequences.	Rodrigo L. Carceroni,Kiriakos N. Kutulakos	1999	Toward Recovering Shape and Motion of 3D Curves from Multi-View Image Sequences.
CVPR	Calibration of Image Sequences for Model Visualization.	Adrian Broadhurst,Roberto Cipolla	1999	Calibration of Image Sequences for Model Visualization.
CVPR	Shape from Video.	Tomás Brodský,Cornelia Fermüller,Yiannis Aloimonos	1999	Shape from Video.
CVPR	Recognition of Strings Using Nonstationary Markovian Models: An Application in ZIP Code Recognition.	Djamel Bouchaffra,Venu Govindaraju,Sargur N. Srihari	1999	Recognition of Strings Using Nonstationary Markovian Models: An Application in ZIP Code Recognition.
CVPR	What Do Planar Shadows Tell About Scene Geometry?.	Jean-Yves Bouguet,Markus Weber,Pietro Perona	1999	What Do Planar Shadows Tell About Scene Geometry?.
CVPR	Edge Detector Evaluation Using Empirical ROC Curves.	Kevin W. Bowyer,Christine Kranenburg,Sean Dougherty	1999	Edge Detector Evaluation Using Empirical ROC Curves.
CVPR	A New Bayesian Framework for Object Recognition.	Yuri Boykov,Daniel P. Huttenlocher	1999	We describe a new approach to feature-based object recognition, using maximum a posteriori (MAP) estimation under a Markov random field (MRF) model. The main advantage of this approach is that it allows explicit modeling of dependencies between individual features of an object. For instance, we use the approach to model the fact that mismatched features due to partial occlusions tend to form spatially coherent groups rather than being independent. Efficient computation of the MAP estimate in our framework can be accomplished by finding a minimum cut on an appropriately defined graph. An even more efficient approximation, that does not use graph cuts, is also presented. This approximation technique, which we call spatially coherent matching (SCM), is closely related to generalized Hausdorff matching. We report some Monte Carlo experiments showing that the SCM technique improves substantially on the tradeoff between correct detection and false alarms compared with previous feature matching methods such as the Hausdorff distance.
CVPR	Fast, Reliable Head Tracking under Varying Illumination.	Marco La Cascia,Stan Sclaroff	1999	"Abstract An improved technique for 3D head tracking under varying illumination conditions is proposed. The head is modeled as a texture mapped cylinder. Tracking is formulated as an image registration problem in the cylinder''s texture map image. To solve the registration problem in the presence of lighting variation and head motion, the residual error of registration is modeled as a linear combination of texture warping templates and orthogonal illumination templates. Fast and stable on-line tracking is then achieved via regularized, weighted least squares minimization of the registration error. The regularization term tends to limit potential ambiguities that arise in the warping and illumination templates. It enables stable tracking over extended sequences. Tracking does not require a precise initial fit of the model; the system is initialized automatically using a simple 2-D face detector. The only assumption is that the target is facing the camera in the first frame of the sequence. The warping templates are computed at the first frame of the sequence. Illumination templates are precomputed off-line over a training set of face images collected under varying lighting conditions. Experiments in tracking are reported."
CVPR	A Multiple Hypothesis Approach to Figure Tracking.	Tat-Jen Cham,James M. Rehg	1999	A Multiple Hypothesis Approach to Figure Tracking.
CVPR	Evaluation of Texture Segmentation Algorithms.	Kyong I. Chang,Kevin W. Bowyer,Munish Sivagurunath	1999	Evaluation of Texture Segmentation Algorithms.
CVPR	Object Recognition with Color Cooccurrence Histograms.	Peng Chang,John Krumm	1999	Object Recognition with Color Cooccurrence Histograms.
CVPR	A Volumetric Stereo Matching Method: Application to Image-Based Modeling.	Qian Chen,Gérard G. Medioni	1999	A Volumetric Stereo Matching Method: Application to Image-Based Modeling.
CVPR	Efficient Iterative Solution to M-View Projective Reconstruction Problem.	Qian Chen,Gérard G. Medioni	1999	Efficient Iterative Solution to M-View Projective Reconstruction Problem.
CVPR	Probabilistic Recognition of Activity using Local Appearance.	Olivier Chomat,James L. Crowley	1999	Probabilistic Recognition of Activity using Local Appearance.
CVPR	An Integral Formulation for Differential Photometric Stereo.	James J. Clark,Holly Pekau	1999	An Integral Formulation for Differential Photometric Stereo.
CVPR	Detecting and Tracking Moving Objects for Video Surveillance.	Isaac Cohen,Gérard G. Medioni	1999	Detecting and Tracking Moving Objects for Video Surveillance.
CVPR	Calibration of an Outdoor Active Camera System.	Robert T. Collins,Yanghai Tsin	1999	Calibration of an Outdoor Active Camera System.
CVPR	A Probabilistic Framework for Embedded Face and Facial Expression Recognition.	Antonio Colmenarez,Brendan J. Frey,Thomas S. Huang	1999	A Probabilistic Framework for Embedded Face and Facial Expression Recognition.
CVPR	Model Based Segmentation of Nuclei.	Ge Cong,Bahram Parvin	1999	Model Based Segmentation of Nuclei.
CVPR	Extracting Textured Vertical Facades From Controlled Close-Range Imagery.	Satyan R. Coorg,Seth J. Teller	1999	Extracting Textured Vertical Facades From Controlled Close-Range Imagery.
CVPR	Simultaneous Extraction of Functional Face Subspaces.	Nicholas Costen,Timothy F. Cootes,Gareth J. Edwards,Christopher J. Taylor	1999	Simultaneous Extraction of Functional Face Subspaces.
CVPR	Dynamic Occluding Contours: A New External-Energy Term for Snakes.	Michele Covell,Trevor Darrell	1999	Dynamic Occluding Contours: A New External-Energy Term for Snakes.
CVPR	Real-Time Periodic Motion Detection, Analysis, and Applications.	Ross Cutler,Larry S. Davis	1999	Real-Time Periodic Motion Detection, Analysis, and Applications.
CVPR	Combining Information using Hard Constraints.	Douglas DeCarlo,Dimitris N. Metaxas	1999	Combining Information using Hard Constraints.
CVPR	Using the Condensation Algorithm for Robust, Vision-based Mobile Robot Localization.	Frank Dellaert,Wolfram Burgard,Dieter Fox,Sebastian Thrun	1999	Using the Condensation Algorithm for Robust, Vision-based Mobile Robot Localization.
CVPR	A Projective Framework for Scene Segmentation in the Presence of Moving Objects.	David Demirdjian,Radu Horaud	1999	A Projective Framework for Scene Segmentation in the Presence of Moving Objects.
CVPR	Color Image Segmentation.	Yining Deng,B. S. Manjunath,Hyundoo Shin	1999	Color Image Segmentation.
CVPR	Stereo Correspondence from Motion Correspondence.	Fadi Dornaika,Ronald Chung	1999	Stereo Correspondence from Motion Correspondence.
CVPR	Visual Tracking and Control using Lie Algebras.	Tom Drummond,Roberto Cipolla	1999	Visual Tracking and Control using Lie Algebras.
CVPR	A Fast Scheme for Altering Resolution in the Compressed Domain.	Rakesh Dugad,Narendra Ahuja	1999	A Fast Scheme for Altering Resolution in the Compressed Domain.
CVPR	Learning 2D Shape Models.	Nicolae Duta,Anil K. Jain,Marie-Pierre Dubuisson-Jolly	1999	Learning 2D Shape Models.
CVPR	The Customized-Queries Approach to CBIR Using EM.	Jennifer G. Dy,Carla E. Brodley,Avinash C. Kak,Chi-Ren Shyu,Lynn S. Broderick	1999	The Customized-Queries Approach to CBIR Using EM.
CVPR	Improving Identification Performance by Integrating Evidence from Sequences.	Gareth J. Edwards,Christopher J. Taylor,Timothy F. Cootes	1999	Improving Identification Performance by Integrating Evidence from Sequences.
CVPR	Elastic Registration of Medical Images Using Radial Basis Functions with Compact Support.	Mike Fornefett,Karl Rohr,H. Siegfried Stiehl	1999	Elastic Registration of Medical Images Using Radial Basis Functions with Compact Support.
CVPR	Sampling, Resampling and Colour Constancy.	David A. Forsyth	1999	Sampling, Resampling and Colour Constancy.
CVPR	Surface Reconstruction from Multiple Aerial Images in Dense Urban Area.	M. Fradkin,Michel Roux,Henri Maître,Ugur M. Leloglu	1999	Surface Reconstruction from Multiple Aerial Images in Dense Urban Area.
CVPR	Estimating Mixture Models of Images and Inferring Spatial Transformations Using the EM Algorithm.	Brendan J. Frey,Nebojsa Jojic	1999	Estimating Mixture Models of Images and Inferring Spatial Transformations Using the EM Algorithm.
CVPR	Separating Reflections and Lighting Using Independent Components Analysis.	Hany Farid,Edward H. Adelson	1999	Separating Reflections and Lighting Using Independent Components Analysis.
CVPR	Progressive Probabilistic Hough Transform for Line Detection.	Charles Galambos,Josef Kittler,Jiri Matas	1999	Progressive Probabilistic Hough Transform for Line Detection.
CVPR	Virtual Snakes for Occlusion Analysis.	Ben Galvin,Brendan McCane,Kevin Novins	1999	Virtual Snakes for Occlusion Analysis.
CVPR	Motion Segmentation: A Synergistic Approach.	Cornelia Fermüller,Tomás Brodský,Yiannis Aloimonos	1999	Motion Segmentation: A Synergistic Approach.
CVPR	Statistical Biases in Optic Flow.	Cornelia Fermüller,Robert Pless,Yiannis Aloimonos	1999	Statistical Biases in Optic Flow.
CVPR	Stochastic Image Segmentation by Typical Cuts.	Yoram Gdalyahu,Daphna Weinshall,Michael Werman	1999	Stochastic Image Segmentation by Typical Cuts.
CVPR	Parameterized Image Varieties and Estimation with Bilinear Constraints.	Yakup Genc,Jean Ponce,Yoram Leedan,Peter Meer	1999	Parameterized Image Varieties and Estimation with Bilinear Constraints.
CVPR	On the Intrinsic Reconstruction of Shape from its Symmetries.	Peter J. Giblin,Benjamin B. Kimia	1999	The main question we address is: What is the minimal information required to generate closed, nonintersecting planar boundaries? For this paper, we restrict ¿shape¿ to this meaning. More precisely, we examine whether the medial axis, together with dynamics, can serve as a language to design shapes and to effect shape changes, e.g., for modeling, to generate a morph sequence, etc. We represent the medial axis together with a direction of flow along the axis as the shock graph and examine the reconstruction of shape along each of the three types of medial axis points, (labeled A_1^2, A_1^3, A_3; see below for the A notation) and the associated six types of shock points. First, we show that the tangent and curvature of the medial axis and the speed and acceleration of the shock with respect to time of propagation, i.e., first and second order geometric and dynamic properties, are sufficient to determine the boundary tangent and curvature at corresponding points of the boundary. This implies that a rather coarse sampling of the symmetry axis, its tangent, curvature, speed, and acceleration is sufficient to regenerate accurately a local neighborhood of shape at regular axis points (A_1^2). We also show how higher order differential properties of the axis can be related to the higher-order differential properties of the boundary of the same order. Second, we examine the reconstruction of shape at branch points (A_1^3) where three regular branches are joined. We show that the three pairs of geometry (that is, curvature) and dynamics (that is, acceleration) must satisfy certain constraints. Finally, we derive similar results for the end points of shock branches (A_3 points). These formulas completely specify the local reconstruction of a shape from its shock-graph or medial axis and the conditions required to form a coherent shape from the medial axis.
CVPR	Gesture Localization and Recognition using Probabilistic Visual Learning.	Raouf Hamdan,Fabrice Heitz,Laurent Thoraval	1999	Gesture Localization and Recognition using Probabilistic Visual Learning.
CVPR	Planar Catadioptric Stereo: Geometry and Calibration.	Joshua Gluckman,Shree K. Nayar	1999	Planar Catadioptric Stereo: Geometry and Calibration.
CVPR	A Fuzzy K-NN Algorithm Using Weights from the Variance of Membership Values.	Joon H. Han,Yoon K. Kim	1999	A Fuzzy K-NN Algorithm Using Weights from the Variance of Membership Values.
CVPR	Yet another Method for Pose Estimation: A Probabilistic Approach using Points, Lines, and Cylinders.	Robert Hanek,Nassir Navab,Mirko Appel	1999	Yet another Method for Pose Estimation: A Probabilistic Approach using Points, Lines, and Cylinders.
CVPR	Multiscale Image Registration Using Scale Trace Correlation.	Bruce B. Hansen,Bryan S. Morse	1999	Multiscale Image Registration Using Scale Trace Correlation.
CVPR	Background Estimation and Removal Based on Range and Color.	Gaile G. Gordon,Trevor Darrell,Michael Harville,John Woodfill	1999	Background Estimation and Removal Based on Range and Color.
CVPR	An Efficient Recursive Factorization Method for Determining Structure from Motion.	Yanhua Li,Michael J. Brooks	1999	An Efficient Recursive Factorization Method for Determining Structure from Motion.
CVPR	Norm-Based Face Recognition.	Daniel B. Graham,Nigel M. Allinson	1999	Norm-Based Face Recognition.
CVPR	Invariant Recognition in Hyperspectral Images.	Glenn Healey,David Slater	1999	Invariant Recognition in Hyperspectral Images.
CVPR	3D Geometric Invariant Alignment of Surfaces with Application in Brain Mapping.	Walid S. Ibrahim,Fernand S. Cohen	1999	3D Geometric Invariant Alignment of Surfaces with Application in Brain Mapping.
CVPR	Visual Recognition of Multi-Agent Action Using Binary Temporal Relations.	Stephen S. Intille,Aaron F. Bobick	1999	Visual Recognition of Multi-Agent Action Using Binary Temporal Relations.
CVPR	Applying Perceptual Grouping to Content-Based Image Retrieval: Building Images.	Qasim Iqbal,Jake K. Aggarwal	1999	Applying Perceptual Grouping to Content-Based Image Retrieval: Building Images.
CVPR	Projective Rectification Without Epipolar Geometry.	Francesco Isgrò,Emanuele Trucco	1999	Projective Rectification Without Epipolar Geometry.
CVPR	Multi-Scale Feature Selection in Stereo.	Hiroshi Ishikawa	1999	Multi-Scale Feature Selection in Stereo.
CVPR	FingerCode: A Filterbank for Fingerprint Representation and Matching.	Anil K. Jain,Salil Prabhakar,Lin Hong,Sharath Pankanti	1999	FingerCode: A Filterbank for Fingerprint Representation and Matching.
CVPR	Reconstruction of Linearly Parameterized Models from Single Images with a Camera of Unknown Focal Length.	David Jelinek,Camillo J. Taylor	1999	This paper deals with the problem of recovering the dimensions of an object and its pose from a single image acquired with a camera of unknown focal length. It is assumed that the object in question can be modeled as a polyhedron where the coordinates of the vertices can be expressed as a linear function of a dimension vector, $\lambda$. The reconstruction program takes as input, a set of correspondences between features in the model and features in the image. From this information, the program determines an appropriate projection model for the camera (scaled orthographic or perspective), the dimensions of the object, its pose relative to the camera and, in the case of perspective projection, the focal length of the camera. This paper describes how the reconstruction problem can be framed as an optimization over a compact set with low dimension¿no more than four. This optimization problem can be solved efficiently by coupling standard nonlinear optimization techniques with a multistart method which generates multiple starting points for the optimizer by sampling the parameter space uniformly. The result is an efficient, reliable solution system that does not require initial estimates for any of the parameters being estimated.
CVPR	Joint Detection for Potsherds of Broken Earthenware.	Kenta Hori,Masakazu Imai,Tsukasa Ogasawara	1999	Joint Detection for Potsherds of Broken Earthenware.
CVPR	Statistics of Natural Images and Models.	Jinggang Huang,David Mumford	1999	Statistics of Natural Images and Models.
CVPR	Statistical Color Models with Application to Skin Detection.	Michael J. Jones,James M. Rehg	1999	The existence of large image datasets such as the set of photos on the World Wide Web make it possible to build powerful generic models for low-level image attributes like color using simple histogram learning techniques. We describe the construction of color models for skin and non-skin classes from a dataset of nearly 1 billion labelled pixels. These classes exhibit a surprising degree of separability which we exploit by building a skin pixel detector achieving a detection rate of 80% with 8.5% false positives. We compare the performance of histogram and mixture models in skin detection and find histogram models to be superior in accuracy and computational cost. Using aggregate features computed from the skin pixel detector we build a surprisingly effective detector for naked people. Our results suggest that color can be a more powerful cue for detecting people in unconstrained imagery than was previously suspected. We believe this work is the most comprehensive and detailed exploration of skin color models to date.
CVPR	Detection and Removal of Line Scratches in Motion Picture Films.	Laurent Joyeux,Olivier Buisson,Bernard Besserer,Samia Boukir	1999	Detection and Removal of Line Scratches in Motion Picture Films.
CVPR	Adaptive Balloon Models.	Xiaobo Li,Jiankang Wang	1999	Adaptive Balloon Models.
CVPR	Critical Motions in Euclidean Structure from Motion.	Fredrik Kahl,Bill Triggs	1999	Critical Motions in Euclidean Structure from Motion.
CVPR	Automatic Aircraft Recognition: Toward Using Human Similarity Measure in a Recognition System.	Behrooz Kamgar-Parsi,Behzad Kamgar-Parsi,Anil K. Jain	1999	Automatic Aircraft Recognition: Toward Using Human Similarity Measure in a Recognition System.
CVPR	A Dependency-based Framework of Combining Multiple Experts for the Recognition of Unconstrained Handwritten Numerals.	Hee-Joong Kang,Seong-Whan Lee	1999	A Dependency-based Framework of Combining Multiple Experts for the Recognition of Unconstrained Handwritten Numerals.
CVPR	Minimum-Entropy Models of Scene Activity.	Vera Kettnaker,Matthew Brand	1999	Minimum-Entropy Models of Scene Activity.
CVPR	Bayesian Multi-Camera Surveillance.	Vera Kettnaker,Ramin Zabih	1999	Bayesian Multi-Camera Surveillance.
CVPR	A Convolver-Based Real-Time Stereo Machine (SAZAN).	Shigeru Kimura,Katsuyiki Nakano,Tetsuya Sinbo,Hiroyoshi Yamaguchi,Eiji Kawamura	1999	A Convolver-Based Real-Time Stereo Machine (SAZAN).
CVPR	Fundamental Bounds on Edge Detection: An Information Theoretic Evaluation of Different Edge Cues.	Scott Konishi,Alan L. Yuille,James M. Coughlan,Song Chun Zhu	1999	Fundamental Bounds on Edge Detection: An Information Theoretic Evaluation of Different Edge Cues.
CVPR	Robust and Efficient Image Alignment with Spatially Varying Illumination Models.	Shang-Hong Lai,Ming Fang	1999	Robust and Efficient Image Alignment with Spatially Varying Illumination Models.
CVPR	A Biprism-Stereo Camera System.	Doo Hyun Lee,In-So Kweon,Roberto Cipolla	1999	A Biprism-Stereo Camera System.
CVPR	Sensor Planning for a Trinocular Active Vision System.	Peter Lehel,Elsayed E. Hemayed,Aly A. Farag	1999	Sensor Planning for a Trinocular Active Vision System.
CVPR	Automatic Differentiation Facilitates OF-Integration into Steering-Angle-Based Road Vehicle Tracking.	Holger Leuck,Hans-Hellmut Nagel	1999	Automatic Differentiation Facilitates OF-Integration into Steering-Angle-Based Road Vehicle Tracking.
CVPR	Optimal Multi-Scale Matching.	Michael S. Lew,Thomas S. Huang	1999	Optimal Multi-Scale Matching.
CVPR	Image Interpolation by Joint View Triangulation.	Maxime Lhuillier,Long Quan	1999	Image Interpolation by Joint View Triangulation.
CVPR	User Assisted Modeling of Buildings from Aerial Images.	J. Li,Ramakant Nevatia,Sanjay Noronha	1999	User Assisted Modeling of Buildings from Aerial Images.
CVPR	Computing Rectifying Homographies for Stereo Vision.	Charles T. Loop,Zhengyou Zhang	1999	Computing Rectifying Homographies for Stereo Vision.
CVPR	Using Planar Parallax to Estimate the Time-to-Contact.	Manolis I. A. Lourakis,Stelios C. Orphanoudakis	1999	Using Planar Parallax to Estimate the Time-to-Contact.
CVPR	Model-Guided Segmentation of Corpus Callosum in MR Images.	Arvid Lundervold,Torfinn Taxt,Nicolae Duta,Anil K. Jain	1999	Model-Guided Segmentation of Corpus Callosum in MR Images.
CVPR	Real-Time Gabor-Type Filtering Using Analog Focal Plane Image Processors.	Bertram Emil Shi	1999	Real-Time Gabor-Type Filtering Using Analog Focal Plane Image Processors.
CVPR	Area-Based Computation of Stereo Disparity with Model-Based Window Size Selection.	Robert Maas,Bart M. ter Haar Romeny,Max A. Viergever	1999	Area-Based Computation of Stereo Disparity with Model-Based Window Size Selection.
CVPR	Deformable Shape Detection and Description via Model-Based Region.	Lifeng Liu,Stan Sclaroff	1999	Deformable Shape Detection and Description via Model-Based Region.
CVPR	Interpolating View and Scene Motion by Dynamic View Morphing.	Russell A. Manning,Charles R. Dyer	1999	Interpolating View and Scene Motion by Dynamic View Morphing.
CVPR	Face Recognition Using Shape and Texture.	Chengjun Liu,Harry Wechsler	1999	Face Recognition Using Shape and Texture.
CVPR	Optimal Rigid Motion Estimation and Performance Evaluation with Bootstrap.	Bogdan Matei,Peter Meer	1999	Optimal Rigid Motion Estimation and Performance Evaluation with Bootstrap.
CVPR	Constrained Self-Calibration.	Jeffrey Mendelsohn,Konstantinos Daniilidis	1999	Constrained Self-Calibration.
CVPR	Estimation of Epipolar Geometry from Apparent Contours: Affine and Circular Motion Cases.	Paulo R. S. Mendonça,Roberto Cipolla	1999	Estimation of Epipolar Geometry from Apparent Contours: Affine and Circular Motion Cases.
CVPR	A Simple Technique for Self-Calibration.	Paulo R. S. Mendonça,Roberto Cipolla	1999	A Simple Technique for Self-Calibration.
CVPR	Deformable Template and Distribution Mixture-Based Data Modeling for the Endocardial Contour Tracking in an Echographic Sequence.	Max Mignotte,Jean Meunier	1999	Deformable Template and Distribution Mixture-Based Data Modeling for the Endocardial Contour Tracking in an Echographic Sequence.
CVPR	Recognizing Color Patterns Irrespective of Viewpoint and Illumination.	Florica Mindru,Theo Moons,Luc J. Van Gool	1999	Recognizing Color Patterns Irrespective of Viewpoint and Illumination.
CVPR	Radiometric Self Calibration.	Tomoo Mitsunaga,Shree K. Nayar	1999	Radiometric Self Calibration.
CVPR	Folded Catadioptric Cameras.	Shree K. Nayar,Venkata Peri	1999	Folded Catadioptric Cameras.
CVPR	Detection of Gradual Transitions through Temporal Slice Analysis.	Chong-Wah Ngo,Ting-Chuen Pong,Roland T. Chin	1999	Detection of Gradual Transitions through Temporal Slice Analysis.
CVPR	Toboggan-Based Intelligent Scissors with a Four-Parameter Edge Model.	Eric N. Mortensen,William A. Barrett	1999	Toboggan-Based Intelligent Scissors with a Four-Parameter Edge Model.
CVPR	Pose Clustering with Density Estimation and Structural Constraints.	Simon Moss,Edwin R. Hancock	1999	Pose Clustering with Density Estimation and Structural Constraints.
CVPR	Integrating Shape from Shading and Range Data Using Neural Networks.	Mostafa G.-H. Mostafa,Sameh M. Yamany,Aly A. Farag	1999	Integrating Shape from Shading and Range Data Using Neural Networks.
CVPR	Eigen-Texture Method: Appearance Compression Based on 3D Model.	Ko Nishino,Yoichi Sato,Katsushi Ikeuchi	1999	Image-based and model-based methods are two representative rendering methods for generating virtual images of objects from their real images. Extensive research on these two methods has been made in CV and CG communities. However, both methods still have several drawbacks when it comes to applying them to the mixed reality where we integrate such virtual images with real background images. To overcome these difficulties, we propose a new method, which we refer to as the Eigen-Texture method. The proposed method samples appearances of a real object under various illumination and viewing conditions, and compresses them in the 2D coordinate system defined on the 3D model surface. The 3D model is generated from a sequence of range images. The Eigen-Texture method is practical because it does not require any detailed reflectance analysis of the object surface, and has great advantages due to the accurate 3D geometric models. This paper describes the method, and reports on its implementation.
CVPR	Visual Signature Verification Using Affine Arc-Length.	Mario E. Munich,Pietro Perona	1999	Visual Signature Verification Using Affine Arc-Length.
CVPR	3D Deformable Image Matching Using Multiscale Minimization of Global Energy Functions.	Olivier Musse,Fabrice Heitz,Jean-Paul Armspach	1999	3D Deformable Image Matching Using Multiscale Minimization of Global Energy Functions.
CVPR	Shape from Recognition and Learning: Recovery of 3-D Face Shapes.	Dibyendu Nandy,Jezekiel Ben-Arie	1999	Shape from Recognition and Learning: Recovery of 3-D Face Shapes.
CVPR	A New Structure--From--Motion Ambiguity.	John Oliensis	1999	This paper demonstrates the existence of a new, approximate, intrinsic ambiguity in Euclidean structure from motion (SFM) which occurs as generically as the bas-relief ambiguity but, unlike it, strengthens for scenes with more depth variation. The ambiguity does not occur in projective SFM, but the reasons for this make projective reconstructions more likely to have large errors. Our analysis gives a semiquantitative characterization of the least-squares error surface over a domain complementary to that analyzed by Jepson, Heeger, and Maybank. As part of our analysis, we show that the least-squares error for infinitesimal motion¿the optical-flow error¿gives a good approximation to the least-squares error for moderate finite motions. We propose that many high-error local minima occur for epipoles in or near the image. We also establish the existence of a new local minimum in minimizing over the rotation, given the translation direction.
CVPR	Toward a Scale-Space Aspect Graph: Solids of Revolution.	Sung-il Pae,Jean Ponce	1999	Toward a Scale-Space Aspect Graph: Solids of Revolution.
CVPR	Exploiting the Dependencies in Information Fusion.	Hao Pan,Zhi-Pei Liang,Thomas S. Huang	1999	Exploiting the Dependencies in Information Fusion.
CVPR	Unifying Boundary and Region-Based Information for Geodesic Active Tracking.	Nikos Paragios,Rachid Deriche	1999	Unifying Boundary and Region-Based Information for Geodesic Active Tracking.
CVPR	Geodesic Active Contours for Supervised Texture Segmentation.	Nikos Paragios,Rachid Deriche	1999	Geodesic Active Contours for Supervised Texture Segmentation.
CVPR	Histogram Clustering for Unsupervised Image Segmentation.	Jan Puzicha,Joachim M. Buhmann,Thomas Hofmann	1999	Histogram Clustering for Unsupervised Image Segmentation.
CVPR	A Computational Approach to Semantic Event Detection.	Richard J. Qian,Niels C. Haering,M. Ibrahim Sezan	1999	A Computational Approach to Semantic Event Detection.
CVPR	Minimal Projective Reconstruction with Missing Data.	Long Quan,Anders Heyden,Fredrik Kahl	1999	Minimal Projective Reconstruction with Missing Data.
CVPR	Time-Series Classification Using Mixed-State Dynamic Bayesian Networks.	Vladimir Pavlovic,Brendan J. Frey,Thomas S. Huang	1999	Time-Series Classification Using Mixed-State Dynamic Bayesian Networks.
CVPR	Stereo Panorama with a Single Camera.	Shmuel Peleg,Moshe Ben-Ezra	1999	Stereo Panorama with a Single Camera.
CVPR	Simultaneous Depth Recovery and Image Restoration from Defocused Images.	A. N. Rajagopalan,Subhasis Chaudhuri	1999	Simultaneous Depth Recovery and Image Restoration from Defocused Images.
CVPR	Toward Learning Visual Discrimination Strategies.	Justus H. Piater,Roderic A. Grupen	1999	Toward Learning Visual Discrimination Strategies.
CVPR	Independent Motion: The Importance of History.	Robert Pless,Tomás Brodský,Yiannis Aloimonos	1999	Independent Motion: The Importance of History.
CVPR	A Framework for Learning Query Concepts in Image Classification.	Aparna Lakshmi Ratan,Oded Maron,W. Eric L. Grimson,Tomás Lozano-Pérez	1999	A Framework for Learning Query Concepts in Image Classification.
CVPR	Vision-Based Speaker Detection Using Bayesian Networks.	James M. Rehg,Kevin P. Murphy,Paul W. Fieguth	1999	Vision-Based Speaker Detection Using Bayesian Networks.
CVPR	The Quotient Image: Class Based Recognition and Synthesis under Varying Illumination Conditions.	Tammy Riklin-Raviv,Amnon Shashua	1999	The Quotient Image: Class Based Recognition and Synthesis under Varying Illumination Conditions.
CVPR	3D Trajectory Recovery for Tracking Multiple Objects and Trajectory Guided Recognition of Actions.	Rómer Rosales,Stan Sclaroff	1999	"Abstract A mechanism is proposed that integrates low-level (image processing), mid-level (recursive 3D trajectory estimation), and high-level (action recognition) processes. It is assumed that the system observes multiple moving objects via a single, uncalibrated video camera. A novel extended Kalman filter formulation is used in estimating the relative 3D motion trajectories up to a scale factor. The recursive estimation process provides a prediction and error measure that is exploited in higher-level stages of action recognition. Conversely, higher-level mechanisms provide feedback that allows the system to reliably segment and maintain the tracking of moving objects before, during, and after occlusion. The 3D trajectory, occlusion, and segmentation information are utlized in extracting stabilized views of the moving object. Trajectory-guided recognition (TGR) is proposed as a new and efficient method for adaptive classification of action. The TGR approach is demonstrated using ``motion history images'''' that are then recognized via a mixture of Gaussian classifier. The system was tested in recognizing various dynamic human outdoor activities; e.g., running, walking, roller blading, and cycling. Experiments with synthetic data sets are used to evaluate stability of the trajectory estimator with respect to noise."
CVPR	A Bayesian Approach To Spread Spectrum Watermark Detection and Secure Copyright Protection for Digital Image Libraries.	Joseph Ó Ruanaidh,Gabriela Csurka	1999	A Bayesian Approach To Spread Spectrum Watermark Detection and Secure Copyright Protection for Digital Image Libraries.
CVPR	Projective Rotations Applied to a Pan-Tilt Stereo Head.	Andreas Ruf,Radu Horaud	1999	Projective Rotations Applied to a Pan-Tilt Stereo Head.
CVPR	Color Edge Detection with the Compass Operator.	Mark A. Ruzon,Carlo Tomasi	1999	Color Edge Detection with the Compass Operator.
CVPR	Shape Reconstruction in Projective Grid Space from Large Number of Images.	Hideo Saito,Takeo Kanade	1999	Shape Reconstruction in Projective Grid Space from Large Number of Images.
CVPR	Measurement of Surface Orientations of Transparent Objects Using Polarization in Highlight.	Megumi Saito,Hiroshi Kashiwagi,Yoichi Sato,Katsushi Ikeuchi	1999	Measurement of Surface Orientations of Transparent Objects Using Polarization in Highlight.
CVPR	Simultaneous Image Classification and Restoration Using a Variational Approach.	Christophe Samson,Laure Blanc-Féraud,Josiane Zerubia,Gilles Aubert	1999	Simultaneous Image Classification and Restoration Using a Variational Approach.
CVPR	Illumination Distribution from Shadows.	Imari Sato,Yoichi Sato,Katsushi Ikeuchi	1999	The image irradiance of a three-dimensional object is known to be the function of three components: the distribution of light sources, the shape, and reflectance of a real object surface. In the past, recovering the shape and reflectance of an object surface from the recorded image brightness has been intensively investigated. On the other hand, there has been little progress in recovering illumination from the knowledge of the shape and reflectance of a real object. In this paper, we propose a new method for estimating the illumination distribution of a real scene from image brightness observed on a real object surface in that scene. More specifically, we recover the illumination distribution of the scene from a radiance distribution inside shadows cast by an object of known shape onto another object surface of known shape and reflectance. By using the occlusion information of incoming light, we are able to reliably estimate the illumination distribution of a real scene, even in a complex illumination environment.
CVPR	Perceptual Organization of Occluding Contours Generated By Opaque Surfaces.	Eric Saund	1999	Perceptual Organization of Occluding Contours Generated By Opaque Surfaces.
CVPR	The Discriminatory Power of Ordinal Measures - Towards a New Coefficient.	Stefan Scherer,Axel Pinz,Peter Werth	1999	The Discriminatory Power of Ordinal Measures - Towards a New Coefficient.
CVPR	A Structured Probabilistic Model for Recognition.	Cordelia Schmid	1999	A Structured Probabilistic Model for Recognition.
CVPR	Optimal Eigenfeature Selection by Optimal Image Registration.	Haim Schweitzer	1999	Optimal Eigenfeature Selection by Optimal Image Registration.
CVPR	Torque-Based Recursive Filtering Approach to the Recovery of 3D Articulated Motion from Image Sequences.	Hiroyuki Segawa,Takashi Totsuka	1999	Torque-Based Recursive Filtering Approach to the Recovery of 3D Articulated Motion from Image Sequences.
CVPR	Shadow Gestures: 3D Hand Pose Estimation Using a Single Camera.	Jakub Segen,Senthil Kumar	1999	Shadow Gestures: 3D Hand Pose Estimation Using a Single Camera.
CVPR	Implicit Representation and Scene Reconstruction from Probability Density Functions.	Steven M. Seitz,P. Anandan	1999	Implicit Representation and Scene Reconstruction from Probability Density Functions.
CVPR	Comparison of Edge Detectors Using an Object Recognition Task.	Min C. Shin,Dmitry B. Goldgof,Kevin W. Bowyer	1999	Comparison of Edge Detectors Using an Object Recognition Task.
CVPR	Indexing using a Spectral Encoding of Topological Structure.	Ali Shokoufandeh,Sven J. Dickinson,Kaleem Siddiqi,Steven W. Zucker	1999	Indexing using a Spectral Encoding of Topological Structure.
CVPR	Efficient Bundle Adjustment with Virtual Key Frames: A Hierarchical Approach to Multi-Frame Structure from Motion.	Heung-Yeung Shum,Zhengyou Zhang,Qifa Ke	1999	Efficient Bundle Adjustment with Virtual Key Frames: A Hierarchical Approach to Multi-Frame Structure from Motion.
CVPR	Material Classification for 3D Objects in Aerial Hyperspectral Images.	David Slater,Glenn Healey	1999	Material Classification for 3D Objects in Aerial Hyperspectral Images.
CVPR	Efficient Techniques for Wide-Angle Stereo Vision Using Surface Projection Models.	Philip W. Smith,Keith B. Johnson,Mongi A. Abidi	1999	Efficient Techniques for Wide-Angle Stereo Vision Using Surface Projection Models.
CVPR	A New Visualization Paradigm for Multispectral Imagery And Data Fusion.	Diego A. Socolinsky,Lawrence B. Wolff	1999	A New Visualization Paradigm for Multispectral Imagery And Data Fusion.
CVPR	Automatic Hierarchical Classification Using Time-Based Co-Occurrences.	Chris Stauffer	1999	Automatic Hierarchical Classification Using Time-Based Co-Occurrences.
CVPR	Adaptive Background Mixture Models for Real-Time Tracking.	Chris Stauffer,W. Eric L. Grimson	1999	Adaptive Background Mixture Models for Real-Time Tracking.
CVPR	Tracking from Multiple View Points: Self-calibration of Space and Time.	Gideon P. Stein	1999	Tracking from Multiple View Points: Self-calibration of Space and Time.
CVPR	Estimating Model Parameters and Boundaries By Minimizing a Joint, Robust Objective Function.	Charles V. Stewart,Kishore Bubna,A. G. Amitha Perera	1999	Estimating Model Parameters and Boundaries By Minimizing a Joint, Robust Objective Function.
CVPR	Robust Hierarchical Algorithm for Constructing a Mosaic from Images of the Curved Human Retina.	Charles V. Stewart,Badrinath Roysam,Ali Can	1999	Robust Hierarchical Algorithm for Constructing a Mosaic from Images of the Curved Human Retina.
CVPR	On Plane-Based Camera Calibration: A General Algorithm, Singularities, Applications.	Peter F. Sturm,Stephen J. Maybank	1999	On Plane-Based Camera Calibration: A General Algorithm, Singularities, Applications.
CVPR	Non-Metric Calibration of Wide-Angle Lenses and Polycameras.	Rahul Swaminathan,Shree K. Nayar	1999	Non-Metric Calibration of Wide-Angle Lenses and Polycameras.
CVPR	Locating Indexing Structures in Engineering Drawing Databases Using Location Hashing.	Tanveer Fathima Syeda-Mahmood	1999	Locating Indexing Structures in Engineering Drawing Databases Using Location Hashing.
CVPR	A Multi-View Approach to Motion and Stereo.	Richard Szeliski	1999	A Multi-View Approach to Motion and Stereo.
CVPR	Low-Cost Interactive Active Monocular Range Finder.	Masahiro Takatsuka,Geoff A. W. West,Svetha Venkatesh,Terry Caelli	1999	Low-Cost Interactive Active Monocular Range Finder.
CVPR	Explanation-Based Facial Motion Tracking Using a Piecewise Bézier Volume Deformation Model.	Hai Tao,Thomas S. Huang	1999	Explanation-Based Facial Motion Tracking Using a Piecewise Bézier Volume Deformation Model.
CVPR	Algebraic Curves That Work Better.	Tolga Tasdizen,Jean-Philippe Tarel,David B. Cooper	1999	Algebraic Curves That Work Better.
CVPR	Robust Visual Servoing based on Relative Orientation.	Camillo J. Taylor,James P. Ostrowski,Sang-Hack Jung	1999	Robust Visual Servoing based on Relative Orientation.
CVPR	Perceptual Organization via the Symmetry Map and Symmetry Transforms.	Hüseyin Tek,Benjamin B. Kimia	1999	Perceptual Organization via the Symmetry Map and Symmetry Transforms.
CVPR	Using a Linear Subspace Approach for Invariant Subpixel Material Identification in Airborne Hyperspectral Imagery.	Bea Thai,Glenn Healey	1999	Using a Linear Subspace Approach for Invariant Subpixel Material Identification in Airborne Hyperspectral Imagery.
CVPR	Spatial Filter Selection for Illumination-Invariant Color Texture Discrimination.	Bea Thai,Glenn Healey	1999	Spatial Filter Selection for Illumination-Invariant Color Texture Discrimination.
CVPR	The Tensors of Three Affine Views.	Torfi Thórhallsson,David W. Murray	1999	The Tensors of Three Affine Views.
CVPR	Multiple-Hand-Gesture Tracking using Multiple Cameras.	Akira Utsumi,Jun Ohya	1999	Multiple-Hand-Gesture Tracking using Multiple Cameras.
CVPR	Generic Object Detection using Model Based Segmentation.	Zhiqian Wang,Jezekiel Ben-Arie	1999	Generic Object Detection using Model Based Segmentation.
CVPR	A Novel Bayesian Method for Fitting Parametric and Non-Parametric Models to Noisy Data.	Michael Werman,Daniel Keren	1999	A Novel Bayesian Method for Fitting Parametric and Non-Parametric Models to Noisy Data.
CVPR	Robot Localization using Uncalibrated Camera Invariants.	Michael Werman,MaoLin Qiu,Subhashis Banerjee,Sumantra Dutta Roy	1999	Robot Localization using Uncalibrated Camera Invariants.
CVPR	Q-Warping: Direct Computation of Quadratic Reference Surfaces.	Yonatan Wexler,Amnon Shashua	1999	Q-Warping: Direct Computation of Quadratic Reference Surfaces.
CVPR	Subset Selection for Active Object Recognition.	Jay Winkeler,B. S. Manjunath,Shivkumar Chandrasekaran	1999	Subset Selection for Active Object Recognition.
CVPR	Data-Driven Shape-from-Shading Using Curvature Consistency.	Philip L. Worthington,Edwin R. Hancock	1999	Data-Driven Shape-from-Shading Using Curvature Consistency.
CVPR	Recognizing Hand Gesture Using Motion Trajectories.	Ming-Hsuan Yang,Narendra Ahuja	1999	Recognizing Hand Gesture Using Motion Trajectories.
CVPR	Detection and Characterization of Multiple Motion Points.	Weichuan Yu,Gerald Sommer,Steven S. Beauchemin,Konstantinos Daniilidis	1999	Detection and Characterization of Multiple Motion Points.
CVPR	Extending Superquadrics with Exponent Functions: Modeling and Reconstruction.	Lin Zhou,Chandra Kambhamettu	1999	Extending Superquadrics with Exponent Functions: Modeling and Reconstruction.
CVPR	High-Level and Generic Models for Visual Search: When Does High Level Knowledge Help?	Alan L. Yuille,James M. Coughlan	1999	High-Level and Generic Models for Visual Search: When Does High Level Knowledge Help?
CVPR	Extracting Nonrigid Motion and 3D Structure of Hurricanes from Satellite Image Sequences without Correspondences.	Lin Zhou,Chandra Kambhamettu,Dmitry B. Goldgof	1999	Extracting Nonrigid Motion and 3D Structure of Hurricanes from Satellite Image Sequences without Correspondences.
CVPR	Multi-Frame Alignment of Planes.	Lihi Zelnik-Manor,Michal Irani	1999	Multi-Frame Alignment of Planes.
CVPR	Panoramic EPI Generation and Analysis of Video from a Moving Platform with Vibration.	Zhigang Zhu,Guangyou Xu,Xueyin Lin	1999	Panoramic EPI Generation and Analysis of Video from a Moving Platform with Vibration.
CVPR	Harmonic Maps and Their Applications in Surface Matching.	Dongmei Zhang,Martial Hebert	1999	Harmonic Maps and Their Applications in Surface Matching.
CVPR	Fast, Robust, and Consistent Camera Motion Estimation.	Tong Zhang,Carlo Tomasi	1999	Fast, Robust, and Consistent Camera Motion Estimation.
CVPR	"1999 Conference on Computer Vision and Pattern Recognition (CVPR '99), 23-25 June 1999, Ft. Collins, CO, USA"		1999	"1999 Conference on Computer Vision and Pattern Recognition (CVPR '99), 23-25 June 1999, Ft. Collins, CO, USA"
ICCV	Dynamic Feature Ordering for Efficient Registration.	Tat-Jen Cham,James M. Rehg	1999	Existing sequential feature-based registration algorithms involving search typically either select features randomly (eg. the RANSAC[8] approach) or assume a predefined, intuitive ordering for the features (e.g. based on size or resolution). This paper presents a formal framework for computing an ordering for features which maximizes search efficiency.Features are ranked according to matching ambiguity measure, and an algorithm is proposed which couples the feature selection with the parameter estimation, resulting in a dynamic feature ordering. The analysis is extended to template features where the matching is non-discrete and a sample-refinement process is proposed.The framework is demonstrated effectively on the localization of a person in an image, using a kinematic model with template features. Different priors are used on the model parameters and the results demonstrate nontrivial variations in the optimal feature hierarchy.
ICCV	Shape Similarity Retrieval under Affine Transform: Application to Multi-View Object Representation and Recognition.	Sadegh Abbasi,Farzin Mokhtarian	1999	Shape Similarity Retrieval under Affine Transform: Application to Multi-View Object Representation and Recognition.
ICCV	Multiway Cut for Stereo and Motion with Slanted Surfaces.	Stan Birchfield,Carlo Tomasi	1999	Multiway Cut for Stereo and Motion with Slanted Surfaces.
ICCV	Real-Time Epipolar Geometry Estimation and Disparity.	Mårten Björkman,Jan-Olof Eklundh	1999	Real-Time Epipolar Geometry Estimation and Disparity.
ICCV	Probabilistic Detection and Tracking of Motion Discontinuities.	Michael J. Black,David J. Fleet	1999	Probabilistic Detection and Tracking of Motion Discontinuities.
ICCV	Neurocalibration: A Neural Network That Can Tell Camera Calibration Parameters.	Moumen T. Ahmed,Elsayed E. Hemayed,Aly A. Farag	1999	Neurocalibration: A Neural Network That Can Tell Camera Calibration Parameters.
ICCV	Recovery and Tracking of Continuous 3D Surfaces from Stereo Data using a Deformable Dual-Mesh.	Yusuf Sinan Akgul,Chandra Kambhamettu	1999	We propose a novel method for continuous 3D depth recovery and tracking using calibrated stereo. The method integrates stereo correspondence, surface reconstruction and tracking by using a new single deformable dual mesh optimization, resulting in simplicity, robustness and efficiency.In order to combine stereo correspondence and structure recovery, the method introduces an external energy function defined for a 3D volume based on cross-correlation between the stereo pairs. The internal energy functional of the deformable dual mesh imposes smoothness on the surfaces and it serves as a communication tool between the two meshes. Under the forces produced by the energy terms, the dual mesh deforms to recover and track the 3D surface.The newly introduced dual-mesh model, which is one of the main contributions of this paper, makes the system robust against local minima and yet it is efficient. A coarse-to-fine minimization approach makes the system even more efficient. Tracking is achieved by using the recovered surface as an initial position for the next time frame. Although the system can effectively utilize initial surface positions and disparity data, they are not needed for a successful operation, which makes this system applicable to a wide range of areas. We present the results of a number of experiments on stereo human face and cloud images, which proves that our new method is very effective.
ICCV	Roxels: Responsibility Weighted 3D Volume Reconstruction.	Jeremy S. De Bonet,Paul A. Viola	1999	Roxels: Responsibility Weighted 3D Volume Reconstruction.
ICCV	Spectral Gradient: A Material Descriptor Invariant to Geometry and Incident Illumination.	Elli Angelopoulou,Sang Wook Lee,Ruzena Bajcsy	1999	The light reflected from a surface depends on the scene geometry, the incident illumination and the surface material. A novel methodology is presented which extracts reflectivity information of the various materials in the scene independent of incident light and scene geometry. A scene is captured under different narrow-band color filters and the spectral derivatives of the scene are computed. The resulting spectral derivatives form a spectral gradient at each pixel. This spectral gradient is a material descriptor which is invariant to scene geometry and incident illumination for smooth diffuse surfaces. Spectral gradients can discriminate among smooth dielectrics with different reflectance properties independent of viewing conditions.
ICCV	Using Model-Driven Bundle-Adjustment to Model Heads from Raw Video Sequences.	Pascal Fua	1999	Using Model-Driven Bundle-Adjustment to Model Heads from Raw Video Sequences.
ICCV	Photometric Surface Analysis in a Tri-Luminal Environment.	Elli Angelopoulou,James P. Williams	1999	Photometric Surface Analysis in a Tri-Luminal Environment.
ICCV	Viewpoint Selection by Navigation through Entropy Maps.	Tal Arbel,Frank P. Ferrie	1999	Viewpoint Selection by Navigation through Entropy Maps.
ICCV	Structure and Motion from Lines under Affine Projections.	Kalle Åström,Anders Heyden,Fredrik Kahl,Magnus Oskarsson	1999	Structure and Motion from Lines under Affine Projections.
ICCV	Statistical Tracking in Video Traffic Surveillance.	Jeffrey E. Boyd,Jean Meloche,Yehuda Vardi	1999	Statistical Tracking in Video Traffic Surveillance.
ICCV	On the Evolution of the Skeleton.	Jonas August,Allen Tannenbaum,Steven W. Zucker	1999	On the Evolution of the Skeleton.
ICCV	Fast Approximate Energy Minimization via Graph Cuts.	Yuri Boykov,Olga Veksler,Ramin Zabih	1999	Many tasks in computer vision involve assigning a label (such as disparity) to every pixel. A common constraint is that the labels should vary smoothly almost everywhere while preserving sharp discontinuities that may exist, e.g., at object boundaries. These tasks are naturally stated in terms of energy minimization. In this paper, we consider a wide class of energies with various smoothness constraints. Global minimization of these energy functions is NP-hard even in the simplest discontinuity-preserving case. Therefore, our focus is on efficient approximation algorithms. We present two algorithms based on graph cuts that efficiently find a local minimum with respect to two types of large moves, namely expansion moves and swap moves. These moves can simultaneously change the labels of arbitrarily large sets of pixels. In contrast, many standard algorithms (including simulated annealing) use small moves where only one pixel changes its label at a time. Our expansion algorithm finds a labeling within a known factor of the global minimum, while our swap algorithm handles more general energy functions. Both of these algorithms allow important cases of discontinuity preserving energies. We experimentally demonstrate the effectiveness of our approach for image restoration, stereo and motion. On real data with ground truth, we achieve 98 percent accuracy.
ICCV	Shadow Puppetry.	Matthew Brand	1999	"The mapping between 3D body poses and 2D shadows is fundamentally many-to-many and defeats regression methods, even with windowed context. We show how to learn a function between paths in the two systems, resolving ambiguities by integrating information over the entire length of a sequence.The basis of this function is a configurable and dynamical manifold that summarizes the target system's behavior. This manifold can be modeled from data with a hidden Markov model having special topological properties that we obtain via entropy minimization. Inference is then a matter of solving for the geodesic on the manifold that best explains the evidence in the cue sequence. We give a closed-form maximum a posteriori solution for geodesics through the learned density space, thereby obtaining optimal paths over the dynamical manifold.These methods give a completely general way to perform inference over time-series; in vision they support analysis, recognition, classification and synthesis of behaviors in linear time. We demonstrate with a prototype that infers 3D from monocular monochromatic sequences (e.g., back-subtractions), without using any articulatory body model. The framework readily accommodates multiple cameras and other sources of evidence such as optical flow or feature tracking."
ICCV	Motion based Decompositing of Video.	Gabriel J. Brostow,Irfan A. Essa	1999	Motion based Decompositing of Video.
ICCV	Optimum Fiducials under Weak Perspective Projection.	Alfred M. Bruckstein,Robert J. Holt,Thomas S. Huang,Arun N. Netravali	1999	We investigate how a given fixed number of points should be located in space so that the pose of a camera viewing them from unknown locations can be estimated with the greatest accuracy. We show that optimum solutions are obtained when the points form concentric complete regular polyhedra. For the case of optimal configurations we provide a worst-case error analysis and use it to analyze the effects of weak perspective approximation to true perspective viewing. Comprehensive computer simulations validate the theoretical results.
ICCV	Projective Alignment with Regions.	Ronen Basri,David W. Jacobs	1999	We have recently proposed an approach to recognition that uses regions to determine the pose of objects while allowing for partial occlusion of the regions. Regions introduce an attractive alternative to existing global and local approaches, since, unlike global features, they can handle occlusion and segmentation errors, and unlike local features they are not as sensitive to sensor errors, and they are easier to match. The region-based approach also uses image information directly, without the construction of intermediate representations, such as algebraic descriptions, which may be difficult to reliably compute. In this paper we further analyze properties of the method for planar objects undergoing projective transformations. In particular, we prove that three visible regions are sufficient to determine the transformation uniquely, and that for a large class of objects two regions are insufficient for this purpose. However, we show that when several regions are available the pose of the object can generally be recovered even when some or all regions are significantly occluded. Our analysis is based on investigating the flow patterns of points under projective transformations in the presence of fixed points.
ICCV	Real-Time Motion Analysis with Linear-Programming.	Moshe Ben-Ezra,Shmuel Peleg,Michael Werman	1999	A method to compute motion models in real time from point-to-line correspondences using linear programming is presented. Point-to-line correspondences are the most reliable motion measurements given the aperture effect, and it is shown how they can approximate other motion measurements as well.Using an L1 error measure for image alignment based on point-to-line correspondences and minimizing this measure using linear programming, achieves results which are more robust than the commonly used L2 metric. While estimators based on L1 are not theoretically robust, experiments show that the proposed method is robust enough to allow accurate motion recovery in hundreds of consecutive frames. The entire computation is performed in real-time on a PC with no special hardware.
ICCV	Multi-View 3D Shape and Motion Recovery on the Spatio-Temporal Curve Manifold.	Rodrigo L. Carceroni,Kiriakos N. Kutulakos	1999	Multi-View 3D Shape and Motion Recovery on the Spatio-Temporal Curve Manifold.
ICCV	Densities and Maximum Likelihood Estimation of Matching Constraints.	Rikard Berthilsson	1999	In this paper we present a theory for obtaining densities that are important for computer vision. As a result of the theory we compute the exact and novel density of the slope of a line fitted to image points. This density makes it possible to obtain confidence intervals for the slope or to make hypothesis testing about if two intersecting lines form a corner or not. The theory also lets us derive a novel technique for maximum likelihood estimation, that can be used for computing the fundamental matrix, conics, or any other constraint that can be expressed by polynomials of degree 2. We present exact and novel densities for the fundamental matrix and conic constraints, that are needed for the estimation. Experiments show how the results can be used in practice to compute maximum likelihood estimates of the fundamental matrix. The ACM Portal is published by the Association for Computing Machinery. Copyright © 2010 ACM, Inc. Terms of Usage Privacy Policy Code of Ethics Contact Us Useful downloads: Adobe Acrobat QuickTime Windows Media Player Real Player
ICCV	Finding Correspondences of Patches by Means of Affine Transformations.	Rikard Berthilsson	1999	In this paper we present a novel method for finding the optimal affine transformation for matching of images. The method requires no feature points, does not rely on normalization of images and can be tuned to highlight interesting parts in the images. Furthermore, the method does not need any derivatives for obtaining the affine transformation and it has a computational cost proportional to n2 logn for n×n images.The problem of finding the optimal affine transformation is solved by an iterative algorithm. In each step a global optimization is performed by the use of FFT. This global characteristic helps the algorithm from getting trapped in a local optimum. Novel theoretical results are presented that show under what restrictions the algorithm can be expected to work properly.Its intended primary use is for reconstruction problems in computer vision. These rely heavily on the establishment of point correspondences in the images. Since the method makes no assumptions on the images it can be used when feature points are difficult to detect. Experiments on real images are included and it is shown that the algorithm is robust and performs well even in difficult situations, with occlusions. The ACM Portal is published by the Association for Computing Machinery. Copyright © 2010 ACM, Inc. Terms of Usage Privacy Policy Code of Ethics Contact Us Useful downloads: Adobe Acrobat QuickTime Windows Media Player Real Player
ICCV	Reconstruction of Curves in R, using Factorization and Bundle Adjustment.	Rikard Berthilsson,Kalle Åström,Anders Heyden	1999	Reconstruction of Curves in R, using Factorization and Bundle Adjustment.
ICCV	Unsupervised Image Classification with a Hierarchical EM Algorithm.	Annabelle Chardin,Patrick Pérez	1999	This work takes place in the context of hierarchical stochastic models for the resolution of discrete inverse problems from low level vision. Some of these models lie on the nodes of a quad-tree which leads to non-iterative inference procedures. Nevertheless, if they circumvent the algorithmic drawbacks of grid-based models (computational load and/or great dependence on the initialization), they admit modeling shortcomings (cumbersome and somehow artificial). We investigate a new hierarchical stochastic model which takes benefit from both the spatial and the hierarchical prior modelings. The independence graph is based on a tree which has been pollarded with the nodes at the coarsest resolution exhibiting a grid-based interaction structure. For this class of models, we address the critical problem of parameter estimation. To this end, we derive an EM algorithm on the hybrid structure which mixes an exact EM algorithm on each subtrees and a low cost Gibbsian EM algorithm on the coarse spatial grid. Experiments on a synthetic image and on multi-spectral satellite images are reported.
ICCV	New Calibration-free Approach for Augmented Reality based on Parameterized Cuboid Structure.	Chu-Song Chen,Chi-Kuo Yu,Yi-Ping Hung	1999	New Calibration-free Approach for Augmented Reality based on Parameterized Cuboid Structure.
ICCV	A Bidirectional Matching Algorithm for Deformable Pattern Detection with Application to Handwritten Word Retrieval.	Kwok-Wai Cheung,Dit-Yan Yeung,Roland T. Chin	1999	A Bayesian framework for deformable pattern classification has been proposed in [1] with promising results for isolated handwritten character recognition. Its performance, however, degrades significantly when it is applied to detect deformable patterns in complex scenes, where the amount of outliers due to other neighboring objects or the background is usually large. Also, the fact that the associated evidence measure does not penalize models resting on white space results in a high false alarm rate.In this paper, another Bayesian framework for deformable pattern detection is proposed. The framework possesses the intrinsic property of matching with only part of an image (segmentation) and its associated evidence measure can penalize white space implicitly. However, limited data exploration capability is the major trade-off. By properly combining the two frameworks, a new matching algorithm called bi-directional matching is proposed.This combined approach possesses the advantages of the two frameworks and gives robust results for non-rigid shape extraction. To evaluate the performance of the proposed approach, we have applied it to shape-based handwritten word retrieval. Using a subset of the bb data-set in the CEDAR database, we can achieve a recall rate of 59% and a precision rate of 43%.
ICCV	Shape from Texture through Deformations.	Maureen Clerc,Stéphane Mallat	1999	Shape from Texture through Deformations.
ICCV	"The Earth Mover's Distance under Transformation Sets."	Scott D. Cohen,Leonidas J. Guibas	1999	"The Earth Mover's Distance (EMD) is a distance measure between distributions with applications in image retrieval and matching. We consider the problem of computing a transformation of one distribution which minimizes its EMD to another.The applications discussed here include estimation of the size at which a color pattern occurs in an image, lighting-invariant object recognition, and point feature matching in stereo image pairs. We present a monotonically convergent iteration which can be applied to a large class of EMD under transformation problems, although the iteration may converge to only a locally optimal transformation. We also provide algorithms that are guaranteed to compute a globally optimal transformation for a few specific problems, including some EMD under translation problems."
ICCV	Generalized Bounds for Time to Collision from First-Order Image Motion.	Carlo Colombo,Alberto Del Bimbo	1999	Generalized Bounds for Time to Collision from First-Order Image Motion.
ICCV	Mean Shift Analysis and Applications.	Dorin Comaniciu,Peter Meer	1999	A non-parametric estimator of density gradient, the mean shift, is employed in the joint, spatial-range (value) domain of gray level and color images for discontinuity preserving filtering and image segmentation. Properties of the mean shift are reviewed and its convergence on lattices is proven.The proposed filtering method associates with each pixel in the image the closest local mode in the density distribution of the joint domain. Segmentation into a piecewise constant structure requires only one more step, fusion of the regions associated with nearby modes.The proposed technique has two parameters controlling the resolution in the spatial and range domains. Since convergence is guaranteed, the technique does not require the intervention of the user to stop the filtering at the desired image quality. Several examples, for gray and color images, show the versatility of the method and compare favorably with results described in the literature for the same images.
ICCV	Resolution Invariant Surfaces for Panoramic Vision Systems.	Tanya L. Conroy,John B. Moore	1999	Resolution Invariant Surfaces for Panoramic Vision Systems.
ICCV	Manhattan World: Compass Direction from a Single Image by Bayesian Inference.	James M. Coughlan,Alan L. Yuille	1999	When designing computer vision systems for the blind and visually impaired it is important to determine the orientation of the user relative to the scene. We observe that most indoor and outdoor (city) scenes are designed on a Manhattan three-dimensional grid. This Manhattan grid structure puts strong constraints on the intensity gradients in the image. We demonstrate an algorithm for detecting the orientation of the user in such scenes based on Bayesian inference using statistics which we have learnt in this domain. Our algorithm requires a single input image and does not involve pre-processing stages such as edge detection and Hough grouping. We demonstrate strong experimental results on a range of indoor and outdoor images. We also show that estimating the grid structure makes it significantly easier to detect target objects which are not aligned with the grid.
ICCV	Single View Metrology.	Antonio Criminisi,Ian D. Reid,Andrew Zisserman	1999	"We describe how 3D affine measurements may be computed from a single perspective view of a scene given only minimal geometric information determined from the image. This minimal information is typically the vanishing line of a reference plane, and a vanishing point for a direction not parallel to the plane. It is shown that affine scene structure may then be determined from the image, without knowledge of the camera's internal calibration (e.g. focal length), nor of the explicit relation between camera and world (pose).In particular, we show how to (i) compute the distance between planes parallel to the reference plane (up to a common scale factor)&semi; (ii) compute area and length ratios on any plane parallel to the reference plane&semi; (iii) determine the camera's location. Simple geometric derivations are given for these results. We also develop an algebraic representation which unifies the three types of measurement and, amongst other advantages, permits a first order error propagation analysis to be performed, associating an uncertainty with each measurement.We demonstrate the technique for a variety of applications, including height measurements in forensic images and 3D graphical modelling from single images."
ICCV	Parallax Geometry of Smooth Surfaces in Multiple Views.	Geoffrey Cross,Andrew W. Fitzgibbon,Andrew Zisserman	1999	Parallax Geometry of Smooth Surfaces in Multiple Views.
ICCV	Direct Identification of Moving Objects and Background from 2D Motion Models.	Gabriela Csurka,Patrick Bouthemy	1999	Direct Identification of Moving Objects and Background from 2D Motion Models.
ICCV	Correlation Model for 3D Texture.	Kristin J. Dana,Shree K. Nayar	1999	While an exact definition of texture is somewhat elusive, texture can be qualitatively described as a distribution of color, albedo or local normal on a surface. In the literature, the word texture is often used to describe a color or albedo variation on a smooth surface. We refer to such texture as 2D texture.In real world scenes, texture is often due to surface height variations and can be termed 3D texture. Because of local foreshortening and masking, oblique views of 3D texture are not simple transformations of the frontal view. Consequently, texture representations such as the correlation function or power spectrum are also affected by local foreshortening and masking.This work presents a correlation model for a particular class of 3D textures. The model characterizes the spatial relationship among neighboring pixels in an image of 3D texture and the change of this spatial relationship with viewing direction.
ICCV	3D Articulated Models and Multi-View Tracking with Silhouettes.	Quentin Delamarre,Olivier D. Faugeras	1999	We propose a method to estimate the motion of a person filmed by two or more fixed cameras. The novelty of our technique is its ability to cope with fast movements, self-occlusions and noisy images. Our algorithms are based on the latest works on calibration and image segmentation developed in our lab. We compare the projections of a 3D model of a person on the images to the detected silhouettes of the person, and create forces that will move the 3D model towards the final estimation of the real pose. We developed a fast algorithm that computes the motion of the articulated 3D model. We show that our results are good, even if the cameras are not synchronized.
ICCV	Tracking through Singularities and Discontinuities by Random Sampling.	Jonathan Deutscher,Ben North,Benedicte Bascle,Andrew Blake	1999	Some issues in marker-less tracking of human body motion are addressed. Extended Kalman filters have commonly been applied to kinematic variables, to combine predictions consistent with plausible motion, with the incoming stream of visual measurements. Kalman filtering is applicable only when the underlying distribution is approximately Gaussian. Often, this assumption proves remarkably robust.There are two pervasive circumstances under which the Gaussianity assumption can break down. The first is kinematic singularity, and the second is at joint end-stops. Failure of Kalman filtering under these circumstance is illustrated.The non-Gaussian nature of the distributions is demonstrated experimentally by means of Monte-Carlo simulation. Random simulation - particle filtering or Condensation -proves to provide a robust alternative algorithm for tracking that can also deal with these difficult conditions.
ICCV	Learning-based Object Detection in Cardiac MR Images.	Nicolae Duta,Anil K. Jain,Marie-Pierre Dubuisson-Jolly	1999	An automated method for left ventricle detection in MR cardiac images is presented. Ventricle detection is the first step in a fully automated segmentation system used to compute volumetric information about the heart.Our method is based on learning the gray level appearance of the ventricle by maximizing the discrimination between positive and negative examples in a training set. The main differences from previously reported methods are feature definition and solution to the optimization problem involved in the learning process.Our method was trained on a set of 1,350 MR cardiac images from which 101,250 positive examples and 123,096 negative examples were generated. The detection results on a test set of 887 different images demonstrate an excellent performance: 98% detection rate, a false alarm rate of 0:05% of the number of windows analyzed (10 false alarms per image) and a detection time of 2 seconds per 256 × 256 image on a Sun Ultra 10 for an 8-scale search. The false alarms are eventually eliminated by a position/scale consistency check along all the images that represent the same anatomical slice.
ICCV	Advances in Active Appearance Models.	Gareth J. Edwards,Timothy F. Cootes,Christopher J. Taylor	1999	Advances in Active Appearance Models.
ICCV	Texture Synthesis by Non-parametric Sampling.	Alexei A. Efros,Thomas K. Leung	1999	A non-parametric method for texture synthesis is proposed. The texture synthesis process grows a new image outward from an initial seed, one pixel at a time. A Markov random field model is assumed, and the conditional distribution of a pixel given all its neighbors synthesized so far is estimated by querying the sample image and finding all similar neighborhoods. The degree of randomness is controlled by a single perceptually intuitive parameter. The method aims at preserving as much local structure as possible and produces good results for a wide variety of synthetic and real-world textures.
ICCV	Optimal Filters for Gradient-based Motion Estimation.	Michael Elad,Patrick C. Teo,Yacov Hel-Or	1999	Optimal Filters for Gradient-based Motion Estimation.
ICCV	Estimation of Structure and Motion Parameters for a Roaming Robot that Scans the Space.	Minoru Etoh,Toshimichi Aoki,Koichi Hata	1999	Estimation of Structure and Motion Parameters for a Roaming Robot that Scans the Space.
ICCV	Colour by Correlation: A Simple, Unifying Approach to Colour Constancy.	Graham D. Finlayson,Steven D. Hordley,Paul M. Hubel	1999	In this paper we consider the problem of color constancy; how given an image of a scene under an unknown illuminant can we recover an estimate of that light? Rather than recovering a single estimate of the illuminant as many previous authors have done, in the first instance we recover a measure of the likelihood that each possible illuminant was the scene illuminant. We do this by correlating image colors with the colors that can occur under each of a set of possible lights. We then recover an estimate of the scene illuminant based on these likelihoods. Computation is expressed and performed in a generic correlation framework which we develop in this paper. We develop a new probabilistic instantiation of this framework which delivers very good color constancy on synthetic and real images. We show that the proposed framework is rich enough to allow many existing algorithms to be expressed within it; e.g. the grey-world and gamut mapping algorithms. We explore too the relationship of these algorithms to other probabilistic and neural network approaches.
ICCV	Bayesian Structure from Motion.	David A. Forsyth,Sergey Ioffe,John A. Haddon	1999	Bayesian Structure from Motion.
ICCV	A Subset Approach to Contour Tracking in Clutter.	Daniel Freedman,Michael S. Brandstein	1999	A Subset Approach to Contour Tracking in Clutter.
ICCV	Learning Low-Level Vision.	William T. Freeman,Egon C. Pasztor	1999	"We show a learning-based method for low-level vision problems{estimating scenes from images. We generate a synthetic world of scenes and their corresponding rendered images. We model that world with a Markov network, learning the network parameters from the examples. Bayesian belief propagation allows us to efficiently find a local maximum of the posterior probability for the scene, given the image. We call this approach VISTA{Vision by Image/Scene TrAining.We apply VISTA to the ""super-resolution"" problem (estimating high frequency details from a low-resolution image), showing good results. For the motion estimation problem, we show figure/ground discrimination, solution of the aperture problem, and filling-in arising from application of the same probabilistic machinery."
ICCV	Transformed Component Analysis: Joint Estimation of Spatial Transformations and Image Components.	Brendan J. Frey,Nebojsa Jojic	1999	"A simple, effective way to model images is to represent each input pattern by a linear combination of ""component"" vectors, where the amplitudes of the vectors are modulated to match the input. This approach includes principal component analysis, independent component analysis and factor analysis. In practice, images are subjected to randomly selected transformations of a known nature, such as translation and rotation. Direct use of the above methods will lead to severely blurred components that tend to ignore the more interesting and useful structure. In previous work, we introduced a clustering algorithm that is invariant to transformations [1].In this paper, we propose a method called transformed component analysis, which incorporates a discrete, hidden variable that accounts for transformations and uses the expectation maximization algorithm to jointly extract components and normalize for transformations. We illustrate the algorithm using a shading problem, facial expression modeling and written digit recognition."
ICCV	Accurate Motion Flow Estimation with Discontinuities.	Lionel Gaucher,Gérard G. Medioni	1999	We address the problem of motion flow estimation for a scene with multiple moving objects, observed from a possibly moving camera. We take as input a (possibly sparse) noisy velocity field, as obtained from local matching, produce a set of motion boundaries, and identify pixels with different velocities in overlapping layers. For a fixed observer, these overlapping layers capture occlusion information. For a moving observer, further processing is required to segment independent objects and infer structure. Unlike previous approaches, which generate layers by iteratively fitting data to a set of predefined parameters, we instead find boundaries first, then infer regions and address occlusion overlap relationships. All computational steps use a common framework of tensors to represent velocity information, together with saliency (confidence), and uncertainty. Communication between sites is performed by convolution-like tensor voting. The scheme is non-iterative, and the only free parameter is the scale, related to neighborhood size. We illustrate the approach with results obtained from synthetic sequences and from real images. The quantitative results compare favorably with those of other methods, especially in the presence of occlusion
ICCV	"Real-Time Object Detection for ""Smart"" Vehicles."	Dariu Gavrila,Vasanth Philomin	1999	"Real-Time Object Detection for ""Smart"" Vehicles."
ICCV	Modeling Bayesian Estimation for Deformable Contours.	Stan Z. Li,Juwei Lu	1999	"A novel trainable snake model, called EigenSnake, is presented in the Bayesian framework. In the EigenSnake, prior knowledge of a specific object shape, such as that of face outlines and facial features, is derived from a training set of the shape and incorporated into a Bayesian snake model in the form of the prior distribution. Further, a ""shape space"", which is constructed on the basis of a set of eigenvectors obtained from principle component analysis, is used to restrict and stabilize the search for the optimal solution. The effectiveness is demonstrated by experiments, which shows that the EigenSnake produces more reliable and accurate results than existing models."
ICCV	Catadioptric Camera Calibration.	Christopher Geyer,Konstantinos Daniilidis	1999	Catadioptric Camera Calibration.
ICCV	On the Local Form and Transitions of Symmetry Sets, Medial Axes, and Shocks.	Peter J. Giblin,Benjamin B. Kimia	1999	In this paper we explore the local geometry of the medial axis (MA) and shocks (SH), and their structural changes under deformations, by viewing these symmetries as subsets of the symmetry set (SS) and present two results. First, we establish that the local form of the medial axis must generically be one of three cases, which we denote by the A notation explained below (here, it merely serves as a reference to sections of the paper): endpoints (A3), interior points (A12), and junctions (A13). The local form of shocks is a sub-classification of these points into six types. Second, we address the (classical) instabilities of the MA, i.e., abrupt changes in the representation with a slight changes in shape, as when a new branch appears with slight protrusion. The identification of these &lsquo;transitions&rsquo; is clearly crucial in robust object recognition. We show that for the medial axis only two such instabilities are generically possible: (i) when four branches come together (A14), and (ii) when a new branch grows out of an existing one (A1A3). Similarly, there are six cases of shock instabilities, derived as sub-classifications of the MA instabilities. We give an explicit example of a dent forming in an ellipse where many of the transitions described in the paper can be seen to appear.
ICCV	Registration of Multiple Point Sets using the EM Algorithm.	Jacob Goldberger	1999	In this paper we address the problem of global registration between multiple dimensional point patterns with a given correspondence. The actual overlapping is not necessarily between pairs. Instead, it can be between any number of patterns. It is assumed that each pattern is a portion of an image of an unobserved object under a distinct rigid transformation. We derive an iterative solution for the problem of global registration of the patterns in order to reconstruct the original object.Our solution is based on the EM algorithm and it generalizes the well known solutions for the two-pattern case. We also suggest a very efficient method to implement the proposed algorithm. Experimental results demonstrate the improved performance of the proposed method.
ICCV	Recognition of Temporal Structures: Learning Prior and Propagating Observation Augmented Densities via Hidden Markov States.	Shaogang Gong,Michael Walter,Alexandra Psarrou	1999	Recognition of Temporal Structures: Learning Prior and Propagating Observation Augmented Densities via Hidden Markov States.
ICCV	Curve Finder Combining Perceptual Grouping and a Kalman like Fitting.	Frederic Guichard,Jean-Philippe Tarel	1999	Curve Finder Combining Perceptual Grouping and a Kalman like Fitting.
ICCV	Backpack: Detection of People Carrying Objects using Silhouettes.	Ismail Haritaoglu,Ross Cutler,David Harwood,Larry S. Davis	1999	Backpack: Detection of People Carrying Objects using Silhouettes.
ICCV	Camera Calibration and the Search for Infinity.	Richard I. Hartley,Lourdes de Agapito,Ian D. Reid,Eric Hayman	1999	Camera Calibration and the Search for Infinity.
ICCV	3D Pose Tracking with Linear Depth and Brightness Constraints.	Michael Harville,Ali Rahimi,Trevor Darrell,Gaile G. Gordon,John Woodfill	1999	3D Pose Tracking with Linear Depth and Brightness Constraints.
ICCV	Zoom-Invariant Tracking using Points and Lines in Affine Views - An Application of the Affine Multifocal Tensors.	Eric Hayman,Torfi Thórhallsson,David W. Murray	1999	Zoom-Invariant Tracking using Points and Lines in Affine Views - An Application of the Affine Multifocal Tensors.
ICCV	Flexible Calibration: Minimal Cases for Auto-Calibration.	Anders Heyden,Kalle Åström	1999	Flexible Calibration: Minimal Cases for Auto-Calibration.
ICCV	Representation Issues in the ML Estimation of Camera Motion.	Joachim Hornegger,Carlo Tomasi	1999	Representation Issues in the ML Estimation of Camera Motion.
ICCV	Sensitivity Analysis for Object Recognition from Large Structural Libraries.	Benoit Huet,Edwin R. Hancock	1999	This paper studies the structural sensitivity of line-pattern recognition using shape-graphs. We compare the recognition performance for four different algorithms. Each algorithm uses a set of pair wise geometric attributes and a neighborhood graph to represent the structure of the line patterns. The first algorithm uses a pair-wise geometric histogram, the second uses a relational histogram on the edges of the shape graph, the third compares the set of attributes on the edges of the shape graph and the final algorithm compares the arrangement of line correspondences using graph-matching. The different algorithms are compared under line deletion, line addition, line fragmentation and line end-point measurement errors. It is the graph-matching algorithm which proves to be the most effective.
ICCV	Affine Reconstruction from Monocular Vision in the Presence of a Symmetry Plane.	D. Q. Huynh	1999	Affine Reconstruction from Monocular Vision in the Presence of a Symmetry Plane.
ICCV	Motion Segmentation based on Factorization Method and Discriminant Criterion.	Naoyuki Ichimura	1999	Motion Segmentation based on Factorization Method and Discriminant Criterion.
ICCV	Finding People by Sampling.	Sergey Ioffe,David A. Forsyth	1999	We show how to use a sampling method to find sparsely clad people in static images. People are modeled as an assembly of nine cylindrical segments. Segments are found using an EM algorithm, and then assembled into hypotheses incrementally, using a learned likelihood model. Each assembly step passes on a set of samples of its likelihood to the next; this yields effective pruning of the space of hypotheses. The collection of available nine-segment hypotheses is then represented by a set of equivalence classes, which yield an efficient pruning process. The posterior for the number of people is obtained from the class representatives. People are counted quite accurately in images of real scenes using an MAP estimate. We show the method allows top-down as well as bottom up reasoning. While the method can be overwhelmed by very large numbers of segments, we show that this problem can be avoided by quite simple pruning steps.
ICCV	Multi-Frame Optical Flow Estimation using Subspace Constraints.	Michal Irani	1999	Multi-Frame Optical Flow Estimation using Subspace Constraints.
ICCV	Recognition of Multi-Agent Interaction in Video Surveillance.	Yuri A. Ivanov,Aaron F. Bobick	1999	Recognition of Multi-Agent Interaction in Video Surveillance.
ICCV	Posture Estimation using Structure and Motion Models.	Yoshio Iwai,Kenichi Ogaki,Masahiko Yachida	1999	Posture Estimation using Structure and Motion Models.
ICCV	Qualitative Probabilities for Image Interpretation.	Allan D. Jepson,Richard Mann	1999	Two basic problems in image interpretation are: a) determining which interpretations are the most plausible amongst many possibilities; and b) controlling the search for plausible interpretations. We address these issues using a Bayesian approach, with the plausibility ordering and search pruning based on the posterior probabilities of interpretations. However, due to the need for detailed quantitative prior probabilities and the need to evaluate complex integrals over various conditional distributions, a full Bayesian approach is currently impractical except in tightly constrained domains.To circumvent these difficulties we introduce the notion of qualitative probabilistic analysis. In particular, given spatial and contrast resolution parameters, we consider only the asymptotic order of the posterior probability for any interpretation as these resolutions are made finer. We introduce this approach for a simple card-world domain, and present computational results for blocks-world images.
ICCV	Globally Optimal Regions and Boundaries.	Ian Jermyn,Hiroshi Ishikawa	1999	We propose a new form of energy functional for the segmentation of regions in images, and an efficient method for finding its global optima. The energy can have contributions from both the region and its boundary, thus combining the best features of region- and boundary-based approaches to segmentation. By transforming the region energy into a boundary energy, we can treat both contributions on an equal footing, and solve the global optimization problem as a minimum mean weight cycle problem on a directed graph. The simple, polynomial-time algorithm requires no initialization and is highly parallelizable
ICCV	View Synthesis and 3D Reconstruction of Piecewise Planar Scenes Using Intersection Lines between the Planes.	Björn Johansson	1999	View Synthesis and 3D Reconstruction of Piecewise Planar Scenes Using Intersection Lines between the Planes.
ICCV	Tracking Self-Occluding Articulated Objects in Dense Disparity Maps.	Nebojsa Jojic,Matthew Turk,Thomas S. Huang	1999	Tracking Self-Occluding Articulated Objects in Dense Disparity Maps.
ICCV	Critical Motions and Ambiguous Euclidean Reconstructions in Auto-Calibration.	Fredrik Kahl	1999	Critical Motions and Ambiguous Euclidean Reconstructions in Auto-Calibration.
ICCV	Accuracy Bounds and Optimal Computation of Homography for Image Mosaicing Applications.	Ken-ichi Kanatani,Naoya Ohta	1999	Accuracy Bounds and Optimal Computation of Homography for Image Mosaicing Applications.
ICCV	United Snakes.	Jianming Liang,Tim McInerney,Demetri Terzopoulos	1999	"Since their debut in 1987, snakes (active contour models) have become a standard image analysis technique with several variants now in common use. We present a portable, reusable, software package called ""United Snakes"". The package unites the most popular snake variants, including finite difference, B-spline, and Hermite polynomial snakes within the mathematical framework of a general finite element formulation with a choice of shape functions. The package furthermore incorporates a recently proposed snake-like technique known as ""livewire"". We integrate snakes and livewire by introducing an effective method for imposing hard constraints on snakes. Our experiments demonstrate that snakes and livewire have complementary strengths and that their union offers a more powerful tool for interactive image analysis, especially for medical imaging applications. United Snakes is implemented in Java as a JavaBean so that it can easily be integrated in end-user application systems."
ICCV	Combining Scene and Auto-Calibration Constraints.	David Liebowitz,Andrew Zisserman	1999	Combining Scene and Auto-Calibration Constraints.
ICCV	Robust Sequence Proximity Estimation by Radial Distance Hashing.	Michael Kertesz,Yehezkel Yeshurun	1999	Robust Sequence Proximity Estimation by Radial Distance Hashing.
ICCV	A Level Line Selection Approach for Object Boundary Estimation.	Charles Kervrann,Mark Hoebeke,Alain Trubuil	1999	"An energy model-based approach for estimating object boundaries is presented. We study a particular energy, which minimizer can be determined. The method estimates the unknown number of objects and draws object boundaries by selecting the ""best"" level lines computed from level sets of the original image. Unlike previous standard methods, the proposed method does not require iteration for minimizing the energy. In addition, our segmentation algorithm combines anisotropic diffusion-based regularization with level line selection to extract smooth object boundaries. Experimental results on 2D biomedical and meteorological images are reported."
ICCV	Tracking Articulated Objects in Real-Time Range Image Sequences.	Michael H. Lin	1999	Tracking Articulated Objects in Real-Time Range Image Sequences.
ICCV	A Representation of Specular Appearance.	Stephen Lin,Sang Wook Lee	1999	The appearance of an object can vary considerably with changes in illumination conditions. Methods have been developed to describe these differences for diffuse reflection using the Lambertian model, but little work has been done in characterizing specular appearance.Towards a more comprehensive global reflectance descriptor, this paper focuses on a representation of specular appearance based on an approximate specular reflection model derived from Torrance-Sparrow. We propose that under certain illumination and surface conditions local specular structure can be expressed by the logarithms of three intensity-normalized photometric images. The total number of photometric images needed for representing global specular appearance depends on the object surface roughness, and we suggest an illumination planning method for determining the number of images. Experimental results demonstrate the effectiveness of this logarithmic model as a specular descriptor.
ICCV	Estimation of Diffuse and Specular Appearance.	Stephen Lin,Sang Wook Lee	1999	To account for the variability of object appearance due to differences in illumination, attention has recently been focused on representing the set of images for all possible lighting conditions. Approaches that address this problem have primarily focused on lighting differences for diffuse reflection using the Lambertian model; however, specular reflections can additionally present considerable disparity in appearance.We present a method for representing illumination appearance for both diffuse and specular reflections for objects of uniform surface roughness using four photometric images. This approach uses separation of reflection components, extracts surface reflectances and roughness, and produces arbitrary lighting images without explicit computation of surface shape. Experimental results demonstrate the validity of the proposed method for constructing diffuse and specular appearances.
ICCV	Approximate Tree Matching and Shape Similarity.	Tyng-Luh Liu,Davi Geiger	1999	Approximate Tree Matching and Shape Similarity.
ICCV	Calibration of Hand-Held Camera Sequences for Plenoptic Modeling.	Reinhard Koch,Marc Pollefeys,Benno Heigl,Luc J. Van Gool,Heinrich Niemann	1999	Calibration of Hand-Held Camera Sequences for Plenoptic Modeling.
ICCV	Correspondenceless Motion Estimation from Range Images.	Yonghuai Liu,Marcos A. Rodrigues	1999	Correspondenceless Motion Estimation from Range Images.
ICCV	Object Recognition from Local Scale-Invariant Features.	David G. Lowe	1999	An object recognition system has been developed that uses a new class of local image features. The features are invariant to image scaling, translation, and rotation, and partially invariant to illumination changes and affine or 3D projection.These features share similar properties with neurons in inferior temporal cortex that are used for object recognition in primate vision. Features are efficiently detected through a staged filtering approach that identifies stable points in scale space. Image keys are created that allow for local geometric deformations by representing blurred image gradients in multiple orientation planes and at multiple scales.The keys are used as input to a nearest-neighbor indexing method that identifies candidate object matches. Final verification of each match is achieved by finding a low-residual least-squares solution for the unknown model parameters. Experimental results show that robust object recognition can be achieved in cluttered partially-occluded images with a computation time of under 2 seconds.
ICCV	Specularities on Surfaces with Tangential Hairs or Grooves.	Rong Lu,Jan J. Koenderink,Astrid M. L. Kappers	1999	Specularities on Surfaces with Tangential Hairs or Grooves.
ICCV	Euclidean Reconstruction and Reprojection up to Subgroups.	Yi Ma,Stefano Soatto,Jana Kosecka,Shankar Sastry	1999	The necessary and sufficient conditions for being able to estimate scene structure, motion and camera calibration from a sequence of images are very rarely satisfied in practice. What exactly can be estimated in sequences of practical importance, when such conditions are not satisfied? In this paper we give a complete answer to this question. For every camera motion that fails to meet the conditions, we give explicit formulas for the ambiguities in the reconstructed scene, motion and calibration. Such a characterization is crucial both for designing robust estimation algorithms (that do not try to recover parameters that cannot be recovered), and for generating novel views of the scene by controlling the vantage point. To this end, we characterize explicitly all the vantage points that give rise to a valid Euclidean reprojection regardless of the ambiguity in the reconstruction. We also characterize vantage points that generate views that are altogether invariant to the ambiguity. All the results are presented using simple notation that involves no tensors nor complex projective geometry, and should be accessible with basic background in linear algebra.
ICCV	A Theory of Shape by Space Carving.	Kiriakos N. Kutulakos,Steven M. Seitz	1999	In this paper we consider the problem of computing the 3D shape of an unknown, arbitrarily-shaped scene from multiple photographs taken at known but arbitrarily-distributed viewpoints. By studying the equivalence class of all 3D shapes that reproduce the input photographs, we prove the existence of a special member of this class, the photo hull, that (1) can be computed directly from photographs of the scene, and (2) subsumes all other members of this class. We then give a provably-correct algorithm, called Space Carving, for computing this shape and present experimental results on complex real-world scenes. The approach is designed to (1) capture photorealistic shapes that accurately model scene appearance from a wide range of viewpoints, and (2) account for the complex interactions between occlusion, parallax, shading, and their view-dependent effects on scene-appearance.
ICCV	A Probabilistic Exclusion Principle for Tracking Multiple Objects.	John MacCormick,Andrew Blake	1999	Tracking multiple targets is a challenging problem, especially when the targets are &ldquo;identical&rdquo;, in the sense that the same model is used to describe each target. In this case, simply instantiating several independent 1-body trackers is not an adequate solution, because the independent trackers tend to coalesce onto the best-fitting target. This paper presents an observation density for tracking which solves this problem by exhibiting a probabilistic exclusion principle. Exclusion arises naturally from a systematic derivation of the observation density, without relying on heuristics. Another important contribution of the paper is the presentation of partitioned sampling, a new sampling method for multiple object tracking. Partitioned sampling avoids the high computational load associated with fully coupled trackers, while retaining the desirable properties of coupling.
ICCV	Removal of Translation Bias when using Subspace Methods.	W. James MacLean	1999	"Given estimates of the motion field (optic flow) from an image sequence, it is possible to recover translational direction, ~ T , using a variety of techniques. One such technique, known as ""subspace methods,"" generates constraints which are perpendicular to ~ T , so that two distinct constraints allow a solution for ~ T . In practice many constraints are used in a least-squares solution, but it has been observed that the recovered estimates for ~ T are biased towards the optical axis. While the cause of the bias is well known, previous attempts to remove it have been awed. This paper outlines a new method which removes the bias. The technique is simple to apply and computationally efficient."
ICCV	Modeling the Process of Aging in Face Images.	Andreas Lanitis,Christopher J. Taylor,Timothy F. Cootes	1999	Modeling the Process of Aging in Face Images.
ICCV	Efficient Recovery of Low-Dimensional Structure from High-Dimensional Data.	Shyjan Mahamud,Martial Hebert	1999	Efficient Recovery of Low-Dimensional Structure from High-Dimensional Data.
ICCV	Segmentation of Salient Closed Contours from Real Images.	Shyjan Mahamud,Karvel K. Thornber,Lance R. Williams	1999	Using a saliency measure based on the global property of contour closure, we have developed a method that reliably segments out salient contours bounding unknown objects from real edge images. The measure also incorporates the Gestalt principles of proximity and smooth continuity that previous methods have exploited. Unlike previous measures, we incorporate contour closure by finding the eigen-solution associated with a stochastic process that models the distribution of contours passing through edges in the scene. The segmentation algorithm utilizes the saliency measure to identify multiple closed contours by finding strongly- connected components on an induced graph. The determination of strongly connected components is a direct consequence of the property of closure. We report for the first time, results on large real images for which segmentation takes an average of about 10 secs per object on a general-purpose workstation. The segmentation is made efficient for such large images by exploiting the inherent symmetry in the task.
ICCV	The Visual Hull of Curved Objects.	Aldo Laurentini	1999	The Visual Hull of Curved Objects.
ICCV	Textons, Contours and Regions: Cue Integration in Image Segmentation.	Jitendra Malik,Serge Belongie,Jianbo Shi,Thomas K. Leung	1999	This paper makes two contributions. It provides (1) an operational definition of textons, the putative elementary units of texture perception, and (2) an algorithm for partitioning the image into disjoint regions of coherent bright-ness and texture, where boundaries of regions are defined by peaks in contour orientation energy and differences in texton densities across the contour.Julesz introduced the term texton, analogous to a phoneme in speech recognition, but did not provide an operational definition for gray-level images. Here we re-invent textons as frequently co-occurring combinations of oriented linear filter outputs. These can be learned using a K-means approach. By mapping each pixel to its nearest texton, the image can be analyzed into texton channels, each of which is a point set where discrete techniques such as Voronoi diagrams become applicable.Local histograms of texton frequencies can be used with a X2 test for significant differences to find texture boundaries. Natural images contain both textured and untextured regions, so we combine this cue with that of the presence of peaks of contour energy derived from outputs of odd- and even-symmetric oriented Gaussian derivative filters. Each of these cues has a domain of applicability, so to facilitate cue combination we introduce a gating operator based on a statistical test for isotropy of Delaunay neighbors. Having obtained a local measure of how likely two nearby pixels are to belong to the same region, we use the spectral graph theoretic framework of normalized cuts to find partitions of the image into regions of coherent texture and brightness. Experimental results on a wide range of images are shown.
ICCV	Correlation-based Estimation of Ego-Motion and Structure from Motion and Stereo.	Robert Mandelbaum,Garbis Salgian,Harpreet S. Sawhney	1999	Correlation-based Estimation of Ego-Motion and Structure from Motion and Stereo.
ICCV	Bayesian Fusion of Color and Texture Segmentations.	Roberto Manduchi	1999	"In many applications one would like to use information from both color and texture features in order to segment an image. We propose a novel technique to combine ""soft"" segmentations computed for two or more features independently. Our algorithm merges models according to a maximum descriptiveness criterion, and allows to choose any number of classes for the final grouping. This technique also allows to improve the quality of supervised classification based on one feature (e.g. color) by merging information from unsupervised segmentation based on another feature (e.g., texture.)"
ICCV	Independent Component Analysis of Textures.	Roberto Manduchi,Javier Portilla	1999	A common method for texture representation is to use the marginal probability densities over the outputs of a set of multi-orientation, multi-scale filters as a description of the texture. We propose a technique, based on Independent Components Analysis, for choosing the set of filters that yield the most informative marginals, meaning that the product over the marginals most closely approximates the joint probability density function of the filter outputs. The algorithm is implemented using a steerable filter space. Experiments involving both texture classification and synthesis show that compared to Principal Components Analysis, ICA provides superior performance for modeling of natural and synthetic textures.
ICCV	Medial Faces from a Concise 3D Thinning Algorithm.	Antoine Manzanera,Thierry M. Bernard,Françoise J. Prêteux,Bernard Longuet	1999	Medial Faces from a Concise 3D Thinning Algorithm.
ICCV	Robust Real-Time Visual Tracking using a 2D-3D Model-based Approach.	Éric Marchand,Patrick Bouthemy,François Chaumette,Valérie Moreau	1999	Robust Real-Time Visual Tracking using a 2D-3D Model-based Approach.
ICCV	Recognizing Surfaces using Three-Dimensional Textons.	Thomas K. Leung,Jitendra Malik	1999	We study the recognition of surfaces made from different materials such as concrete, rug, marble or leather on the basis of their textural appearance. Such natural textures arise from spatial variation of two surface attributes: (1) reflectance and (2) surface normal. In this paper, we provide a unified model to address both these aspects of natural texture. The main idea is to construct a vocabulary of prototype tiny surface patches with associated local geometric and photometric properties. We call these 3D textons. Examples might be ridges, grooves, spots or stripes or combinations thereof. Associated with each texton is an appearance vector, which characterizes the local irradiance distribution, represented as a set of linear Gaussian derivative filter outputs, under different lighting and viewing conditions.Given a large collection of images of different materials, a clustering approach is used to acquire a small (on the order of 100) 3D texton vocabulary. Given a few (1 to 4) images of any material, it can be characterized using these textons. We demonstrate the application of this representation for recognition of the material viewed under novel lighting and viewing conditions.
ICCV	Control in a 3D Reconstruction System using Selective Perception.	Maurício Marengoni,Allen R. Hanson,Shlomo Zilberstein,Edward M. Riseman	1999	This paper presents a control structure for general purpose image understanding that addresses both the high level of uncertainty in local hypotheses and the computational complexity of image interpretation. The control of vision algorithms is performed by an independent subsystem that uses Bayesian networks and utility theory to compute the marginal value of information provided by alternative operators and selects the ones with the highest value.We have implemented and tested this control structure with several aerial image data-sets. The results show that the knowledge base used by the system can be acquired using standard learning techniques and that the value-driven approach to the selection of vision algorithms leads to performance gains. Moreover, the modular system architecture simplifies the addition of both control knowledge and new vision algorithms.
ICCV	Fluid Motion Recovery by Coupling Dense and Parametric Vector Fields.	Étienne Mémin,Patrick Pérez	1999	Fluid Motion Recovery by Coupling Dense and Parametric Vector Fields.
ICCV	Capturing Articulated Human Hand Motion: A Divide-and-Conquer Approach.	Ying Wu,Thomas S. Huang	1999	Capturing Articulated Human Hand Motion: A Divide-and-Conquer Approach.
ICCV	A Pattern Classification Approach to Dynamical Object Detection.	Constantine Papageorgiou,Tomaso Poggio	1999	Current systems for object detection in video sequences rely on explicit dynamical models like Kalman filters or hidden Markov models. There is significant overhead needed in the development of such systems as well as the a priori assumption that the object dynamics can be described with such a dynamical model. This paper describes a new pattern classification technique for object detection in video sequences that uses a rich, over complete dictionary of wavelet features to describe an object class. Unlike previous work where a small subset of features was selected from the dictionary, this system does no feature selection and learns the model in the full 1,326 dimensional feature space. Comparisons using different sized sets of several types of features are given. We extend this representation into the time domain without assuming any explicit model of dynamics. This data driven approach produces a model of the physical structure and short-time dynamical characteristics of people from a training set of examples; no assumptions are made about the motion of people, just that short sequences characterize their dynamics sufficiently for the purposes of detection. One of the main benefits of this approach is that transient false positives are reduced. This technique compares favorably with the static detection approach and could be applied to other object classes. We also present a real-time version of one of our static people detection systems.
ICCV	Geodesic Active Regions for Motion Estimation and Tracking.	Nikos Paragios,Rachid Deriche	1999	Geodesic Active Regions for Motion Estimation and Tracking.
ICCV	Geodesic Active Regions for Supervised Texture Segmentation.	Nikos Paragios,Rachid Deriche	1999	This paper presents a novel variational method for supervised texture segmentation. The textured feature space is generated by filtering the given textured images using isotropic and anisotropic filters, and analyzing their responses as multi-component conditional probability density functions.The texture segmentation is obtained by unifying region and boundary-based information as an improved Geodesic Active Contour Model. The defined objective function is minimized using a gradient-descent method where a level set approach is used to implement the obtained PDE.According to this PDE, the curve propagation towards the final solution is guided by boundary and region-based segmentation forces, and is constrained by a regularity force. The level set implementation is performed using a fast front propagation algorithm where topological changes are naturally handled. The performance of our method is demonstrated on a variety of synthetic and real textured frames.
ICCV	Principal Manifolds and Bayesian Subspaces for Visual Recognition.	Baback Moghaddam	1999	"We investigate the use of linear and nonlinear principal manifolds for learning low-dimensional representations for visual recognition. Three techniques: Principal Component Analysis (PCA), Independent Component Analysis (ICA) and Nonlinear PCA (NLPCA) are examined and tested in a visual recognition experiment using a large gallery of facial images from the ""FERET"" database. We compare the recognition performance of a nearest-neighbor matching rule with each principal manifold representation to that of a maximum a posteriori (MAP) matching rule using a Bayesian similarity measure derived from probabilistic subspaces and demonstrate the superiority of the latter."
ICCV	Cluster-based Segmentation of Natural Scenes.	Eric J. Pauwels,Greet Frederix	1999	In cluster-based segmentation pixels are mapped into various feature-spaces whereupon they are subjected to a grouping-algorithm. In this paper we develop a robust and versatile non-parametric clustering algorithm that is able to handle the unbalanced and irregular clusters encountered in such segmentation- applications.The strength of our approach lies in the definition and use of two cluster-validity indices that are independent of the cluster-topology. By combining them, an excellent clustering can be identified, and experiments confirm that the associated clusters do indeed correspond to perceptually salient image-regions.
ICCV	Exploiting Human Actions and Object Context for Recognition Tasks.	Darnell J. Moore,Irfan A. Essa,Monson H. Hayes	1999	Exploiting Human Actions and Object Context for Recognition Tasks.
ICCV	A Dynamic Bayesian Network Approach to Figure Tracking using Learned Dynamic Models.	Vladimir Pavlovic,James M. Rehg,Tat-Jen Cham,Kevin P. Murphy	1999	A Dynamic Bayesian Network Approach to Figure Tracking using Learned Dynamic Models.
ICCV	Depth Estimation using Defocused Stereo Image Pairs.	Uma Mudenagudi,Subhasis Chaudhuri	1999	Depth Estimation using Defocused Stereo Image Pairs.
ICCV	The PDAF based Active Contour.	Natan Peterfreund	1999	The PDAF based Active Contour.
ICCV	The Recovery of Object Shape and Camera Motion using a Sensing System with a Video Camera and a Gyro Sensor.	Toshiharu Mukai,Noboru Ohnishi	1999	The Recovery of Object Shape and Camera Motion using a Sensing System with a Video Camera and a Gyro Sensor.
ICCV	Resynthesizing Facial Animation through 3D Model-based Tracking.	Frederic H. Pighin,Richard Szeliski,David Salesin	1999	Resynthesizing Facial Animation through 3D Model-based Tracking.
ICCV	Continuous Dynamic Time Warping for Translation-Invariant Curve Alignment with Applications to Signature Verification.	Mario E. Munich,Pietro Perona	1999	Continuous Dynamic Time Warping for Translation-Invariant Curve Alignment with Applications to Signature Verification.
ICCV	A Simple and Efficient Rectification Method for General Motion.	Marc Pollefeys,Reinhard Koch,Luc J. Van Gool	1999	A Simple and Efficient Rectification Method for General Motion.
ICCV	Vision in Bad Weather.	Shree K. Nayar,Srinivasa G. Narasimhan	1999	Current vision systems are designed to perform in clear weather. Needless to say, in any outdoor application, there is no escape from &ldquo;bad&rdquo; weather. Ultimately, computer vision systems must include mechanisms that enable them to function (even if somewhat less reliably) in the presence of haze, fog, rain, hail and snow. We begin by studying the visual manifestations of different weather conditions. For this, we draw on what is already known about atmospheric optics. Next, we identify effects caused by bad weather that can be turned to our advantage. Since the atmosphere modulates the information carried from a scene point to the observer it can be viewed as a mechanism of visual information coding. Based on this observation, we develop models and methods for recovering pertinent scene properties, such as three-dimensional structure, from images taken under poor weather conditions
ICCV	Measuring Convexity for Figure/Ground Separation.	Hsing-Kuo Pao,Davi Geiger,Nava Rubin	1999	"In human perception, convex surfaces have a strong tendency to be perceived as the ""figure"". Convexity has a stronger influence on figural organization than other global shape properties, such as symmetry ([9]). And yet, there has been very little work on convexity properties in computer vision.We present a model for figure/ground segregation which exhibits a preference for convex regions as the figure (i.e., the foreground). The model also shows a preference for smaller regions to be selected as figures, which is also known to hold for human visual perception (e.g., Koffka [11]). The model is based on the machinery of Markov random fields/random walks/diffusion processes, so that the global shape properties are obtained via local and stochastic computations. Experimental results demonstrate that our model performs well on ambiguous figure/ground displays which were not captured before. In particular, in ambiguous displays where neither region is strictly convex, the model shows preference to the ""more convex"" region, thus offering a continuous measure of convexity in agreement with human perception."
ICCV	Empirical Evaluation of Dissimilarity Measures for Color and Texture.	Jan Puzicha,Yossi Rubner,Carlo Tomasi,Joachim M. Buhmann	1999	Empirical Evaluation of Dissimilarity Measures for Color and Texture.
ICCV	Appearance Compression and Synthesis based on 3D Model for Mixed Reality.	Ko Nishino,Yoichi Sato,Katsushi Ikeuchi	1999	Appearance Compression and Synthesis based on 3D Model for Mixed Reality.
ICCV	Inherent Two-Way Ambiguity in 2D Projective Reconstruction from Three Uncalibrated 1D Images.	Long Quan	1999	Inherent Two-Way Ambiguity in 2D Projective Reconstruction from Three Uncalibrated 1D Images.
ICCV	Gesture Recognition using Character Recognition Techniques on Two-Dimensional Eigenspace.	Hiroshi Ohno,Masanobu Yamamoto	1999	Gesture Recognition using Character Recognition Techniques on Two-Dimensional Eigenspace.
ICCV	Higher Order Statistical Learning for Vehicle Detection in Images.	A. N. Rajagopalan,Philippe Burlina,Rama Chellappa	1999	"The paper describes a scheme for detecting vehicles in images. The proposed method approximately models the unknown distribution of the images of vehicles by learning higher order statistics (HOS) information of the `vehicle class' from sample images. Given a test image, statistical information about the background is learnt `on the y'. An HOS-based decision measure then classifies test patterns as vehicles or otherwise.When tested on real images of aerial views of vehicular activity, the method gives good results even on complicated scenes. It does not require any a priori information about the site. However, it is amenable to augmentation with contextual information. The method can serve as an important step towards building an automated roadway monitoring system."
ICCV	Fast and Accurate Self-Calibration.	John Oliensis	1999	This paper describes new techniques for self calibration and for recovering the motion from a projective reconstruction when the calibration is known. We show that our approach deals with the ambiguities in self calibration produced by special motions. We extend our techniques to deal with varying calibration parameters. In passing, we prove convergence for the iterative projective reconstruction algorithm of Sturm/Triggs and Berthilsson/Heyden/Sparr.
ICCV	Fast Algorithms for Projective Multi-Frame Structure from Motion.	John Oliensis,Yakup Genc	1999	Fast Algorithms for Projective Multi-Frame Structure from Motion.
ICCV	New Algorithms for Two-Frame Structure from Motion.	John Oliensis,Yakup Genc	1999	We describe two new algorithms for two frame structure from motion from tracked point features. One is the first fast algorithm for computing an exact least squares estimate. It exploits our observation that the rotationally invariant least squares error can be written in a simple form that depends just on the motion. The other is essentially as accurate as the least squares estimate and is more efficient, probably faster, and potentially more robust than previous algorithms of comparable accuracy. We also analyze theoretically the accuracy of the optical flow approximation to the least squares error.
ICCV	A Cluster-based Statistical Model for Object Detection.	Thomas D. Rikert,Michael J. Jones,Paul A. Viola	1999	This paper presents an approach to object detection which is based on recent work in statistical models for texture synthesis and recognition (Heeger and Bergen 1995, De Bonet and Viola 1998, Zhu et al. 1998, Simoncelli and Portilla 1998). Our method follows the texture recognition work of De Bonet and Viola. We use feature vectors which capture the joint occurrence of local features at multiple resolutions. The distribution of feature vectors for a set of training images of an object class is estimated by clustering the data and then forming a mixture of gaussian model. The mixture model is further refined by determining which clusters are the most discriminative for the class and retaining only those clusters. After the model is learned, test images are classified by computing the likelihood of their feature vectors with respect to the model. We present promising results in applying our technique to face detection and car detection.
ICCV	Classification of Human Body Motion.	Jens Rittscher,Andrew Blake	1999	Classification of Human Body Motion.
ICCV	Trajectory Triangulation over Conic Sections.	Amnon Shashua,Shai Avidan,Michael Werman	1999	Trajectory Triangulation over Conic Sections.
ICCV	Texture-based Image Retrieval without Segmentation.	Yossi Rubner,Carlo Tomasi	1999	Image segmentation is not only hard and unnecessary for texture-based image retrieval, but can even be harmful. Images of either individual or multiple textures are best described by distributions of spatial frequency descriptors, rather than single descriptor vectors over pre-segmented regions.A retrieval method based on the Earth Movers Distance with an appropriate ground distance is shown to handle both complete and partial multi-textured queries. As an illustration, different images of the same type of animal are easily retrieved together. At the same time, animals with subtly different coats, like cheetahs and leopards, are properly distinguished.
ICCV	Omnivergent Stereo.	Heung-Yeung Shum,Adam Kalai,Steven M. Seitz	1999	Omnivergent Stereo.
ICCV	Stereo Reconstruction from Multiperspective Panoramas.	Heung-Yeung Shum,Richard Szeliski	1999	Stereo Reconstruction from Multiperspective Panoramas.
ICCV	Rigid and Articulated Motion Seen with an Uncalibrated Stereo Rig.	Andreas Ruf,Radu Horaud	1999	This paper establishes a link between uncalibrated stereo vision and the motion of rigid and articulated bodies. The variation in the projective reconstruction of a dynamic scene over time allows an uncalibrated stereo rig to be used as a faithful motion capturing device.We introduce an original theoretical framework - projective kinematics - which allows rigid and articulated motion to be represented within the transformation group of projective space. Corresponding projective velocities are defined in the tangent space. Most importantly, these projective motions inherit the Lie-group structure of the displacement group.These theoretical results lead immediately to non-metric formulations of visual serving, tracking, motion capturing and motion synthesis systems, that no longer require the metric geometry of a stereo camera or of the articulated body to be known. We report on such a non-metric formulation of a visual serving system and present simulated experimental results.
ICCV	The Hamilton-Jacobi Skeleton.	Kaleem Siddiqi,Sylvain Bouix,Allen Tannenbaum,Steven W. Zucker	1999	The eikonal equation and variants of it are of significant interest for problems in computer vision and image processing. It is the basis for continuous versions of mathematical morphology, stereo, shape-from-shading and for recent dynamic theories of shape. Its numerical simulation can be delicate, owing to the formation of singularities in the evolving front, and is typically based on level set methods. However, there are more classical approaches rooted in Hamiltonian physics, which have received little consideration in computer vision.In this paper we first introduce a new algorithm for simulating the eikonal equation, which offers a number of computational and conceptual advantages over the earlier methods when it comes to shock tracking. Next, we introduce a very efficient algorithm for shock detection, where the key idea is to measure the net outward flux of a vector field per unit volume, and to detect locations where a conservation of energy principle is violated. We illustrate the approach with several numerical examples including skeletons of complex 2D and 3D shapes.
ICCV	Corner Detection in Textured Color Images.	Mark A. Ruzon,Carlo Tomasi	1999	Corner models in the literature have lagged behind edge models with respect to color and shading. We use both a region model, based on distributions of pixel colors, and an edge model, which removes false positives, to perform corner detection on color images whose regions contain texture. We show results on a variety of natural images at different scales that highlight the problems that occur when boundaries between regions have curvature.
ICCV	Learning and Evaluating Visual Features for Pose Estimation.	Robert Sim,Gregory Dudek	1999	We present a method for learning a set of visual landmarks which are useful for pose estimation. The landmark learning mechanism is designed to be applicable to a wide range of environments, and generalized for different approaches to computing a pose estimate. Initially, each landmark is detected as a local extreme of a measure of distinctiveness and represented by a principal components encoding which is exploited for matching. Attributes of the observed landmarks can be parameterized using a generic parameterization method and then evaluated in terms of their utility for pose estimation. We present experimental evidence that demonstrates the utility of the method.
ICCV	Coupled Lighting Direction and Shape Estimation from Single Images.	Dimitris Samaras,Dimitris N. Metaxas	1999	This paper presents a new method for the simultaneous estimation of lighting direction and shape from shading. The method estimates the shape and the lighting direction using a two step iterative process. We assume an initial (possibly incorrect) estimate of the lighting position. A stiff deformable model is then fitted to the image, assuming this lighting position. Next, a least-squares estimate of the lighting position is derived from the model using the Levenberg-Marquart method.The two steps - model fitting and lighting-position estimation - are iterated. Once the light direction has converged to a stable solution the deformable model stiffness is lowered and the model fits accurately given the lighting model. In addition, we show how the method can be used with either orthographic or perspective projection assumptions. In a variety of experiments on real and synthetic data, the method is robust to errors both to the initial light position and shape estimates.
ICCV	Illumination Distribution from Brightness in Shadows: Adaptive Estimation of Illumination Distribution with Unknown Reflectance Properties in Shadow Regions.	Imari Sato,Yoichi Sato,Katsushi Ikeuchi	1999	Illumination Distribution from Brightness in Shadows: Adaptive Estimation of Illumination Distribution with Unknown Reflectance Properties in Shadow Regions.
ICCV	Independent Motion Detection in 3D Scenes.	Harpreet S. Sawhney,Yanlin Guo,Jane C. Asmuth,Rakesh Kumar	1999	Independent Motion Detection in 3D Scenes.
ICCV	The Optimal Axial Interval in Estimating Depth from Defocus.	Yoav Y. Schechner,Nahum Kiryati	1999	We analyze the effect of perturbations on the estimation of Depth from Defocus (DFD) implemented by changing the focus setting (e.g., axially moving the sensor). The analysis yields the optimal change of focus setting, and the spatial frequencies for which estimation is most robust. For stable estimation at all spatial frequencies, the change in focus setting should be less than twice the depth of field. For the most robust estimation in the highest spatial frequencies the axial interval should be equal to the depth of field.
ICCV	Polarization-based Decorrelation of Transparent Layers: The Inclination Angle of an Invisible Surface.	Yoav Y. Schechner,Joseph Shamir,Nahum Kiryati	1999	When a transparent surface is present between an observer and an object, an image reflected by the surface may be superimposed on the image of the observed object. We present a new approach to recover the scenes (layers) and to classify which is the reflected/transmitted one, based on imaging through a polarizing filter at two orientations. Estimates of the separate layers are obtained by weighted pixel-wise differences of these images, inverting the image formation process. However, the weights depend on the angle of incidence, hence on the inclination of the transparent (invisible) surface. This angle is estimated by seeking the angle-value which (through the weights) leads to decorrelation of the estimated layers. Experimental results, obtained using real photos of actual objects, demonstrate the success of angle estimation and consequent layer separation and labeling. The method is shown to be superior to earlier methods where only raw optical data was used.
ICCV	Probabilistic Object Recognition and Localization.	Bernt Schiele,Alex Pentland	1999	Probabilistic Object Recognition and Localization.
ICCV	Monocular Perception of Biological Motion - Detection and Labeling.	Yang Song,Luis Goncalves,Enrico Di Bernardo,Pietro Perona	1999	Computer perception of biological motion is key to developing convenient and powerful human-computer inter-faces. Successful body tracking algorithms have been developed; however, initialization is done by hand.We propose a method for detecting a moving human body and for labeling its parts automatically. It is based on maximizing the joint probability density function (PDF) of the position and velocity of the body parts. The PDF is estimated from training data. Dynamic programming is used for calculating efficiently the best global labeling on an approximation of the PDF. The computational cost is on the order of N 4 where N is the number of features detected.We explore the performance of our method with experiments carried on a variety of periodic and non-periodic body motions viewed monocularly for a total of approximately 30,000 frames. Point-markers were strapped to the joints of the subject for facilitating image analysis. We find an average of 2.3% labeling error; the experiments also suggest a high degree of viewpoint-invariance.
ICCV	Error Detection and DEM Fusion using Self-Consistency.	Howard J. Schultz,Dong-Min Woo,Frank Stolle,Edward M. Riseman	1999	Error Detection and DEM Fusion using Self-Consistency.
ICCV	Utilizing Scatter for Pixel Subspace Selection.	Haim Schweitzer	1999	Measures of scatter are used in statistical pattern recognition to identify and select important features, computed as linear combinations of the given features. Examples include principal components and linear discriminants. The classic computational procedures require eigenvector decomposition of large matrices, and in the case of images they are only practical for identifying a low dimensional feature subspace.We investigate the case in which the selected features are required to be a subset of the given features. It is shown that the same scatter measures used in the general case can also be used in this discrete selection case, but the computational procedure no longer involves matrix eigenvector decomposition.Instead, the selection of pixels that optimize scatter measures can be accomplished by a very simple and efficient discrete optimization technique that runs in linear time regardless of the subspace size. Applications to clustering and content based indexing are discussed.
ICCV	Fast Partial Search Solution to the 3D SFM Problem.	S. Srinivasan	1999	Fast Partial Search Solution to the 3D SFM Problem.
ICCV	About the Self-Calibration of a Rotating and Zooming Camera: Theory and Practice.	Yongduek Seo,Ki-Sang Hong	1999	About the Self-Calibration of a Rotating and Zooming Camera: Theory and Practice.
ICCV	Subpixel-Precise Extraction of Watersheds.	Carsten Steger	1999	An approach to extract watersheds and watercourses, as well as their corresponding valleys and hills, from images with subpixel precision is proposed. The critical points of the terrain are essential as the starting points for the construction of these separatrices. They are extracted efficiently with subpixel precision using an approach based on derivatives of Gaussian filters. The separatrices are extracted by integrating their defining differential equation.Finally, the hills and valleys are constructed by an efficient graph search algorithm. Examples show the quality of the results that can be achieved with the proposed approach.
ICCV	Object Localization by Bayesian Correlation.	Josephine Sullivan,Andrew Blake,Michael Isard,John MacCormick	1999	Maximization of cross-correlation is a commonly used principle for intensity-based object localization that gives a single estimate of location. However, to facilitate sequential inference (eg over time or scale) and to allow the representation of ambiguity, it is desirable to represent an entire probability distribution for object location. Although the cross-correlation itself (or some function of it) has sometimes been treated as a probability distribution, this is not generally justifiable.Bayesian correlation achieves a consistent probabilistic treatment by combining several developments. The first is the interpretation of correlation matching functions in probabilistic terms, as observation likelihoods. Second, probability distributions of filter-bank responses are learned from training examples. Inescapably, response-learning also demands statistical modeling of background intensities, and there are links here with image coding and Independent Component Analysis. Lastly, multi-scale processing is achieved, in a Bayesian context, by means of a new algorithm, layered sampling, for which asymptotic properties are derived.
ICCV	Prediction Error as a Quality Metric for Motion and Stereo.	Richard Szeliski	1999	This paper presents a new methodology for evaluating the quality of motion estimation and stereo correspondence algorithms. Motivated by applications such as novel view generation and motion-compensated compression, we suggest that the ability to predict new views or frames is a natural metric for evaluating such algorithms.Our new metric has several advantages over comparing algorithm outputs to true motions or depths. First of all, it does not require the knowledge of ground truth data, which may be difficult or laborious to obtain. Second, it more closely matches the ultimate requirements of the application, which are typically tolerant of errors in uniform color regions, but very sensitive to isolated pixel errors or disocclusion errors.In the paper, we develop a number of error metrics based on this paradigm, including forward and inverse prediction errors, residual motion error, and local motion-compensated prediction error. We show results on a number of widely used motion and stereo sequences, many of which do not have associated ground truth data.
ICCV	Robust Estimation of Curvature Information from Noisy 3D Data for Shape Description.	Chi-Keung Tang,Gérard G. Medioni	1999	Robust Estimation of Curvature Information from Noisy 3D Data for Shape Description.
ICCV	Epipolar Geometry Estimation by Tensor Voting in 8D.	Chi-Keung Tang,Gérard G. Medioni,Mi-Suen Lee	1999	Epipolar Geometry Estimation by Tensor Voting in 8D.
ICCV	Direction Diffusion.	Bei Tang,Guillermo Sapiro,Vicent Caselles	1999	In a number of disciplines, directional data provides a fundamental source of information. A novel framework for isotropic and anisotropic diffusion of directions is presented in this paper. The framework can be applied both to regularize directional data and to obtain multi-scale representations of it. The basic idea is to apply and extend results from the theory of harmonic maps in liquid crystals.This theory deals with the regularization of vectorial data, while satisfying the unit norm constraint of directional data. We show the corresponding variational and partial differential equations formulations for isotropic diffusion, obtained from an L2 norm, and edge preserving diffusion, obtained from an L1 norm.In contrast with previous approaches, the framework is valid for directions in any dimensions, supports non-smooth data, and gives both isotropic and anisotropic formulations. We present a number of theoretical results, open questions, and examples for gradient vectors, optical flow, and color images.
ICCV	Symmetry Maps of Free-Form Curve Segments Via Wave Propagation.	Hüseyin Tek,Benjamin B. Kimia	1999	"This paper presents an approach for computing the symmetries (skeletons) of an edge map consisting of a collection of curve segments. This approach is a combination of analytic computations in the style of computational geometry and discrete propagations on a grid in the style of the numerical solutions of PDE's. Specifically, waves from each of the initial curve segments are initialized and propagated as a discrete wavefront along discrete directions. In addition, to avoid error built up due to the discrete nature of propagation, shockwaves are detected and explicitly propagated along a secondary dynamic grid. The propagation of shockwaves, integrated with the propagation of the wavefront along discrete directions, leads to an exact simulation of propagation by the Eikonal equation. The resulting symmetries are simply the collection of shockwaves formed in this process which can be manipulated locally, exactly, and efficiently under local changes in an edge map (gap completion, removal of spurious elements, etc). The ability to express grouping operations in the language of symmetry maps makes it an appropriate intermediate representation between low-level edge maps and high level object hypotheses."
ICCV	An Integrated Bayesian Approach to Layer Extraction from Image Sequences.	Philip H. S. Torr,Richard Szeliski,P. Anandan	1999	This paper describes a Bayesian approach for modeling 3D scenes as a collection of approximately planar layers that are arbitrarily positioned and oriented in the scene. In contrast to much of the previous work on layer-based motion modeling, which computes layered descriptions of 2D image motion, our work leads to a 3D description of the scene. There are two contributions within the paper. The first is to formulate the prior assumptions about the layers and scene within a Bayesian decision making framework which is used to automatically determine the number of layers and the assignment of individual pixels to layers. The second is algorithmic. In order to achieve the optimization, a Bayesian version of RANSAC is developed with which to initialize the segmentation. Then, a generalized expectation maximization method is used to find the MAP solution.
ICCV	Semantic Organization of Scenes using Discriminant Structural Templates.	Antonio Torralba,Aude Oliva	1999	In this paper, we present a procedure for organizing real world scenes along semantic axes. The approach is based on the output energies of linear discriminant filters that take into account, or not, spatial information.We introduce three semantic axes along which pictures are ordered. The main semantic axis computes the degree of naturalness of a scene. Then, urban pictures are evaluated according to their degree of verticalness and natural scenes, according to their degree of openness. We observe the emergence of typical scene categories such as beach, mountain, skyscrapers, city center, etc., along the axes.
ICCV	Wallflower: Principles and Practice of Background Maintenance.	Kentaro Toyama,John Krumm,Barry Brumitt,Brian Meyers	1999	Wallflower: Principles and Practice of Background Maintenance.
ICCV	Camera Pose and Calibration from 4 or 5 Known 3D Points.	Bill Triggs	1999	Camera Pose and Calibration from 4 or 5 Known 3D Points.
ICCV	Differential Matching Constraints.	Bill Triggs	1999	Differential Matching Constraints.
ICCV	Three-Dimensional Scene Flow.	Sundar Vedula,Simon Baker,Peter Rander,Robert T. Collins,Takeo Kanade	1999	Just as optical flow is the two-dimensional motion of points in an image, scene flow is the three-dimensional motion of points in the world. The fundamental difficulty with optical flow is that only the normal flow can be computed directly from the image measurements, without some form of smoothing or regularization. In this paper, we begin by showing that the same fundamental limitation applies to scene flow; however, many cameras are used to image the scene. There are then two choices when computing scene flow: 1) perform the regularization in the images or 2) perform the regularization on the surface of the object in the scene. In this paper, we choose to compute scene flow using regularization in the images. We describe three algorithms, the first two for computing scene flow from optical flows and the third for constraining scene structure from the inconsistencies in multiple optical flows.
ICCV	Parallel Hidden Markov Models for American Sign Language Recognition.	Christian Vogler,Dimitris N. Metaxas	1999	Parallel Hidden Markov Models for American Sign Language Recognition.
ICCV	Segmentation using Eigenvectors: A Unifying View.	Yair Weiss	1999	Automatic grouping and segmentation of images remains a challenging problem in computer vision. Recently, a number of authors have demonstrated good performance on this task using methods that are based on eigenvectors of the affinity matrix. These approaches are extremely attractive in that they are based on simple eigen-decomposition algorithms whose stability is well understood. Nevertheless, the use of eigen-decompositions in the context of segmentation is far from well understood.In this paper we give a unified treatment of these algorithms, and show the close connections between them while highlighting their distinguishing features. We then prove results on eigenvectors of block matrices that allow us to analyze the performance of these algorithms in simple grouping settings. Finally, we use our analysis to motivate a variation on the existing methods that combines aspects from different eigenvector segmentation algorithms. We illustrate our analysis with results on real and synthetic images.
ICCV	3D Surface Topography from Intensity Images.	Philip L. Worthington,Edwin R. Hancock	1999	The paper demonstrates how a new shape from shading scheme can be used to extract topographic information from 2D intensity imagery. The shape-from-shading scheme has two novel ingredients. Firstly, it uses a geometric update procedure which allows the image irradiance equation to be satisfied as a hard constraint. This not only improves the data closeness of the recovered needle-map, but also removes the necessity for extensive parameter tuning. Secondly, we use curvature information to impose topographic constraints on the recovered needle-map. The topographic information is captured using the shape index of J.J. Koenderink and A.J. Van Doorn (1995) and consistency is imposed using a robust error function. We show that the new shape-from-shading scheme leads to a meaningful topographic labelling of 3D surface structures
ICCV	Equivalence of Julesz and Gibbs Texture Ensembles.	Ying Nian Wu,Song Chun Zhu,Xiuwen Liu	1999	Research on texture has been pursued along two different lines. The first line of research, pioneered by Julesz (1962), seeks the essential ingredients in terms of features and statistics in human texture perception. This leads us to a mathematical definition of texture as a Julesz ensemble. A Julesz ensemble is the maximum set of images that share the same value of some basic feature statistics as the image \math, or equivalently it is a uniform distribution on this set. The second line of research studies statistical models, in particular, Markov random field (MRF) and FRAME models (Zhu, Wu, and Mumford 1997), to characterize texture patterns locally.In this article, we bridge the two lines by the fundamental principle of equivalence of ensembles in statistical mechanics (Gibbs, 1902). We prove that 1). The conditional probability of an arbitrary image patch given its environment, under the Julesz ensemble or the uniform model, is inevitably a FRAME (MRF) model, and 2). The limit of the FRAME (MRF) model, which we called the Gibbs ensemble, is equivalent to a Julesz ensemble as \math. Thus the advantages of the two methodologies can be fully utilized.
ICCV	Invariant-based Registration of Surface Patches.	Joris Vanden Wyngaerd,Luc J. Van Gool,Reinhard Koch,Marc Proesmans	1999	Invariant-based Registration of Surface Patches.
ICCV	Estimation of Composite Object and Camera Image Motion.	Yaser Yacoob,Larry S. Davis	1999	Estimation of Composite Object and Camera Image Motion.
ICCV	Tracking Rigid Motion using a Compact-Structure Constraint.	Yaser Yacoob,Larry S. Davis	1999	Tracking Rigid Motion using a Compact-Structure Constraint.
ICCV	Free-Form Surface Registration using Surface Signatures.	Sameh M. Yamany,Aly A. Farag	1999	This paper introduces a new free-form surface representation scheme for the purpose of fast and accurate registration and matching. Accurate registration of surfaces is a common task in computer vision. The proposed representation scheme captures the surface curvature information, seen from certain points and produces images, called surface signatures, at these points. Matching signatures of different surfaces enables the recovery of the transformation parameters between these surfaces. We propose to use template matching to compare the signature images. To enable partial matching, another criterion, the overlap ratio, is used. This representation scheme can be used as a global representation of the surface as well as a local one and performs near real-time registration. We show that the signature representation can be used to match objects in 3-D scenes in the presence of clutter and occlusion. Applications presented include free-form object matching, multimodal medical volumes registration and dental teeth reconstruction from intra-oral images.
ICCV	A Statistical Approach to Snakes for Bimodal and Trimodal Imagery.	Anthony J. Yezzi,Andy Tsai,Alan S. Willsky	1999	In this paper, we describe a new region-based approach to active contours for segmenting images composed of two or three types of regions characterizable by a given statistic. The essential idea is to derive curve evolutions which separate two or more values of a predetermined set of statistics computed over geometrically determined subsets of the image. Both global and local image information is used to evolve the active contour. Image derivatives, however, are avoided, thereby giving rise to a further degree of noise robustness compared to most edge-based snake algorithms.
ICCV	Detecting Salient Motion by Accumulating Directionally-Consistent Flow.	Lambert E. Wixson,Michael W. Hansen	1999	"Motion detection can play an important role in many vision tasks. Yet image motion can arise from ""uninteresting"" events as well as interesting ones. In this paper, salient motion is defined as motion that is likely to result from a typical surveillance target (e.g., a person or vehicle traveling with a sense of direction through a scene) as opposed to other distracting motions (e.g., the scintillation of specularities on water, the oscillation of vegetation in the wind).We propose an algorithm for detecting this salient motion that is based on intermediate-stage vision integration of optical flow. Empirical results are presented that illustrate the applicability of the proposed methods to real-world video. Unlike many motion detection schemes, no knowledge about expected object size or shape is necessary for rejecting the distracting motion."
ICCV	Multi-View Subspace Constraints on Homographies.	Lihi Zelnik-Manor,Michal Irani	1999	"The motion of a planar surface between two camera views induces a homography. The homography depends on the camera intrinsic and extrinsic parameters, as well as on the 3D plane parameters. While camera parameters vary across different views, the plane geometry remains the same. Based on this fact, we derive linear subspace constraints on the relative motion of multiple (\math) planes across multiple views.The paper has three main contributions: (i) We show that the collection of all ""relative homographies"" of a pair of planes (homologies) across multiple views, spans a 4-dimensional linear subspace. (ii) We show how this constraint can be extended to the case of multiple planes across multiple views. (iii) We suggest two potential application areas which can benefit from these constraints: (a) The accuracy of homography estimation can be improved by enforcing the multi-view subspace constraints. (b) Violations of these multi-view constraints can be used as a cue for moving object detection. All the results derived in this paper are true for uncalibrated cameras."
ICCV	Flexible Camera Calibration by Viewing a Plane from Unknown Orientations.	Zhengyou Zhang	1999	Flexible Camera Calibration by Viewing a Plane from Unknown Orientations.
ICCV	What Can Be Determined from a Full and a Weak Perspective Image?	Zhengyou Zhang,P. Anandan,Heung-Yeung Shum	1999	What Can Be Determined from a Full and a Weak Perspective Image?
IEEE Visualization	"""Whole Field Modelling"" - Effective Real-Time and Post-Survey Visualization of Underwater Pipelines."	Paul Chapman,Derek Wills,Peter Stevens,Graham Brookes	1999	"""Whole Field Modelling"" - Effective Real-Time and Post-Survey Visualization of Underwater Pipelines."
IEEE Visualization	Visualizing the Evolution of a Subject Domain: A Case Study.	Chaomei Chen,Les Carr	1999	We explore the potential of information visualization techniquesin enhancing existing methodologies for domain analysis andmodeling. In this case study, we particularly focus on visualizingthe evolution of the hypertext field based on author co-citationpatterns, including the use of a sliding-window scheme to generatea series of annual snapshots of the domain structure, and a factor-referencedcolor-coding scheme to highlight predominantspecialties in the field.
IEEE Visualization	Optimal Triangular Haar Bases for Spherical Data.	Georges-Pierre Bonneau	1999	Multiresolution analysis based on FWT (Fast Wavelet Transform)is now widely used in Scientific Visualization. Spherical bi-orthogonalwavelets for spherical triangular grids where introducedin [5]. In order to improve on the orthogonality of the wavelets, theconcept of nearly orthogonality, and two new piecewise-constant(Haar) bases were introduced in [4]. In our paper, we extend theresults of [4]. First we give two one-parameter families of triangu-larHaar wavelet bases that are nearly orthogonal in the sense of [4].Then we introduce a measure of orthogonality. This measure vanishesfor orthogonal bases. Eventually, we show that we can find anoptimal parameter of our wavelet families, for which the measure oforthogonality is minimized. Several numerical and visual examplesfor a spherical topographic data set illustrates our results.
IEEE Visualization	Interacitve Exploration of Extra- and Interacranial Blood Vessels.	Dirk Bartz,Wolfgang Straßer,Martin Skalej,Dorothea Welte	1999	We present a system for interactive explorations of extra- andintracranial blood vessels. Starting with a stack of images from3D angiography, we use virtual clips to limit the segmentation ofthe vessel tree to the parts the neuroradiologists are interested in.Furthermore, methods of interactive virtual endoscopy are appliedin order to provide an interior view of the blood vessels.
IEEE Visualization	"Feature Comparisons of 3-D Vector Fields Using Earth Mover's Distance."	Rajesh Batra,Lambertus Hesselink	1999	A method for comparing three-dimensional vector fields constructed from simple critical points is described. This method is a natural extension of the previous work [1] which defined a distance metric for comparing two-dimensional fields.The extension to three-dimensions follows the path of our previous work, rethinking the representation of a critical point signature and the distance measure between the points. Since the method relies on topologically based information, problems such as grid matching and vector alignment which often complicate other comparison techniques are avoided. In addition, since only feature information is used to represent, and therefore stored for each field, a significant amount of compression occurs.
IEEE Visualization	"Visualization of Conflicts and Resolutions in a ""Free Flight"" Scenario."	Ronald Azuma,Howard Neely,Michael Daily,Mario Correa	1999	"Free Flight will change today's air traffic control systemby giving pilots increased flexibility to choose and modify theirroutes in real time, reducing costs and increasing system capacity.This increased flexibility comes at the price of increasedcomplexity. If Free Flight is to become a reality, future air trafficcontrollers, pilots, and airline managers will require new conflictdetection, resolution and visualization decision support tools.This paper describes a testbed system for building and evaluatingsuch tools, including its current capabilities, lessons we learned,and feedback received from expert users. The visualization systemprovides an overall plan view supplemented with a detailedperspective view, allowing a user to examine highlighted conflictsand select from a list of proposed solutions, as the scenario runsin real time. Future steps needed to improve this system aredescribed."
IEEE Visualization	Progressive Compression and Transmission of Arbitrary Triangular Meshes.	Chandrajit L. Bajaj,Valerio Pascucci,Guozhong Zhuang	1999	The recent growth in the size and availability of large triangularsurface models has generated interest in compactmulti-resolution progressive representation and data transmission.An ongoing challenge is to design an efficient datastructure that encompasses both compactness of geometricrepresentations and visual quality of progressive representations.In this paper we introduce a topological layering baseddata-structure and an encoding scheme to build a compactprogressive representation of an arbitrary triangular mesh (a2D simplicial complex in 3D) with attached attribute data.This compact representation is composed of multiple levelsof detail that can be progressively transmitted and displayed.The global topology, which is the number of holesand connected components, can be flexibly changed amongsuccessive levels while still achieving guaranteed size of thecoarsest level mesh for very complex models. The flexibilityin our encoding scheme also allows topology preservingprogressivity.
IEEE Visualization	"DELTA's Virtual Physics Laboratory: A Comprehensive Learning Platform on Physics & Astronomy."	Sepideh Chakaveh,Udo Zlender,Detlef Skaley,Konstantinos Fostiropoulos,Dieter Breitschwerdt	1999	"DELTA's Virtual Physics Laboratory: A Comprehensive Learning Platform on Physics & Astronomy."
IEEE Visualization	Forward Image Mapping.	Baoquan Chen,Frank Dachille,Arie E. Kaufman	1999	We present a new forward image mapping algorithm, which speedsup perspective warping -- as in texture mapping. It processes thesource image in a special scanline order instead of the normalraster scanline order. This special scanline has the property of preservingparallelism when projecting to the target image. The algorithmreduces the complexity of perspective-correct image warpingby eliminating the division per pixel and replacing it with a divisionper scanline. The method also corrects the perspective distortionin Gouraud shading with negligible overhead. Furthermore, thespecial scanline order is suitable for antialiasing using a more accurateantialiasing conic filter, with minimum additional cost. Thealgorithm is highlighted by incremental calculations and optimizedmemory bandwidth by reading each source pixel only once, suggestinga potential hardware implementation.
IEEE Visualization	LOD-Sprite Technique for Accelerated Terrain Rendering.	Baoquan Chen,J. Edward Swan II,Eddy Kuo,Arie E. Kaufman	1999	We present a new rendering technique, termed LOD-sprite rendering, which uses a combination of a level-of-detail (LOD) representation of the scene together with reusing image sprites (previously rendered images). Our primary application is accelerating terrain rendering. The LOD-sprite technique renders an initial frame using a high-resolution model of the scene geometry. It renders subsequent frames with a much lower-resolution model of the scene geometry and texture-maps each polygon with the image sprite from the initial high-resolution frame. As it renders these subsequent frames the technique measures the error associated with the divergence of the view position from the position where the initial frame was rendered. Once this error exceeds a user-defined threshold, the technique re-renders the scene from the high-resolution model. We have efficiently implemented the LOD-sprite technique with texture-mapping graphics hardware. Although to date we have only applied LOD-sprite to terrain rendering, it could easily be extended to other applications. We feel LOD-sprite holds particular promise for real-time rendering systems.
IEEE Visualization	Progressive Compression of Arbitrary Triangular Meshes.	Daniel Cohen-Or,David Levin,Offir Remez	1999	In this paper we present a mesh compression method based ona multiresolution decomposition whose detail coefficients have acompact representation and thus smaller entropy than the originalmesh. Given an arbitrary triangular mesh with an irregular connectivity,we use a hierarchical simplification scheme, which generatesa multiresolution model. By reversing the process we define a hierarchicalprogressive refinement process, where a simple predictionplus a correction is used for inserting vertices to form a finerlevel. We show how the connectivity of an arbitrary triangulationcan be encoded efficiently by a coloring technique, and recoveredincrementally during the progressive reconstruction of the originalmesh.
IEEE Visualization	Skip Strips: Maintaining Triangle Strips for View-Dependent Rendering.	Jihad El-Sana,Elvir Azanli,Amitabh Varshney	1999	View-dependent simplification has emerged as a powerful toolfor graphics acceleration in visualization of complex environments.However, view-dependent simplification techniqueshave not been able to take full advantage of the underlyinggraphics hardware. Specifically, triangle strips are a widelyused hardware-supported mechanism to compactly representand efficiently render static triangle meshes. However, ina view-dependent framework, the triangle mesh connectivitychanges at every frame making it difficult to use trianglestrips. In this paper we present a novel data-structure,Skip Strip, that efficiently maintains triangle strips during suchview-dependent changes. A Skip Strip stores the vertex hierarchynodes in a skip-list-like manner with path compression.We anticipate that Skip Strips will provide a road-map tocombine rendering acceleration techniques for static datasets,typical of retained-mode graphics applications, with those fordynamic datasets found in immediate-mode applications.
IEEE Visualization	Spiraling Edge: Fast Surface Reconstruction from Partially Organized Sample Points.	Patricia Crossno,Edward Angel	1999	"Many applications produce three-dimensional points thatmust be further processed to generate a surface. Surfacereconstruction algorithms that start with a set of unorganizedpoints are extremely time-consuming. Sometimes, however,points are generated such that there is additional informationavailable to the reconstruction algorithm. We present SpiralingEdge, a specialized algorithm for surface reconstruction that isthree orders of magnitude faster than algorithms for the generalcase. In addition to sample point locations, our algorithm startswith normal information and knowledge of each point'sneighbors. Our algorithm produces a localized approximation tothe surface by creating a star-shaped triangulation between a pointand a subset of its nearest neighbors. This surface patch isextended by locally triangulating each of the points along the edgeof the patch. As each edge point is triangulated, it is removedfrom the edge and new edge points along the patch's edge areinserted in its place. The updated edge spirals out over the surfaceuntil the edge encounters a surface boundary and stops growing inthat direction, or until the edge reduces to a small hole that isfilled by the final triangle."
IEEE Visualization	Visual Debugging of Visualization Software: A Case Study for Particle Systems.	Patricia Crossno,Edward Angel	1999	Visualization systems are complex dynamic softwaresystems. Debugging such systems is difficult using conventionaldebuggers because the programmer must try to imagine the three-dimensionalgeometry based on a list of positions and attributes.In addition, the programmer must be able to mentally animatechanges in those positions and attributes to grasp dynamicbehaviors within the algorithm. In this paper we shall show thatrepresenting geometry, attributes, and relationships graphicallypermits visual pattern recognition skills to be applied to thedebugging problem. The particular application is a particlesystem used for isosurface extraction from volumetric data.Coloring particles based on individual attributes is especiallyhelpful when these colorings are viewed as animations oversuccessive iterations in the program. Although we describe aparticular application, the types of tools that we discuss can beapplied to a variety of problems.
IEEE Visualization	Isosurface Extraction Techniques for Web-Based Volume Visualization.	Klaus D. Engel,Rüdiger Westermann,Thomas Ertl	1999	The reconstruction of isosurfaces from scalar volume data has positioneditself as a fundamental visualization technique in many differentapplications. But the dramatically increasing size of volumetricdata sets often prohibits the handling of these models on affordablelow-end single processor architectures. Distributed client-serversystems integrating high-bandwidth transmission channelsand Web-based visualization tools are one alternative to attack thisparticular problem, but therefore new approaches to reduce the loadof numerical processing and the number of generated primitives arerequired. In this paper we outline different scenarios for distributedisosurface reconstruction from large-scale volumetric data sets. Wedemonstrate how to directly generate stripped surface representationsand we introduce adaptive and hierarchical concepts to minimizethe number of vertices that have to be reconstructed, transmittedand rendered. Furthermore, we propose a novel computationscheme, which allows the user to flexibly exploit locally availableresources. The proposed algorithms have been merged together inorder to build a platform-independent Web-based application. Extensiveuse of VRML and Java OpenGL-bindings allows for theexploration of large-scale volume data quite efficiently.
IEEE Visualization	Real-Time Visualization of Scalably Large Collections of Heterogeneous Objects.	Douglass Davis,William Ribarsky,T. Y. Jiang,Nickolas Faust,Sean Ho	1999	This paper presents results for real-time visualization ofout-of-core collections of 3D objects. This is a significantextension of previous methods and shows the generality ofhierarchical paging procedures applied both to globalterrain and any objects that reside on it. Applied tobuildings, the procedure shows the effectiveness of using ascreen-based paging and display criterion within ahierarchical framework. The results demonstrate that themethod is scalable since it is able to handle multiplecollections of buildings (e.g., cities) placed around theearth with full interactivity and without extensive memoryload. Further the method shows efficient handling ofculling and is applicable to larger, extended collections ofbuildings. Finally, the method shows that levels of detailcan be incorporated to provide improved detailmanagement.
IEEE Visualization	Geo Spatial Visualization for Situational Awareness.	Eliot Feibush,Nikhil Gagvani,Daniel Williams	1999	Situational awareness applications require a highly detailed geospatialvisualization covering a large geographic area. Conventionalpolygon based terrain modeling would exceed the capacity of currentcomputer rendering. Terrain visualization techniques for a situationalawareness application are described in this case study. Visualizinglarge amounts of terrain data has been achieved using verylarge texture maps. Sun shading is applied to the terrain texture mapto enhance perception of relief features. Perception of submarinepositions has been enhanced using a translucent, textured water surface.Each visualization technique is illustrated in the accompanyingvideo tape.
IEEE Visualization	Visualizing Gridded Datasets with Large Number of Missing Values.	Suzana Djurcilov,Alex Pang	1999	Much of the research in scientific visualization has focused on completesets of gridded data. This paper presents our experience dealingwith gridded data sets with large number of missing or invaliddata, and some of our experiments in addressing the shortcomingsof standard off-the-shelf visualization algorithms. In particular, wediscuss the options in modifying known algorithms to adjust to thespecifics of sparse datasets, and provide a new technique to smoothout the side-effects of the operations. We apply our findings todata acquired from NEXRAD (NEXt generation RADars) weatherradars, which usually have no more than 3 to 4 percent of all possiblecell points filled.
IEEE Visualization	Exploring Geo-Scientific Data in Virtual Environments.	Bernd Fröhlich,Stephen Barrass,Björn Zehner,John Plate,Martin Göbel	1999	"This paper describes tools and techniques for the exploration ofgeo-scientific data from the oil and gas domain in stereoscopic virtualenvironments. The two main sources of data in the explorationtask are seismic volumes and multivariate well logs of physicalproperties down a bore hole. We have developed a props-basedinteraction device called the cubic mouse to allow more direct andintuitive interaction with a cubic seismic volume. This device effectivelyplaces the seismic cube in the user's hand. Geologists whohave tried this device have been enthusiastic about the ease of use,and were adept only a few moments after picking it up.We have also developed a multi-modal visualisation and sonificationtechnique for the dense, multivariate well log data. Thevisualisation can show two well log variables mapped along thewell geometry in a bivariate colour scheme, and another variableon a sliding lens. A sonification probe is attached to the lens so thatother variables can be heard. The sonification is based on a Geiger-countermetaphor that is widely understood and which makes iteasy to explain. The data is sonified at higher or lower resolutionsdepending on the speed of the lens. Sweeps can be made at slowerrates and over smaller intervals to home in on peaks, boundaries orother features in the full resolution data set."
IEEE Visualization	Hierarchical Parallel Coordinates for Exploration of Large Datasets.	Ying-Huey Fua,Matthew O. Ward,Elke A. Rundensteiner	1999	Our ability to accumulate large, complex (multivariate) data setshas far exceeded our ability to effectively process them in search ofpatterns, anomalies, and other interesting features. Conventionalmultivariate visualization techniques generally do not scale wellwith respect to the size of the data set. The focus of this paper ison the interactive visualization of large multivariate data sets basedon a number of novel extensions to the parallel coordinates displaytechnique. We develop a multiresolutional view of the data via hierarchicalclustering, and use a variation on parallel coordinates toconvey aggregation information for the resulting clusters. Users canthen navigate the resulting structure until the desired focus regionand level of detail is reached, using our suite of navigational andfiltering tools. We describe the design and implementation of ourhierarchical parallel coordinates system which is based on extendingthe XmdvTool system. Lastly, we show examples of the toolsand techniques applied to large (hundreds of thousands of records)multivariate data sets.
IEEE Visualization	Automating Transfer Function Design for Comprehensible Volume Rendering Based on 3D Field Topology Analysis.	Issei Fujishiro,Taeko Azuma,Yuriko Takeshima	1999	Automating Transfer Function Design for Comprehensible Volume Rendering Based on 3D Field Topology Analysis.
IEEE Visualization	Multi-Resolution Multi-Field Ray Tracing: A Mathematical Overview.	Charidimos E. Gasparakis	1999	A rigorous mathematical review of ray tracing is presented. Theconcept of a generic voxel decoder acting on flexible voxel formatsis introduced. The necessity of interpolating opacity weightedcolors is proved, using a new definition of the blending processin terms of functional integrals. The continuum limit of the discreteopacity accumulation formula is presented, and its convexityproperties are investigated. The issues pertaining to interpola-tion/classification order are discussed. The lighting equation is expressedin terms of opacity weighted colors. The multi-resolution(along the ray) correction of the opacity-weighted color is derived.The mathematics of filtering on the image plane are studied, and anupper limit of the local pixel size on the image plane is obtained.Interpolation of pixel values on the image plane is shown to be inequivalentto blending of interpolated samples. The ACM Portal is published by the Association for Computing Machinery. Copyright © 2010 ACM, Inc. Terms of Usage Privacy Policy Code of Ethics Contact Us Useful downloads: Adobe Acrobat QuickTime Windows Media Player Real Player
IEEE Visualization	Time-Critical Multiresolution Scene Rendering.	Enrico Gobbetti,Eric Bouvier	1999	We describe a framework for time-critical rendering of graphicsscenes composed of a large number of objects having complex geometricdescriptions. Our technique relies upon a scene descriptionin which objects are represented as multiresolution meshes. We performa constrained optimization at each frame to choose the resolutionof each potentially visible object that generates the best qualityimage while meeting timing constraints. The technique providessmooth level-of-detail control and aims at guaranteeing a uniform,bounded frame rate even for widely changing viewing conditions.The optimization algorithm is independent from the particular datastructure used to represent multiresolution meshes. The only requirementsare the ability to represent a mesh with an arbitrarynumber of triangles and to traverse a mesh structure at an arbitraryresolution in a short predictable time. A data structure satisfyingthese criteria is described and experimental results are discussed.
IEEE Visualization	VizCraft: A Multidimensional Visualization Tool for Aircraft Configuration Design.	Amit Goel,Chuck Baker,Clifford A. Shaffer,Bernard Grossman,Raphael T. Haftka,William H. Mason,Layne T. Watson	1999	We describe a visualization tool to aid aircraft designers duringthe conceptual design stage. The conceptual design for anaircraft is defined by a vector of 10-30 parameters. The goalis to find a vector that minimizes an objective function whilemeeting a series of constraints. VizCraft integrates the simulationcode that evaluates the design with visualizations foranalyzing the design individually or in contrast to other designs.VizCraft allows the designer to easily switch betweenthe view of a design in the form of a parameter set, and avisualization of the corresponding aircraft. The user can easilysee which, if any, constraints are violated. VizCraft alsoallows the user to view a database of designs using parallelcoordinates.
IEEE Visualization	Visualizing Simulated Room Fires.	Jayesh Govindarajan,Matthew O. Ward,Jonathan Barnett	1999	Recent advances in fire science and computer modeling of fires allow scientists to predict fire growth and spread through structures. In this paper we describe a variety of visualizations of simulated room fires for use by both fire protection engineers and fire suppression personnel. We also introduce the concept of fuzzy visualization, which results from the superposition of data from several separate simulations into a single visualization.
IEEE Visualization	An Interactive Framework for Visualizing Foreign Currency Exchange Options.	Donna L. Gresh,Bernice E. Rogowitz,M. S. Tignor,E. J. Mayland	1999	Analyzing options is a complex, multi-variate process. Option behaviordepends on a variety of market conditions which vary overthe time course of the option. The goal of this project is to providean interactive visual environment which allows the analyst toexplore these complex interactions, and to select and construct specificviews for communicating information to non-analysts (e.g.,marketing managers and customers).In this paper we describe an environment for exploring 2- and3-dimensional representations of options data, dynamically varyingparameters, examining how multi-variate relationships developover time, and exploring the likelihood of the development of differentoutcomes over the life of the option. We also demonstratehow this tool has been used by analysts to communicate to non-analystshow particular options no longer deliver the behavior theywere originally intended to provide.
IEEE Visualization	Efficient Compression of Non-Manifold Polygonal Meshes.	André Guéziec,Frank Bossen,Gabriel Taubin,Cláudio T. Silva	1999	We present a method for compressing non-manifold polygonalmeshes, i.e. polygonal meshes with singularities, which occur veryfrequently in the real-world. Most efficient polygonal compressionmethods currently available are restricted to a manifold mesh: theyrequire a conversion process, and fail to retrieve the original modelconnectivity after decompression. The present method works byconverting the original model to a manifold model, encoding themanifold model using an existing mesh compression technique, andclustering, or stitching together during the decompression processvertices that were duplicated earlier to faithfully recover the originalconnectivity. This paper focuses on efficiently encoding and decodingthe stitching information. By separating connectivity fromgeometry and properties, the method avoids encoding vertices (andproperties bound to vertices) multiple times; thus a reduction of thesize of the bit-stream of about 10% is obtained compared with encodingthe model as a manifold.
IEEE Visualization	Tetrahedral Mesh Compression with the Cut-Border Machine.	Stefan Gumhold,Stefan Guthe,Wolfgang Straßer	1999	In recent years, substantial progress has been achieved in the areaof volume visualization on irregular grids, which is mainly basedon tetrahedral meshes. Even moderately fine tetrahedral meshesconsume several mega-bytes of storage. For archivation and transmissioncompression algorithms are essential. In scientific applicationslossless compression schemes are of primary interest. Thispaper introduces a new lossless compression scheme for the connectivityof tetrahedral meshes. Our technique can handle all tetrahedralmeshes in three dimensional euclidean space even with nonmanifold border. We present compression and decompression algorithmswhich consume for reasonable meshes linear time in thenumber of tetrahedra. The connectivity is compressed to less than2.4 bits per tetrahedron for all measured meshes. Thus a tetrahedralmesh can almost be reduced to the vertex coordinates, whichconsume in a common representation about one quarter of the totalstorage space.We complete our work with solutions for the compression of vertexcoordinates and additional attributes, which might be attachedto the mesh.
IEEE Visualization	Animating Wrinkles on Clothes.	Sunil Hadap,Endre Bangerter,Pascal Volino,Nadia Magnenat-Thalmann	1999	This paper describes a method to simulate realistic wrinkles onclothes without fine mesh and large computational overheads. Clothhas very little in-plane deformations, as most of the deformationscome from buckling. This can be looked at as area conservationproperty of cloth. The area conservation formulation of the methodmodulates the user defined wrinkle pattern, based on deformationof individual triangle. The methodology facilitates use of smallin-plane deformation stiffnesses and a coarse mesh for the numericalsimulation, this makes cloth simulation fast and robust. Moreover,the ability to design wrinkles (even on generalized deformablemodels) makes this method versatile for synthetic image generation.The method inspired from cloth wrinkling problem, being geometricin nature, can be extended to other wrinkling phenomena.
IEEE Visualization	New Quadric Metric for Simplifying Meshes with Appearance Attributes.	Hugues Hoppe	1999	Complex triangle meshes arise naturally in many areas of computergraphics and visualization. Previous work has shown that a quadricerror metric allows fast and accurate geometric simplification ofmeshes. This quadric approach was recently generalized to handlemeshes with appearance attributes. In this paper we present an improvedquadric error metric for simplifying meshes with attributes.The new metric, based on geometric correspondence in 3D, requiresless storage, evaluates more quickly, and results in more accuratesimplified meshes.Meshes often have attribute discontinuities, such as surfacecreases and material boundaries, which require multiple attributevectors per vertex. We show that a wedge-based mesh data structurecaptures such discontinuities efficiently and permits simultaneousoptimization of these multiple attribute vectors. In addition to thenew quadric metric, we experiment with two techniques proposedin geometric simplification, memoryless simplification and volumepreservation, and show that both of these are beneficial within thequadric framework. The new scheme is demonstrated on a varietyof meshes with colors and normals.
IEEE Visualization	Construction of Vector Field Hierarchies.	Bjørn Heckel,Gunther H. Weber,Bernd Hamann,Kenneth I. Joy	1999	We present a method for the hierarchical representation of vectorfields. Our approach is based on iterative refinement using clusteringand principal component analysis. The input to our algorithmis a discrete set of points with associated vectors. The algorithmgenerates a top-down segmentation of the discrete field by splittingclusters of points. We measure the error of the various approximationlevels by measuring the discrepancy between streamlines generatedby the original discrete field and its approximations based onmuch smaller discrete data sets. Our method assumes no particularstructure of the field, nor does it require any topological connectivityinformation. It is possible to generate multiresolution representationsof vector fields using this approach.
IEEE Visualization	A Distributed Graphics System for Large Tiled Displays.	Greg Humphreys,Pat Hanrahan	1999	"Recent interest in large displays has led to renewed development of tiled displays, which are comprised of several individual displays arranged in an array and used as one large logical display. Stanford's Interactive Mural is an example of such a display, using an overlapping four by two array of projectors that back-project onto a diffuse screen to form a 6' by 2' display area with a resolution of over 60 dpi. Writing software to make effective use of the large display space is a challenge because normal window system interaction metaphors break down. One promising approach is to switch to immersive applications; another approach, the one we are investigating, is to emulate office, conference room or studio environments which use the space to display a collection of visual material to support group activities.In this paper we describe a virtual graphics system that is designed to support multiple simultaneous rendering streams from both local and remote sites. The system abstracts the physical number of computers, graphics subsystems and projectors used to createthe display. We provide performance measurements to show that the system scales well and thus supports a variety of different hardware configurations. The system is also interesting because it uses transparent layers, instead of windows, to manage the screen."
IEEE Visualization	Accelerating 3D Convolution Using Graphics Hardware.	Matthias Hopf,Thomas Ertl	1999	Many volume filtering operations used for image enhancement,data processing or feature detection can be written in terms of three-dimensionalconvolutions. It is not possible to yield interactiveframe rates on todays hardware when applying such convolutionson volume data using software filter routines. As modern graphicsworkstations have the ability to render two-dimensional convolutedimages to the frame buffer, this feature can be used to acceleratethe process significantly. This way generic 3D convolutioncan be added as a powerful tool in interactive volume visualizationtoolkits.
IEEE Visualization	Image Graphs - A Novel Approach to Visual Data Exploration.	Kwan-Liu Ma	1999	For types of data visualization where the cost of producing images is high, and the relationship between the rendering parameters and the image produced is less than obvious, a visual representation of the exploration process can make the process more efficient and effective. Image graphs represent not only the results but also the process of data visualization. Each node in an image graph consists of an image and the corresponding visualization parameters used to produce it. Each edge in a graph shows the change in rendering parameters between the two nodes it connects. Image graphs are not just static representations: users can interact with a graph to review a previous visualization session or to perform new rendering. Operations which cause changes in rendering parameters can propagate through the graph. The user can take advantage of the information in image graphs to understand how certain parameter changes affect visualization results. Users can also share image graphs to streamline the process of collaborative visualization. We have implemented a volume visualization system using the image graph interface, and our examples in the paper come from this application.
IEEE Visualization	Multiresolution Techniques for Interactive Texture-Based Volume Visualization.	Eric LaMar,Bernd Hamann,Kenneth I. Joy	1999	We present a multiresolution technique for interactive texture-basedvolume visualization of very large data sets. This method uses anadaptive scheme that renders the volume in a region-of-interest ata high resolution and the volume away from this region at progressivelylower resolutions. The algorithm is based on the segmentationof texture space into an octree, where the leaves of the tree definethe original data and the internal nodes define lower-resolutionversions. Rendering is done adaptively by selecting high-resolutioncells close to a center of attention and low-resolution cells awayfrom this area. We limit the artifacts introduced by this method bymodifying the transfer functions in the lower-resolution data setsand utilizing spherical shells as a proxy geometry. It is possibleto use this technique to produce viewpoint-dependent renderings ofvery large data sets.
IEEE Visualization	"Automation or Interaction: What's Best for Big Data?"	David N. Kenwright,David C. Banks,Steve Bryson,Robert Haimes,Robert van Liere,Samuel P. Uselton	1999	"Automation or Interaction: What's Best for Big Data?"
IEEE Visualization	Hue-Balls and Lit-Tensors for Direct Volume Rendering of Diffusion Tensor Fields.	Gordon L. Kindlmann,David M. Weinstein	1999	"With the development of magnetic resonance imaging techniquesfor acquiring diffusion tensor data from biological tissue, visualizationof tensor data has become a new research focus. The diffusiontensor describes the directional dependence of water molecules'diffusion and can be represented by a three-by-three symmetricmatrix. Visualization of second-order tensor fields is difficult becausethe data values have many degrees of freedom. Existing visualizationtechniques are best at portraying the tensor's propertiesover a two-dimensional field, or over a small subset of locationswithin a three-dimensional field. A means of visualizing theglobal structure in measured diffusion tensor data is needed. Wepropose the use of direct volume rendering, with novel approachesfor the tensors' coloring, lighting, and opacity assignment. Hue-ballsuse a two-dimensional colormap on the unit sphere to illustratethe tensor's action as a linear operator. Lit-tensors providea lighting model for tensors which includes as special cases bothlit-lines (from streamline vector visualization) and standard Phongsurface lighting. Together with an opacity assignment based on anovel two-dimensional barycentric space of anisotropy, these methodsare shown to produce informative renderings of measured diffusiontensor data from the human brain."
IEEE Visualization	Visualizing Multivalued Data from 2D Incompressible Flows Using Concepts from Painting.	Robert M. Kirby,H. Marmanis,David H. Laidlaw	1999	We present a new visualization method for 2d flows which allowsus to combine multiple data values in an image for simultaneousviewing. We utilize concepts from oil painting, art, and design asintroduced in [1] to examine problems within fluid mechanics. Weuse a combination of discrete and continuous visual elements arrangedin multiple layers to visually represent the data. The representationsare inspired by the brush strokes artists apply in layers tocreate an oil painting. We display commonly visualized quantitiessuch as velocity and vorticity together with three additional mathematicallyderived quantities: the rate of strain tensor (defined insection 4), and the turbulent charge and turbulent current (definedin section 5). We describe the motivation for simultaneously examiningthese quantities and use the motivation to guide our choiceof visual representation for each particular quantity. We present visualizationsof three flow examples and observations concerningsome of the physical relationships made apparent by the simultaneousdisplay technique that we employed.
IEEE Visualization	A Multi-Threaded Streaming Pipeline Architecture for Large Structured Data Sets.	C. Charles Law,Ken Martin,William J. Schroeder,Joshua Temkin	1999	Computer simulation and digital measuring systems arenow generating data of unprecedented size. The size of datais becoming so large that conventional visualization toolsare incapable of processing it, which is in turn is impactingthe effectiveness of computational tools. In this paper wedescribe an object-oriented architecture that addresses thisproblem by automatically breaking data into pieces, andthen processes the data piece-by-piece within a pipeline offilters. The piece size is user specified and can be controlledto eliminate the need for swapping (i.e., relying on virtualmemory). In addition, because piece size can be controlled,any size problem can be run on any size computer, at theexpense of extra computational time. Furthermore, piecesare automatically broken into sub-pieces and each pieceassigned to a different thread for parallel processing. Thispaper includes numerical performance studies and referencesto the source code which is freely available on theWeb.
IEEE Visualization	Rendering on a Budget: A Framework for Time-Critical Rendering.	James T. Klosowski,Cláudio T. Silva	1999	"We present a technique for optimizing the rendering of high-depthcomplexity scenes. Prioritized-Layered Projection(PLP) does this by rendering an estimation of the visible setfor each frame. The novelty in our work lies in the fact thatwe do not explicitly compute visible sets. Instead, our work isbased on computing on demand a priority order for the polygonsthat maximizes the likelihood of rendering visible polygonsbefore occluded ones for any given scene. Given a fixedbudget, e.g. time or number of triangles, our rendering algorithmmakes sure to render geometry respecting the computedpriority.There are two main steps to our technique: (1) anoccupancy-based tessellation of space; and (2) a solidity-basedtraversal algorithm. PLP works by computing an occupancy-basedtessellation of space, which tends to have smaller cellswhere there are more geometric primitives, e.g., polygons. Inthis spatial tessellation, each cell is assigned a solidity value,which is directly proportional to its likelihood of occludingother cells. In its simplest form, a cell's solidity value is directlyproportional to the number of polygons contained withinit. During our traversal algorithm, cells are marked for projection,and the geometric primitives contained within them actuallyrendered. The traversal algorithm makes use of the cells'solidity, and other view-dependent information to determinethe ordering in which to project cells. By tailoring the traversalalgorithm to the occupancy-based tessellation, we can achievevery good frame rates with low preprocessing and renderingcosts.In this paper, we describe our technique and its implementationin detail. We also provide experimental evidence of itsperformance and briefly discuss extensions of our algorithm."
IEEE Visualization	Collapsing Flow Topology Using Area Metrics.	Wim C. de Leeuw,Robert van Liere	1999	Visualization of topological information of a vector field can provideuseful information on the structure of the field. However, inturbulent flows standard critical point visualization will result in acluttered image which is difficult to interpret. This paper presents atechnique for collapsing topologies. The governing idea is to classifythe importance of the critical points in the topology. By onlydisplaying the more important critical points, a simplified depictionof the topology can be provided. Flow consistency is maintainedwhen collapsing the topology, resulting in a visualization whichis consistent with the original topology. We apply the collapsingtopology technique to a turbulent flow field.
IEEE Visualization	Mixing Translucent Polygons with Volumes.	Kevin Kreeger,Arie E. Kaufman	1999	We present an algorithm which renders opaque and/or translucentpolygons embedded within volumetric data. The processing occurssuch that all objects are composited in the correct order, by renderingthin slabs of the translucent polygons between volume slicesusing slice-order volume rendering. We implemented our algorithmwith OpenGL on current general-purpose graphics systems. We discussour system implementation, speed and image quality, as wellas the renderings of several mixed scenes.
IEEE Visualization	Visualizing Large-Scale Telecommunication Networks and Services.	Eleftherios Koutsofios,Stephen C. North,Russell Truscott,Daniel A. Keim	1999	Visualizing Large-Scale Telecommunication Networks and Services.
IEEE Visualization	Splatting Without the Blur.	Klaus Mueller,Torsten Möller,Roger Crawfis	1999	Splatting is a volume rendering algorithm that combines efficientvolume projection with a sparse data representation: Onlyvoxels that have values inside the iso-range need to be considered,and these voxels can be projected via efficient rasterizationschemes. In splatting, each projected voxel is represented as aradially symmetric interpolation kernel, equivalent to a fuzzy ball.Projecting such a basis function leaves a fuzzy impression, called afootprint or splat, on the screen. Splatting traditionally classifiesand shades the voxels prior to projection, and thus each voxel footprintis weighted by the assigned voxel color and opacity. Projectingthese fuzzy color balls provides a uniform screen image forhomogeneous object regions, but leads to a blurry appearance ofobject edges. The latter is clearly undesirable, especially when theview is zoomed on the object. In this work, we manipulate the renderingpipeline of splatting by performing the classification andshading process after the voxels have been projected onto thescreen. In this way, volume contributions outside the iso-rangenever affect the image. Since shading requires gradients, we notonly splat the density volume, using regular splats, but we alsoproject the gradient volume, using gradient splats. However, alternativeto gradient splats, we can also compute the gradients on theprojection plane, using central differencing. This latter schemecuts the number of footprint rasterization by a factor of four, sinceonly the voxel densities have to be projected. Our new method rendersobjects with crisp edges and well-preserved surface detail.Added overhead is the calculation of the screen gradients and theper-pixel shading. Both of these operations, however, may be performedusing fast techniques employing lookup tables.
IEEE Visualization	Detecting Null Alleles with Vasarely Charts.	Carl Manaster,Elizabeth Nanthakumar,Phillip A. Morin	1999	Microsatellite genotypes can have problems that are difficult todetect with existing tools. One such problem is null alleles. Thispaper presents a new visualization tool that helps to find andcharacterize these errors. The paper explains how the tool is usedto analyze groups of genotypes and proposes other possible uses.
IEEE Visualization	A Computer Animation Representing the Molecular Events of G Protein-Coupled Receptor Activation.	Zoya Maslak,Douglas J. Steel,Robert J. McDermott	1999	The molecular events involved in the activation of G protein-coupled receptors, represent a fundamental biochemical process. These events were selected for animation because the mechanism involves both a ligand-receptor conformational shape change, and an enzyme-substrate conformational shape change. Expository animation brought this biochemical process to life.
IEEE Visualization	Cracking the Cracking Problem with Coons Patches.	Gregory M. Nielson,Dave J. Holliday,Tom Roxborough	1999	We present a new approach to solving the crackingproblem. The cracking problem arises in many contexts inscientific visualization and computer graphics modeling wherethere is need for an approximation based upon domaindecomposition that is fine in certain regions and coarse inothers. This includes surface rendering, approximation ofimages and multiresolution terrain visualization. In general,algorithms based upon adaptive refinement strategies mustdeal with this problem. The new approach presented here issimple and general. It is based upon the use of a triangularCoons patch. Both the basic idea of using a triangular Coonspatch in this context and the particular Coons patch that isused constitute the novel contributions of this paper.
IEEE Visualization	Enabling Calssification and Shading for 3D Texture Mapping Based Volume Rendering Using OpenGL and Extensions.	Michael Meißner,Ulrich Hoffmann,Wolfgang Straßer	1999	We present a new technique which enables direct volume rendering based on 3D texture mapping hardware, enabling shading as well as classification of the interpolated data. Our technique supports accurate lighting for a one directional light source, semi-transparent classification, and correct blending. To circumvent the limitations of one general classification, we introduce multiple classification spaces which are very valuable to understand the visualized data, and even mandatory to comprehensively grasp the 3D relationship of different materials present in the volumetric data. Furthermore, we illustrate how multiple classification spaces can be realized using existing graphics hardware.In contrast to previously reported algorithms, our technique is capable of performing all the above mentioned tasks within the graphics pipeline. Therefore, it is very efficient: The three dimensional texture needs to be stored only once and no load is put onto the CPU. Besides using standard OpenGL functionality, we exploit advanced per pixel operations and make use of available OpenGL extensions.
IEEE Visualization	Large Field Visualization with Demand-Driven Calculation.	Patrick J. Moran,Chris Henze	1999	We present a system designed for the interactive definition and visualizationof fields derived from large data sets: the Demand-DrivenVisualizer (DDV). The system allows the user to write arbitrary expressionsto define new fields, and then apply a variety of visualizationtechniques to the result. Expressions can include differentialoperators and numerous other built-in functions. Determination offield values, both in space and in time, is directed automatically bythe demands of the visualization techniques. The payoff of followinga demand-driven design philosophy throughout the visualizationsystem becomes particularly evident when working with largetime-series data, where the costs of eager evaluation alternativescan be prohibitive.
IEEE Visualization	Detecting Vortical Phenomena in Vector Data by Medium-Scale Correlation.	Hans-Georg Pagendarm,B. Henne,M. Rutten	1999	The detection of vortical phenomena in vector data is one ofthe key issues in many technical applications, in particularin flow visualization. Many existing approaches rely onpurely local evaluation of the vector data. In order to overcomethe limits of a local approach we choose to combine alocal method with a correlation of a pre-defined genericvortex with the data in a medium-scale region. Two differentconcepts of a generic vortices were tested on various sets offlow velocity vector data. The approach is not limited to thetwo generic patterns suggested here. The method was foundto successfully detect vortices in cases were other methodsfail.
IEEE Visualization	Implant Sprays: Compression of Progressive Tetrahedral Mesh Connectivity.	Renato Pajarola,Jarek Rossignac,Andrzej Szymczak	1999	Irregular tetrahedral meshes, which are popular in many engineeringand scientific applications, often contain a large numberof vertices. A mesh of V vertices and T tetrahedra requires 48·Vbits or less to store the vertex coordinates, 4·T·log 2 (V) bits to storethe tetrahedra-vertex incidence relations, also called connectivityinformation, and k·V bits to store the k-bit value samples associatedwith the vertices. Given that T is 5 to 7 times larger than Vand that V often exceeds 32 3 , the storage space required for theconnectivity is larger than 300·V bits and thus dominates the overallstorage cost. Our implants spray compression approachintroduced in this paper reduces this cost to about 30·V bits or less-- a 10:1 compression ratio. Furthermore, implant spray supportsthe progressive refinement of a crude model through a series ofvertex-splits operations.
IEEE Visualization	"The ""Parallel Vectors"" Operator - A Vector Field Visualization Primitive."	Ronald Peikert,Martin Roth	1999	In this paper we propose an elementary operation on a pair ofvector fields as a building block for defining and computingglobal line-type features of vector or scalar fields. While usualfeature definitions often are procedural and therefore implicit,our operator allows precise mathematical definitions. It canserve as a basis for comparing feature definitions and for reuseof algorithms and implementations. Applications focus on vortexcore methods.
IEEE Visualization	Why the PC Will Be the Most Pervasive Visualization Platform in 2001.	Hanspeter Pfister,Michel Cox,Peter N. Glaskowsky,William E. Lorensen,Richard Greco	1999	Why the PC Will Be the Most Pervasive Visualization Platform in 2001.
IEEE Visualization	Anisotropic Nonlinear Diffusion in Flow Visualization.	Tobias Preußer,Martin Rumpf	1999	Vector field visualization is an important topic in scientific visualization.Its aim is to graphically represent field data in an intuitivelyunderstandable and precise way. Here a new approach based onanisotropic nonlinear diffusion is introduced. It enables an easy perceptionof flowdata and serves as an appropriate scale space methodfor the visualization of complicated flow pattern. The approachis closely related to nonlinear diffusion methods in image analysiswhere images are smoothed while still retaining and enhancingedges. Here an initial noisy image is smoothed along streamlines,whereas the image is sharpened in the orthogonal direction. Themethod is based on a continuous model and requires the solutionof a parabolic PDE problem. It is discretized only in the final implementationalstep. Therefore, many important qualitative aspectscan already be discussed on a continuous level. Applications areshown in 2D and 3D and the provisions for flow segmentation areoutlined.
IEEE Visualization	Multi-Projector Displays Using Camera-Based Registration.	Ramesh Raskar,Michael S. Brown,Ruigang Yang,Wei-Chao Chen,Greg Welch,Herman Towles,W. Brent Seales,Henry Fuchs	1999	Conventional projector-based display systems are typically designed around precise and regular configurations of projectors and display surfaces. While this results in rendering simplicity and speed, it also means painstaking construction and ongoing maintenance. In previously published work, we introduced a vision of projector-based displays constructed from a collection of casually-arranged projectors and display surfaces.In this paper, we present exible yet practical methods for realizing this vision, enabling low-cost mega-pixel display systems with large physical dimensions, higher resolution, or both. The techniques afford new opportunities to build personal 3D visualization systems in offices, conference rooms, theaters, or even your living room. As a demonstration of the simplicity and effectiveness of the methods that we continue to perfect, we show in the included video that a10-year old child can construct and calibrate a two-camera, two-projector, head-tracked display system, all in about 15 minutes.
IEEE Visualization	Interacitve Exploration of Volume Line Integral Convolution Based on 3D-Texture Mapping.	Christof Rezk-Salama,Peter Hastreiter,Christian Teitzel,Thomas Ertl	1999	Line integral convolution (LIC) is an effective technique for visualizing vector fields. The application of LIC to 3D flow fields has yet been limited by difficulties to efficiently display and animate the resulting 3D-images. Texture-based volume rendering allows interactive visualization and manipulation of 3D-LIC textures. In order to ensure the comprehensive and convenient exploration of flow fields, we suggest interactive functionality including transfer functions and different clipping mechanisms. Thereby, we efficiently substitute the calculation of LIC based on sparse noise textures and show the convenient visual access of interior structures. Further on, we introduce two approaches for animating static 3D-flow fields without the computational expense and the immense memory requirements for pre-computed 3D-textures and without loss of interactivity. This is achieved by using a single 3D-LIC texture and a set of time surfaces as clipping geometries. In our first approach we use the clipping geometry to pre-compute a special 3D-LIC texture that can be animated by time-dependent color tables. Our second approach uses time volumes to actually clip the 3D-LIC volume interactively during rasterization. Additionally, several examples demonstrate the value of our strategy in practice.
IEEE Visualization	Structured Spatial Domain Image and Data Comparison Metrics.	Nivedita Sahasrabudhe,John E. West,Raghu Machiraju,Mark Janus	1999	Often, images or datasets have to be compared, to facilitatechoices of visualization and simulation parameters respectively.Common comparison techniques include side-by-side viewing andjuxtaposition, in order to facilitate visual verification of verisimilitude.In this paper, we propose quantitative techniques whichaccentuate differences in images and datasets. The comparison isenabled through a collection of partial metrics which, essentially,measure the lack of correlation between the datasets or imagesbeing compared. That is, they attempt to expose and measure theextent of the inherent structures in the difference between imagesor datasets. Besides yielding numerical attributes, the metrics alsoproduce images, which can visually highlight differences. Our metricsare simple to compute and operate in the spatial domain. Wedemonstrate the effectiveness of our metrics through examples forcomparing images and datasets.
IEEE Visualization	Visualizing Planar Vector Fields with Normal Component Using Line Integral Convolution.	Gerik Scheuermann,Holger Burbach,Hans Hagen	1999	"We present a method for visualizing three dimensional vector fieldswhich are defined on a two dimensional manifold only. These vectorfields do exist in real application, as we show by an example ofan optical measuring instrument which can gauge the displacementat the surface of a mechanical part. The general idea is to computeLIC textures in the manifold's tangent space and to deformthe manifold according to the normal information. The resultingLIC texture is mapped onto the deformed manifold and is renderedas a three dimensional scene. Due to the light's reflection on thedeformed manifold, one can interactively explore the result of thedeformation."
IEEE Visualization	C1-Interpolation for Vector Field Topology Visualization.	Gerik Scheuermann,Xavier Tricoche,Hans Hagen	1999	"An application of C¹ scalar interpolation for 2D vector field topologyvisualization is presented. Powell-Sabin and Nielson interpolantsare considered which both make use of Nielson's MinimumNorm Network for the precomputation of the derivatives in our implementation.A comparison of their results to the commonly usedlinear interpolant underlines their significant improvement of singularitylocation and topological skeleton depiction. Evalution isbased upon the processing of polynomial vector fields with knowntopology containing higher order singularities."
IEEE Visualization	Interacitve Visualization of Fluid Dynamics Simulations in Locally Refined Cartesian Grids.	Martin Schulz,Frank Reck,Wolf Bartelheimer,Thomas Ertl	1999	This work presents interactive flow visualization techniques specificallyadapted for PowerFLOW, a lattice-based CFD code fromthe EXA corporation. Their Digital Physics fluid simulationtechnique is performed on a hierarchy of locally refined cartesiangrids with a fine voxel resolution in areas of interesting flow features.Among other applications the PowerFLOW solver is usedfor aerodynamic simulations in car body development where the advantagesof automatic grid generation from CAD models is of greatinterest. In a joint project with BMWand EXA we are developing avisualization tool which incorporates virtual reality techniques forthe interactive exploration of the large scalar and vector data sets.In this paper we describe the specific data structures and interpolationtechniques and we report on fast particle tracing taking intoaccount collisions with the car body geometry. An OpenGL Optimizerbased implementation allows for the inspection of the flowwith particle probes and slice probes at interactive frame rates.
IEEE Visualization	Interactive Lens Visualization Techniques.	Christopher D. Shaw,James A. Hall,David S. Ebert,D. Aaron Roberts	1999	This paper describes new techniques for minimally immersive visualization of 3D scalar and vector fields, and visualization of document corpora. In our glyph-based visualization system, the user interacts with the 3D volume of glyphs using a pair of button-enhanced 3D position and orientation trackers. The user may also examine the volume using an interactive lens, which is a rectangle that slices through the 3Dvolume and displaysscalar information on its surface. Alens allows the display of scalar data in the 3Dvolume using a contour diagram, and a texture-based volume rendering.
IEEE Visualization	A Fast Volume Rendering Algorithm for Time-Varying Fields Using a Time-Space Partitioning (TSP) Tree.	Han-Wei Shen,Ling-Jan Chiang,Kwan-Liu Ma	1999	This paper presents a fast volume rendering algorithm for time-varyingfields. We propose a new data structure, called Time-SpacePartitioning (TSP) tree, that can effectively capture both the spatialand the temporal coherence from a time-varying field. Usingthe proposed data structure, the rendering speed is substantially improved.In addition, our data structure helps to maintain the memoryaccess locality and to provide the sparse data traversal so thatour algorithm becomes suitable for large-scale out-of-core applications.Finally, our algorithm allows flexible error control for boththe temporal and the spatial coherence so that a trade-off betweenimage quality and rendering speed is possible. We demonstratethe utility and speed of our algorithm with data from several time-varyingCFD simulations. Our rendering algorithm can achievesubstantial speedup while the storage space overhead for the TSPtree is kept at a minimum.
IEEE Visualization	Keynote Address: Data and Visualization Corridors.	Paul Smith,John Van Rosendale	1999	Keynote Address: Data and Visualization Corridors.
IEEE Visualization	Capstone Address: ActiveSpaces - The Access Grid, Active Mural and Advanced Visualization Systems.	Rick L. Stevens	1999	Capstone Address: ActiveSpaces - The Access Grid, Active Mural and Advanced Visualization Systems.
IEEE Visualization	Isosurface Extraction in Time-varying Fields Using a Temporal Branch-on-Ned Tree (T-BON).	Philip M. Sutton,Charles D. Hansen	1999	The Temporal Branch-on-Need Tree (T-BON) extends the three-dimensionalbranch-on-need octree for time-varying isosurface extraction.At each time step, only those portions of the tree and datanecessary to construct the current isosurface are read from disk.This algorithm can thus exploit the temporal locality of the isosurfaceand, as a geometric technique, spatial locality between cellsin order to improve performance. Experimental results demonstratethe performance gained and memory overhead saved usingthis technique.
IEEE Visualization	Visualization Needs More Visual Design!	J. Edward Swan II,Theresa-Marie Rhyne,David H. Laidlaw,Tamara Munzner,Victoria Interrante	1999	Visualization Needs More Visual Design!
IEEE Visualization	Simplified Representation of Vector Fields.	Alexandru Telea,Jarke J. van Wijk	1999	Vector field visualization remains a difficult task. Although manylocal and global visualization methods for vector fields such as flowdata exist, they usually require extensive user experience on settingthe visualization parameters in order to produce images communicatingthe desired insight. We present a visualization method thatproduces simplified but suggestive images of the vector field automatically,based on a hierarchical clustering of the input data. Theresulting clusters are then visualized with straight or curved arrowicons. The presented method has a few parameters with which userscan produce various simplified vector field visualizations that communicatedifferent insights on the vector data.
IEEE Visualization	PLIC: Briding the Gap Between Streamlines and LIC.	Vivek Verma,David T. Kao,Alex Pang	1999	This paper explores mapping strategies for generating LIC-like imagesfrom streamlines and streamline-like images from LIC. Themain contribution of this paper is a technique which we call pseudo-LICor PLIC. By adjusting a small set of key parameters, PLIC cangenerate flow visualizations that span the spectrum of streamline-liketo LIC-like images. Among the advantages of PLIC are: imagequality comparable with LIC, performance speedup over LIC,use of a template texture that is independent of the size of the flowfield, handles the problem of multiple streamlines occupying thesame pixel in image space, reduced aliasing, applicability to timevarying data sets, and variable speed animation.
IEEE Visualization	High Performance Presence-Accelerated Ray Casting.	Ming Wan,Arie E. Kaufman,Steve Bryson	1999	We present a novel presence acceleration for volumetric ray casting.A highly accurate estimation for object presence is obtainedby projecting all grid cells associated with the object boundaryon the image plane. Memory space and access time are reducedby run-length encoding of the boundary cells, while boundary cellprojection time is reduced by exploiting projection templates andmultiresolution volumes. Efforts have also been made towards afast perspective projection as well as interactive classification. Wefurther present task partitioning schemes for effective parallelizationof both boundary cell projection and ray traversal procedures.Good load balancing has been reached by taking full advantage ofboth the optimizations in the serial rendering algorithm and shared-memoryarchitecture. Our experimental results on a 16-processorSGI Power Challenge have shown interactive rendering rates for256³ volumetric data sets at 10 -- 30 Hz. This paper describes thetheory and implementation of our algorithm, and shows its superiorityover the shear-warp factorization approach.
IEEE Visualization	Volume Rendering Based Interactive Navigation within the Human Colon.	Ming Wan,Qingyu Tang,Arie E. Kaufman,Zhengrong Liang,Mark Wax	1999	"We present an interactive navigation system for virtualcolonoscopy, which is based solely on high performance volumerendering. Previous colonic navigation systems haveemployed either a surface rendering or a Z-buffer-assisted volumerendering method that depends on the surface rendering results.Our method is a fast direct volume rendering technique thatexploits distance information stored in the potential field of thecamera control model, and is parallelized on a multiprocessor.Experiments have been conducted on both a simulated pipe andpatients' data sets acquired with a CT scanner."
IEEE Visualization	Tensorlines: Advection-Diffusion Based Propagation through Diffusion Tensor Fields.	David M. Weinstein,Gordon L. Kindlmann,Eric C. Lundberg	1999	Tracking linear features through tensor field datasets is an open researchproblem with widespread utility in medical and engineeringdisciplines. Existing tracking methods, which consider only thepreferred local diffusion direction as they propagate, fail to accuratelyfollow features as they enter regions of local complexity. Thisshortcoming is a result of partial voluming; that is, voxels in theseregions often contain contributions from multiple features. Thesecombined contributions result in ambiguities when deciding localprimary feature orientation based solely on the preferred diffusiondirection. In this paper, we introduce a novel feature extractionmethod, which we term tensorline propagation. Our method resolvesthe above ambiguity by incorporating information about thenearby orientation of the feature, as well as the anisotropic classificationof the local tensor. The nearby orientation informationis added in the spirit of an advection term in a standard diffusion-basedpropagation technique, and has the effect of stabilizing thetracking. To demonstrate the efficacy of tensorlines, we apply thismethod to the neuroscience problem of tracking white-matter bundleswithin the brain.
IEEE Visualization	A Framework for Assisted Exploration with Collaboration.	Eric A. Wernert,Andrew J. Hanson	1999	"We approach the problem of exploring a virtual space by exploitingpositional and camera-model constraints on navigation to provideextra assistance that focuses the user's explorational wanderingson the task objectives. Our specific design incorporates not onlytask-based constraints on the viewer's location, gaze, and viewingparameters, but also a personal guide that serves two importantfunctions: keeping the user oriented in the navigation space, andpointing to interesting subject areas as they are approached. Theguide's cues may be ignored by continuing in motion, but if the userstops, the gaze shifts automatically toward whatever the guide wasinterested in. This design has the serendipitous feature that it automaticallyincorporates a nested collaborative paradigm simply byallowing any given viewer to be seen as the guide of one or moreviewers following behind; the leading automated guide (we tend toselect a guide dog for this avatar) can remind the leading live humanguide of interesting sites to point out, while each real humancollaborator down the chain has some choices about whether to followthe local leader's hints. We have chosen VRML as our initialdevelopment medium primarily because of its portability, and wehave implemented a variety of natural modes for leading and collaborating, including ways for collaborators to attach to and detach from a particular leader."
IEEE Visualization	Design and Implementation of an Immersive Geoscience Toolkit.	Christophe Winkler,Fabien Bosquet,Xavier Cavin,Jean-Claude Paul	1999	Design and Implementation of an Immersive Geoscience Toolkit.
IEEE Visualization	Digital Design of a Surgical Simulator for Interventional Magnetic Resonance Imaging.	Terry S. Yoo,Penny Rheingans	1999	We present the design of a simulator for a prototype interventionalmagnetic resonance imaging scanner. This MRI scanner isintegrated with an operating theater, enabling new techniques inminimally invasive surgery. The simulator is designed with athreefold purpose: (1) to provide a rehearsal apparatus forpracticing and modifying conventional procedures for use in themagnetic environment, (2) to serve as a visualization workstationfor procedure planning and previewing as well as a post-operativereview, and (3) to form the foundation of a laboratory workbenchfor the development of new surgical tools and procedures forminimally invasive surgery. The simulator incorporates pre-operativedata, either MRI or CT exams, as well as data fromcommercial surgical planning systems. Dynamic control of thesimulation and interactive display of preoperative data in lieu ofintra-operative data is handled via an opto-electronic trackingsystem. The resulting system is contributing insights into how bestto perform visualization for this new surgical environment.
IEEE Visualization	IEEE Visualization 1999, 24-29 October 1999, San Francisco, CA, USA, Proceedings.	David S. Ebert,Markus H. Gross,Bernd Hamann	1999	IEEE Visualization 1999, 24-29 October 1999, San Francisco, CA, USA, Proceedings.
IEEE Trans. Vis. Comput. Graph.	The Ball-Pivoting Algorithm for Surface Reconstruction.	Fausto Bernardini,Joshua Mittleman,Holly E. Rushmeier,Cláudio T. Silva,Gabriel Taubin	1999	"The Ball-Pivoting Algorithm (BPA) computes a triangle mesh interpolating a given point cloud. Typically, the points are surface samples acquired with multiple range scans of an object. The principle of the BPA is very simple: Three points form a triangle if a ball of a user-specified radius $\rho$ touches them without containing any other point. Starting with a seed triangle, the ball pivots around an edge (i.e., it revolves around the edge while keeping in contact with the edge's endpoints) until it touches another point, forming another triangle. The process continues until all reachable edges have been tried, and then starts from another seed triangle, until all points have been considered. The process can then be repeated with a ball of larger radius to handle uneven sampling densities. We applied the BPA to datasets of millions of points representing actual scans of complex 3D objects. The relatively small amount of memory required by the BPA, its time efficiency, and the quality of the results obtained compare favorably with existing techniques."
IEEE Trans. Vis. Comput. Graph.	Fast Iterative Refinement of Articulated Solid Dynamics.	François Faure	1999	A new dynamics algorithm for articulated solid animation is presented. It provides enhancements of computational efficiency and accuracy control with respect to previous solutions. Iterative refinement allows us to perform interactive animations which could be only computed off-line using previous methods. The efficiency results from managing two sets of constraints associated with the kinematic graph, and proceeding in two steps. First, the acyclic constraints are solved in linear time. An iterative process then reduces the closed loop errors while maintaining the acyclic constraints. This allows the user to efficiently trade off accuracy for computation time. We analyze the complexity and investigate practical efficiency compared with other approaches. In contrast with previous research, we present a single method which is computationally efficient for acyclic bodies as well as for mesh-like bodies. The accuracy control is provided by the iterative improvement performed by the algorithm and also from the existence of two constraint priority levels induced by the method. Used in conjunction with a robust integration scheme, this new algorithm allows the interactive animation of scenes containing a few thousand geometric constraints, including closed loops. It has been successfully applied to real-time simulations.
IEEE Trans. Vis. Comput. Graph.	Using Linked Volumes to Model Object Collisions, Deformation, Cutting, Carving, and Joining.	Sarah F. Frisken Gibson	1999	"In volume graphics, objects are represented by arrays or clusters of sampled 3D data. A volumetric object representation is necessary in computer modeling whenever interior structure affects an object's behavior or appearance. However, existing volumetric representations are not sufficient for modeling the behaviors expected in applications such as surgical simulation, where interactions between both rigid and deformable objects and the cutting, tearing, and repairing of soft tissues must be modeled in real time. Three-dimensional voxel arrays lack the sense of connectivity needed for complex object deformation, while finite element models and mass-spring systems require substantially reduced geometric resolution for interactivity and they can not be easily cut or carved interactively. This paper discusses a linked volume representation that enables physically realistic modeling of object interactions such as: collision detection, collision response, 3D object deformation, and interactive object modification by carving, cutting, tearing, and joining. The paper presents a set of algorithms that allow interactive manipulation of linked volumes that have more than an order of magnitude more elements and considerably more flexibility than existing methods. Implementation details, results from timing tests, and measurements of material behavior are presented."
IEEE Trans. Vis. Comput. Graph.	Real-Time Elastic Deformations of Soft Tissues for Surgery Simulation.	Stephane Cotin,Hervé Delingette,Nicholas Ayache	1999	In this paper, we describe a new method for surgery simulation including a volumetric model built from medical images and an elastic modeling of the deformations. The physical model is based on elasticity theory which suitably links the shape of deformable bodies and the forces associated with the deformation. A real-time computation of the deformation is possible thanks to a preprocessing of elementary deformations derived from a finite element method. This method has been implemented in a system including a force feedback device and a collision detection algorithm. The simulator works in real-time with a high resolution liver model.
IEEE Trans. Vis. Comput. Graph.	Locally Toleranced Surface Simplification.	André Guéziec	1999	We present a technique for simplifying a triangulated surface. Simplifying consists of approximating the surface with another surface of lower triangle count. Our algorithm can preserve the volume of a solid to within machine accuracy; it favors the creation of near-equilateral triangles. We develop novel methods for reporting and representing a bound to the approximation error between a simplified surface and the original, and respecting a variable tolerance across the surface. A different positive error value is reported at each vertex. By linearly blending the error values in between vertices, we define a volume of space, called the error volume, as the union of balls of linearly varying radii. The error volume is built dynamically as the simplification progresses, on top of preexisting error volumes that it contains. We also build a tolerance volume to forbid simplification errors exceeding a local tolerance. The information necessary to compute error values is local to the star of a vertex; accordingly, the complexity of the algorithm is either linear or in $O(n \log n)$ in the original number of surface edges, depending on the variant. We extend the mechanisms of error and tolerance volumes to preserve during simplification scalar and vector attributes associated with surface vertices. Assuming a linear variation across triangles, error and tolerance volumes are defined in the same fashion as for positional error. For normals, a corrective term is applied to the error measured at the vertices to compensate for nonlinearities.
IEEE Trans. Vis. Comput. Graph.	"Editor's Note."	Hans Hagen	1999	"Editor's Note."
IEEE Trans. Vis. Comput. Graph.	On a Construction of a Hierarchy of Best Linear Spline Approximations Using Repeated Bisection.	Bernd Hamann,Benjamin W. Jordan,David F. Wiley	1999	We present a method for the construction of hierarchies of single-valued functions in one, two, and three variables. The input to our method is a coarse decomposition of the compact domain of a function in the form of an interval (univariate case), triangles (bivariate case), or tetrahedra (trivariate case). We compute best linear spline approximations, understood in an integral least squares sense, for functions defined over such triangulations and refine triangulations using repeated bisection. This requires the identification of the interval (triangle, tetrahedron) with largest error and splitting it into two intervals (triangles, tetrahedra). Each bisection step requires the recomputation of all spline coefficients due to the global nature of the best approximation problem. Nevertheless, this can be done efficiently by bisecting multiple intervals (triangles, tetrahedra) in one step and by reducing the bandwidths of the matrices resulting from the normal equations.
IEEE Trans. Vis. Comput. Graph.	"Errata: Corrections to ""On a Construction of a Hierarchy of Best Linear Spline Approximations Using Repeated Bisection""."	Bernd Hamann,Benjamin W. Jordan,David F. Wiley	1999	"Errata: Corrections to ""On a Construction of a Hierarchy of Best Linear Spline Approximations Using Repeated Bisection""."
IEEE Trans. Vis. Comput. Graph.	Large Datasets at a Glance: Combining Textures and Colors in Scientific Visualization.	Christopher G. Healey,James T. Enns	1999	This paper presents a new method for using texture and color to visualize multivariate data elements arranged on an underlying height field. We combine simple texture patterns with perceptually uniform colors to increase the number of attribute values we can display simultaneously. Our technique builds multicolored perceptual texture elements (or pexels) to represent each data element. Attribute values encoded in an element are used to vary the appearance of its pexel. Texture and color patterns that form when the pexels are displayed can be used to rapidly and accurately explore the dataset. Our pexels are built by varying three separate texture dimensions: height, density, and regularity. Results from computer graphics, computer vision, and human visual psychophysics have identified these dimensions as important for the formation of perceptual texture patterns. The pexels are colored using a selection technique that controls color distance, linear separation, and color category. Proper use of these criteria guarantees colors that are equally distinguishable from one another. We describe a set of controlled experiments that demonstrate the effectiveness of our texture dimensions and color selection criteria. We then discuss new work that studies how texture and color can be used simultaneously in a single display. Our results show that variations of height and density have no effect on color segmentation, but that random color patterns can interfere with texture segmentation. As the difficulty of the visual detection task increases, so too does the amount of color on texture interference increase. We conclude by demonstrating the applicability of our approach to a real-world problem, the tracking of typhoon conditions in Southeast Asia.
IEEE Trans. Vis. Comput. Graph.	Fast Projection-Based Ray-Casting Algorithm for Rendering Curvilinear Volumes.	Lichan Hong,Arie E. Kaufman	1999	We present an efficient and robust ray-casting algorithm for directly rendering a curvilinear volume of arbitrarily-shaped cells. By projecting cell-faces onto the image plane, we have effectively addressed three critical steps of the ray-casting process, namely finding the entry cell-faces for a ray, traversing along the ray from one cell to another, and reconstructing data values at the ray/cell-face intersections. Our algorithm significantly reduces rendering time, alleviates memory space consumption, and overcomes the conventional limitation requiring cells to be convex. Application of this algorithm to several commonly used curvilinear data sets has produced a favorable performance when compared with recently reported algorithms.
IEEE Trans. Vis. Comput. Graph.	Evaluation of Memoryless Simplification.	Peter Lindstrom,Greg Turk	1999	This paper investigates the effectiveness of the Memoryless Simplification approach described by Lindstrom and Turk [14]. Like many polygon simplification methods, this approach reduces the number of triangles in a model by performing a sequence of edge collapses. It differs from most recent methods, however, in that it does not retain a history of the geometry of the original model during simplification. We present numerical comparisons showing that the memoryless method results in smaller mean distance measures than many published techniques that retain geometric history. We compare a number of different vertex placement schemes for an edge collapse in order to identify the aspects of the Memoryless Simplification that are responsible for its high level of fidelity. We also evaluate simplification of models with boundaries, and we show how the memoryless method may be tuned to trade between manifold and boundary fidelity. We found that the memoryless approach yields consistently low mean errors when measured by the Metro mesh comparison tool. In addition to using complex models for the evaluations, we also perform comparisons using a sphere and portions of a sphere. These simple surfaces turn out to match the simplification behaviors for the more complex models that we used.
IEEE Trans. Vis. Comput. Graph.	Feature Extraction of Separation and Attachment Lines.	David N. Kenwright,Chris Henze,Creon Levit	1999	Separation and attachment lines are topologically significant curves that exist on 2D surfaces in 3D vector fields. Two algorithms are presented, one point-based and one element-based, that extract separation and attachment lines using eigenvalue analysis of a locally linear function. Unlike prior techniques based on piecewise numerical integration, these algorithms use robust analytical tests that can be applied independently to any point in a vector field. The feature extraction is fully automatic and suited to the analysis of large-scale numerical simulations. The strengths and weaknesses of the two algorithms are evaluated using analytic vector fields and also results from computational fluid dynamics (CFD) simulations. We show that both algorithms detect open separation lines¿a type of separation that is not captured by conventional vector field topology algorithms.
IEEE Trans. Vis. Comput. Graph.	Partitioning 3D Surface Meshes Using Watershed Segmentation.	Alan P. Mangan,Ross T. Whitaker	1999	This paper describes a method for partitioning 3D surface meshes into useful segments. The proposed method generalizes morphological watersheds, an image segmentation technique, to 3D surfaces. This surface segmentation uses the total curvature of the surface as an indication of region boundaries. The surface is segmented into patches, where each patch has a relatively consistent curvature throughout, and is bounded by areas of higher, or drastically different, curvature. This algorithm has applications for a variety of important problems in visualization and geometrical modeling including 3D feature extraction, mesh reduction, texture mapping 3D surfaces, and computer aided design.
IEEE Trans. Vis. Comput. Graph.	High-Quality Splatting on Rectilinear Grids with Efficient Culling of Occluded Voxels.	Klaus Mueller,Naeem Shareef,Jian Huang,Roger Crawfis	1999	Splatting is a popular volume rendering algorithm that pairs good image quality with an efficient volume projection scheme. The current axis-aligned sheet-buffer approach, however, bears certain inaccuracies. The effect of these is less noticeable in still images, but clearly revealed in animated viewing, where disturbing popping of object brightness occurs at certain view angle transitions. In previous work, we presented a new variant of sheet-buffered splatting in which the compositing sheets are oriented parallel to the image plane. This scheme not only eliminates the condition for popping, but also produces images of higher quality. In this paper, we summarize this new paradigm and extend it in a number of ways. We devise a new solution to render rectilinear grids of equivalent cost to the traditional approach that treats the anisotropic volume as being warped into a cubic grid. This enables us to use the usual radially symmetric kernels, which can be projected without inaccuracies. Next, current splatting approaches necessitate the projection of all voxels in the iso-interval(s), although only a subset of these voxels may eventually be visible in the final image. To eliminate these wasteful computations we propose a novel front-to-back approach that employs an occlusion map to determine if a splat contributes to the image before it is projected, thus skipping occluded splats. Additional measures are presented for further speedups. In addition, we present an efficient list-based volume traversal scheme that facilitates the quick modification of transfer functions and iso-values.
IEEE Trans. Vis. Comput. Graph.	Tools for Computing Tangent Curves for Linearly Varying Vector Fields over Tetrahedral Domains.	Gregory M. Nielson,Il-Hong Jung	1999	We present some very efficient and accurate methods for computing tangent curves for three-dimensional flows. Our methods work directly in physical coordinates, eliminating the usual need to switch back and forth with computational coordinates. Unlike conventional methods, such as Runge-Kutta, for computing tangent curves which give only approximations, our methods produce exact values based upon piece-wise linear variation over a tetrahedrization of the domain of interest. We use barycentric coordinates in order to efficiently track cell-to-cell movement of the tangent curves.
IEEE Trans. Vis. Comput. Graph.	A Rule-Based Interactive Behavioral Animation System for Humanoids.	Hansrudi Noser,Daniel Thalmann	1999	We present a versatile, behavioral, and rule-based animation system that includes autonomous humanoid actors whose behavior is based on synthetic sensors that are used for perceiving the virtual environment. We combine the following in a consistent approach: L-systems, a behavioral production rule system; a particle system; an acoustic environment model, including a speech recognition module; a virtual life network; and a humanoid library. Together, these systems create a real-time-structured virtual environment that both high-level autonomous humanoids and interactive users can easily share.
IEEE Trans. Vis. Comput. Graph.	Adaptive Projection Operators in Multiresolution Scientific Visualization.	Mario Ohlberger,Martin Rumpf	1999	Adaptive Projection Operators in Multiresolution Scientific Visualization.
IEEE Trans. Vis. Comput. Graph.	Interactive Ray Tracing for Volume Visualization.	Steven G. Parker,Michael Parker,Yarden Livnat,Peter-Pike J. Sloan,Charles D. Hansen,Peter Shirley	1999	We present a brute-force ray tracing system for interactive volume visualization. The system runs on a conventional (distributed) shared-memory multiprocessor machine. For each pixel we trace a ray through a volume to compute the color for that pixel. Although this method has high intrinsic computational cost, its simplicity and scalability make it ideal for large datasets on current high-end parallel systems. To gain efficiency several optimizations are used including a volume bricking scheme and a shallow data hierarchy. These optimizations are used in three separate visualization algorithms: isosurfacing of rectilinear data, isosurfacing of unstructured data, and maximum-intensity projection on rectilinear data. The system runs interactively (i.e., several frames per second) on an SGI Reality Monster. The graphics capabilities of the Reality Monster are used only for display of the final color image.
IEEE Trans. Vis. Comput. Graph.	Ray Casting Architectures for Volume Visualization.	Harvey Ray,Hanspeter Pfister,Deborah Silver,Todd A. Cook	1999	--Real-time visualization of large volume datasets demands high performance computation, pushing the storage, processing, and data communication requirements to the limits of current technology. General purpose parallel processors have been used to visualize moderate size datasets at interactive frame rates; however, the cost and size of these supercomputers inhibits the widespread use for real-time visualization. This paper surveys several special purpose architectures that seek to render volumes at interactive rates. These specialized visualization accelerators have cost, performance, and size advantages over parallel processors. All architectures implement ray casting using parallel and pipelined hardware. We introduce a new metric that normalizes performance to compare these architectures. The architectures included in this survey are VOGUE, VIRIM, Array Based Ray Casting, EM-Cube, and VIZARD II. We also discuss future applications of special purpose accelerators.Index Terms--Volume rendering, interactive, ray casting, parallel architecture, VOGUE, VIRIM, array based ray casting, EM-Cube, VIZARD II.
IEEE Trans. Vis. Comput. Graph.	Edgebreaker: Connectivity Compression for Triangle Meshes.	Jarek Rossignac	1999	"Edgebreaker is a simple scheme for compressing the triangle/vertex incidence graphs (sometimes called connectivity or topology) of three-dimensional triangle meshes. Edgebreaker improves upon the storage required by previously reported schemes, most of which can guarantee only an O(t log(t)) storage cost for the incidence graph of a mesh of t triangles. Edgebreaker requires at most 2t bits for any mesh homeomorphic to a sphere and supports fully general meshes by using additional storage per handle and hole. For large meshes, entropy coding yields less than 1.5 bits per triangle. Edgebreaker's compression and decompression processes perform identical traversals of the mesh from one triangle to an adjacent one. At each stage, compression produces an op-code describing the topological relation between the current triangle and the boundary of the remaining part of the mesh. Decompression uses these op-codes to reconstruct the entire incidence graph. Because Edgebreaker's compression and decompression are independent of the vertex locations, they may be combined with a variety of vertex-compressing techniques that exploit topological information about the mesh to better estimate vertex locations. Edgebreaker may be used to compress the connectivity of an entire mesh bounding a 3D polyhedron or the connectivity of a triangulated surface patch whose boundary need not be encoded. The paper also offers a comparative survey of the rapidly growing field of geometric compression."
IEEE Trans. Vis. Comput. Graph.	"Guest Editor's Introduction: Special Section on Visualization."	Holly E. Rushmeier	1999	"Guest Editor's Introduction: Special Section on Visualization."
IEEE Trans. Vis. Comput. Graph.	Acquiring a Radiance Distribution to Superimpose Virtual Objects onto a Real Scene.	Imari Sato,Yoichi Sato,Katsushi Ikeuchi	1999	This paper describes a new method for superimposing virtual objects with correct shadings onto an image of a real scene. Unlike the previously proposed methods, our method can measure a radiance distribution of a real scene automatically and use it for superimposing virtual objects appropriately onto a real scene. First, a geometric model of the scene is constructed from a pair of omnidirectional images by using an omnidirectional stereo algorithm. Then, radiance of the scene is computed from a sequence of omnidirectional images taken with different shutter speeds and mapped onto the constructed geometric model. The radiance distribution mapped onto the geometric model is used for rendering virtual objects superimposed onto the scene image. As a result, even for a complex radiance distribution, our method can superimpose virtual objects with convincing shadings and shadows cast onto the real scene. We successfully tested the proposed method by using real images to show its effectiveness.
IEEE Trans. Vis. Comput. Graph.	Alias-Free Voxelization of Geometric Objects.	Milos Srámek,Arie E. Kaufman	1999	We introduce a new concept for alias-free voxelization of geometric objects based on a Voxelization model (V-model). The V-model of an object is its representation in three-dimensional continuous space by a trivariate density function. This function is sampled during the voxelization and the resulting values are stored in a volume buffer. This concept enables us to study general issues of sampling and rendering separately from object specific design issues. It provides us with a possibility to design such V-models, which are correct from the point of view of both the sampling and rendering, thus leading to both alias-free volumetric representation and alias-free rendered images. We performed numerous experiments with different combinations of V-models and reconstruction techniques. We have shown that the V-model with a Gaussian surface density profile combined with tricubic interpolation and Gabor derivative reconstruction outperforms the previously published technique with a linear density profile. This enables higher fidelity of images rendered from volume data due to increased sharpness of edges and thinner surface patches.
IEEE Trans. Vis. Comput. Graph.	Dynamic Scene Occlusion Culling.	Oded Sudarsky,Craig Gotsman	1999	"Large, complex 3D scenes are best rendered in an output-sensitive way, i.e., in time largely independent of the entire scene model's complexity. Occlusion culling is one of the key techniques for output-sensitive rendering. We generalize existing occlusion culling algorithms, intended for static scenes, to handle dynamic scenes having numerous moving objects. The data structure used by an occlusion culling method is updated to reflect the objects' possible positions. To avoid updating the structure for every dynamic object at each frame, a temporal bounding volume (TBV) is created for each occluded dynamic object, using some known constraints on the object's motion. The TBV is inserted into the structure instead of the object. Subsequently, the object is ignored as long as the TBV is occluded and guaranteed to contain the object. The generalized algorithms' rendering time is linearly affected only by the scene's visible parts, not by hidden parts or by occluded dynamic objects. Our techniques also save communications in distributed graphic systems, e.g., multiuser virtual environments, by eliminating update messages for hidden dynamic objects. We demonstrate the adaptation of two occlusion culling algorithms to dynamic scenes: hierarchical Z-buffering and BSP tree projection."
IEEE Trans. Vis. Comput. Graph.	Simplification of Tetrahedral Meshes with Error Bounds.	Issac J. Trotts,Bernd Hamann,Kenneth I. Joy	1999	We present a method for the construction of multiple levels of tetrahedral meshes approximating a trivariate scalar-valued function at different levels of detail. Starting with an initial, high-resolution triangulation of a three-dimensional region, we construct coarser representation levels by collapsing edges of the mesh. Each triangulation defines a linear spline function, where the function values associated with the vertices are the spline coefficients. Error bounds are stored for individual tetrahedra and are updated as the mesh is simplified. Two algorithms are presented that simplify the mesh within prescribed error bounds. Each algorithm treats simplification on the mesh boundary. The result is a hierarchical data description suited for efficient visualization of large data sets at varying levels of detail.
IEEE Trans. Vis. Comput. Graph.	Efficient Skeletonization of Volumetric Objects.	Yong Zhou,Arthur W. Toga	1999	Skeletonization promises to become a powerful tool for compact shape description, path planning, and other applications. However, current techniques can seldom efficiently process real, complicated 3D data sets, such as MRI and CT data of human organs. In this paper, we present an efficient voxel-coding based algorithm for skeletonization of 3D voxelized objects. The skeletons are interpreted as connected centerlines, consisting of sequences of medial points of consecutive clusters. These centerlines are initially extracted as paths of voxels, followed by medial point replacement, refinement, smoothness, and connection operations. The voxel-coding techniques have been proposed for each of these operations in a uniform and systematic fashion. In addition to preserving basic connectivity and centeredness, the algorithm is characterized by straightforward computation, no sensitivity to object boundary complexity, explicit extraction of ready-to-parameterize and branch-controlled skeletons, and efficient object hole detection. These issues are rarely discussed in traditional methods. A range of 3D medical MRI and CT data sets were used for testing the algorithm, demonstrating itsutility.
CVPR	Segmentation with Invisible Keying Signal.	Moshe Ben-Ezra	2000	Segmentation with Invisible Keying Signal.
CVPR	A Region Extraction Method Using Multiple Active Contour Models.	Toru Abe,Yuki Matsuzawa	2000	A Region Extraction Method Using Multiple Active Contour Models.
CVPR	ROR: Rejection of Outliers by Rotations in Stereo Matching.	Amit Adam,Ehud Rivlin,Ilan Shimshoni	2000	ROR: Rejection of Outliers by Rotations in Stereo Matching.
CVPR	Limits on Super-Resolution and How to Break Them.	Simon Baker,Takeo Kanade	2000	Nearly all super-resolution algorithms are based on the fundamental constraints that the super-resolution image should generate the low resolution input images when appropriately warped and down-sampled to model the image formation process. (These reconstruction constraints are normally combined with some form of smoothness prior to regularize their solution.) In the first part of this paper, we derive a sequence of analytical results which show that the reconstruction constraints provide less and less useful information as the magnification factor increases. We also validate these results empirically and show that, for large enough magnification factors, any smoothness prior leads to overly smooth results with very little high-frequency content (however, many low resolution input images are used). In the second part of this paper, we propose a super-resolution algorithm that uses a different kind of constraint, in addition to the reconstruction constraints. The algorithm attempts to recognize local features in the low-resolution images and then enhances their resolution in an appropriate manner. We call such a super-resolution algorithm a hallucination or recogstruction algorithm. We tried our hallucination algorithm on two different data sets, frontal images of faces and printed Roman text. We obtained significantly better results than existing reconstruction-based algorithms, both qualitatively and in terms of RMS pixel error.
CVPR	Efficient Query Modification for Image Retrieval.	Gaurav Aggarwal,Sugata Ghosal,Pradeep K. Dubey	2000	Efficient Query Modification for Image Retrieval.
CVPR	A Neural Optimization Framework for Zoom Lens Camera Calibration.	Moumen T. Ahmed,Aly A. Farag	2000	A Neural Optimization Framework for Zoom Lens Camera Calibration.
CVPR	Probabilistic vs. Geometric Similarity Measures for Image Retrieval.	Selim Aksoy,Robert M. Haralick	2000	Probabilistic vs. Geometric Similarity Measures for Image Retrieval.
CVPR	Estimating Anthropometry and Pose from a Single Image.	Carlos Barron,Ioannis A. Kakadiaris	2000	Estimating Anthropometry and Pose from a Single Image.
CVPR	Recursive Estimation of Motion and Planar Structure.	Jonathan Alon,Stan Sclaroff	2000	"Abstract A specialized formulation of Azarbayejani and Pentland''s framework for recursive recovery of motion, structure and focal length from feature correspondences tracked through an image sequence is presented. The specialized formulation addresses the case where all tracked points lie on a plane. This planarity constraint reduces the dimension of the original state vector, and consequently the number of feature points needed to estimate the state. Experiments with synthetic data and real imagery illustrate the system performance. The experiments confirm that the specialized formulation provides improved accuracy, stability to observation noise, and rate of convergence in estimation for the case where the tracked points lie on a plane."
CVPR	Reliable Feature Matching across Widely Separated Views.	Adam Baumberg	2000	Reliable Feature Matching across Widely Separated Views.
CVPR	Automatic Recovery of Relative Camera Rotations for Urban Scenes.	Matthew E. Antone,Seth J. Teller	2000	Automatic Recovery of Relative Camera Rotations for Urban Scenes.
CVPR	A Novel Approach to Depth Ordering in Monocular Image Sequences.	Lothar Bergen,Fernand Meyer	2000	A Novel Approach to Depth Ordering in Monocular Image Sequences.
CVPR	Visual Venture: Investigations with Images and Videos for Middle School Education.	Lisa M. G. Brown	2000	Visual Venture: Investigations with Images and Videos for Middle School Education.
CVPR	A Generalized Optical Flow Constraint and its Physical Interpretation.	Dominique Béréziat,Isabelle Herlin,Laurent Younes	2000	A Generalized Optical Flow Constraint and its Physical Interpretation.
CVPR	Distortion-Invariant Recognition via Jittered Querie.	Michael C. Burl	2000	Distortion-Invariant Recognition via Jittered Querie.
CVPR	A Real-Time System for Epipolar Geometry and Ego-Motion Estimation.	Mårten Björkman,Jan-Olof Eklundh	2000	A Real-Time System for Epipolar Geometry and Ego-Motion Estimation.
CVPR	On the Space Requirements of Indexing 3D Models from 2D Perspective Images.	Vincenzo Caglioti	2000	On the Space Requirements of Indexing 3D Models from 2D Perspective Images.
CVPR	Object Recognition for an Intelligent Room.	Richard Campbell,John Krumm	2000	Object Recognition for an Intelligent Room.
CVPR	A Feature-Based Technique for Joint, Linear Estimation of High-Order Image-to-Mosaic Transformations: Application to Mosaicing the Curved Human Retina.	Ali Can,Charles V. Stewart,Badrinath Roysam,Howard L. Tanenbaum	2000	A Feature-Based Technique for Joint, Linear Estimation of High-Order Image-to-Mosaic Transformations: Application to Mosaicing the Curved Human Retina.
CVPR	A Bayesian Framework for Radar Shape-from-Shading.	Adrian G. Bors,Edwin R. Hancock,Richard C. Wilson	2000	A Bayesian Framework for Radar Shape-from-Shading.
CVPR	Point Pattern Matching with Robust Spectral Correspondence.	Marco Carcassoni,Edwin R. Hancock	2000	Point Pattern Matching with Robust Spectral Correspondence.
CVPR	Catadioptric Self-Calibration.	Sing Bing Kang	2000	Catadioptric Self-Calibration.
CVPR	Curve and Surface Reconstruction from Regular and Non-Regular Point Sets.	Edmond Boyer,Sylvain Petitjean	2000	Curve and Surface Reconstruction from Regular and Non-Regular Point Sets.
CVPR	Adaptive Bayesian Recognition in Tracking Rigid Objects.	Yuri Boykov,Daniel P. Huttenlocher	2000	Adaptive Bayesian Recognition in Tracking Rigid Objects.
CVPR	Step towards Sequence-to-Sequence Alignment.	Yaron Caspi,Michal Irani	2000	Step towards Sequence-to-Sequence Alignment.
CVPR	"Intel's Computer Vision Library: Applications in Calibration, Stereo, Segmentation, Tracking, Gesture, Face and Object Recognition."	Gary R. Bradski,Vadim Pisarevsky	2000	"Intel's Computer Vision Library: Applications in Calibration, Stereo, Segmentation, Tracking, Gesture, Face and Object Recognition."
CVPR	Nearest Neighbor Search Using Additive Binary Tree.	Sung-Hyuk Cha,Sargur N. Srihari	2000	Nearest Neighbor Search Using Additive Binary Tree.
CVPR	Parallel Projections for Stereo Reconstruction.	Jinxiang Chai,Heung-Yeung Shum	2000	Parallel Projections for Stereo Reconstruction.
CVPR	Recovering Non-Rigid 3D Shape from Image Streams.	Christoph Bregler,Aaron Hertzmann,Henning Biermann	2000	"This paper addresses the problem of recovering 3D non-rigid shape models from image sequences. For example, given a video recording of a talking person, we would like to estimate a 3D model of the lips and the full head and its internal modes of variation. Many solutions that recover 3D shape from 2D image sequences have been proposed; these so-called structure-from-motion techniques usually assume that the 3D object is rigid. For example, Tomasi and Kanade''s factorization technique is based on a rigid shape matrix, which produces a tracking matrix of rank 3 under orthographic projection. We propose a novel technique based on a non-rigid model, where the 3D shape in each frame is a linear combination of a set of basis shapes. Under this model, the tracking matrix is of higher rank, and can be factored in a three step process to yield to pose, configuration and shape. We demonstrate this simple but effective algorithm on video sequences of speaking people. We were able to recover 3D non-rigid facial models with high accuracy."
CVPR	Robust Visual Recognition of Color Images Rozenn Dahyot, Pierre.	Pierre Charbonnier,Fabrice Heitz	2000	Robust Visual Recognition of Color Images Rozenn Dahyot, Pierre.
CVPR	In Search of Illumination Invariants.	Hansen F. Chen,Peter N. Belhumeur,David W. Jacobs	2000	In Search of Illumination Invariants.
CVPR	Wide Area Camera Calibration Using Virtual Calibration Object.	Xing Chen,James Davis,Philipp Slusallek	2000	Wide Area Camera Calibration Using Virtual Calibration Object.
CVPR	A Real Time System for Robust 3D Voxel Reconstruction of Human Motions.	German K. M. Cheung,Takeo Kanade,Jean-Yves Bouguet,Mark Holler	2000	A Real Time System for Robust 3D Voxel Reconstruction of Human Motions.
CVPR	Sensor Networked Mobile Robotics.	Thomas Christensen,Mads Noergaard,Claus B. Madsen,Adam Hoover	2000	Sensor Networked Mobile Robotics.
CVPR	A New Algorithm for Non-Rigid Point Matching.	Haili Chui,Anand Rangarajan	2000	A New Algorithm for Non-Rigid Point Matching.
CVPR	Real-Time Tracking of Non-Rigid Objects Using Mean Shift.	Dorin Comaniciu,Visvanathan Ramesh,Peter Meer	2000	Real-Time Tracking of Non-Rigid Objects Using Mean Shift.
CVPR	A New Regularized Approach for Contour Morphing.	Ge Cong,Bahram Parvin	2000	A New Regularized Approach for Contour Morphing.
CVPR	Articulated-Pose Estimation Using Brightness and Depth-Constancy Constraints.	Michele Covell,Ali Rahimi,Michael Harville,Trevor Darrell	2000	Articulated-Pose Estimation Using Brightness and Depth-Constancy Constraints.
CVPR	obust Periodic Motion and Motion Symmetry Detection.	Ross Cutler,Larry S. Davis	2000	obust Periodic Motion and Motion Symmetry Detection.
CVPR	Computing 3D Object Parts from Similarities among Object Views.	Florin Cutzu	2000	Computing 3D Object Parts from Similarities among Object Views.
CVPR	Categorical Representation and Recognition of Oscillatory Motion Patterns.	James W. Davis,Whitman Richards,Aaron F. Bobick	2000	Categorical Representation and Recognition of Oscillatory Motion Patterns.
CVPR	Structure from Motion without Correspondence.	Frank Dellaert,Steven M. Seitz,Charles E. Thorpe,Sebastian Thrun	2000	Structure from Motion without Correspondence.
CVPR	Articulated Body Motion Capture by Annealed Particle Filtering.	Jonathan Deutscher,Andrew Blake,Ian D. Reid	2000	Articulated Body Motion Capture by Annealed Particle Filtering.
CVPR	Robust and Efficient Skeletal Graphs.	Pavel Dimitrov,Carlos Phillips,Kaleem Siddiqi	2000	Robust and Efficient Skeletal Graphs.
CVPR	Adaptive Metric nearest Neighbor Classification.	Carlotta Domeniconi,Dimitrios Gunopulos,Jing Peng	2000	Adaptive Metric nearest Neighbor Classification.
CVPR	Detecting Dynamic Behavior in Compressed Fingerprint Videos: Distortion.	Chitra Dorai,Nalini K. Ratha,Ruud M. Bolle	2000	Detecting Dynamic Behavior in Compressed Fingerprint Videos: Distortion.
CVPR	Matching Images with Different Resolutions.	Yves Dufournaud,Cordelia Schmid,Radu Horaud	2000	Matching Images with Different Resolutions.
CVPR	A General Scheme for Training and Optimization of the Grenander Deformable Template Model.	Rune Fisker,Nette Schultz,Jens Michael Carstensen,Nicolae Duta	2000	A General Scheme for Training and Optimization of the Grenander Deformable Template Model.
CVPR	Towards Reliable Fusion of Segmented Surface Descriptions.	Daniel Fischer,Peter Kohlhepp	2000	Towards Reliable Fusion of Segmented Surface Descriptions.
CVPR	Detecting Binocular Half-Occlusions: Empirical Comparisons of Four Approaches.	Geoffrey Egnal,Richard P. Wildes	2000	Binocular half-occlusion points are those that are visible in one of the two views provided by a binocular imaging system. Due to their importance in binocular matching as well as, subsequent interpretation tasks, a number of approaches have been developed for dealing with such points. In the current paper, we consider five methods that explicitly detect half-occlusions and report on a more uniform comparison than has previously been performed. Taking a disparity image and its associated match goodness image as input, we generate images that show the half-occluded points in the underlying scene. We quantitatively and qualitatively compare these methods under a variety of conditions.
CVPR	Inconsistencies in Edge Detector Evaluation.	Lance A. Forbes,Bruce A. Draper	2000	Inconsistencies in Edge Detector Evaluation.
CVPR	Provably Fast Algorithms for Contour Tracking.	Daniel Freedman,Michael S. Brandstein	2000	Provably Fast Algorithms for Contour Tracking.
CVPR	Filling In Scenes by Propagating Probabilities through Layers and Into Appearance Models.	Brendan J. Frey	2000	Filling In Scenes by Propagating Probabilities through Layers and Into Appearance Models.
CVPR	A Combined Feature-Texture Similarity Measure for Face Alignment under.	Lixin Fan,Kah Kay Sung	2000	A Combined Feature-Texture Similarity Measure for Face Alignment under.
CVPR	Detection of Obstacles in the Flight Path of an Aircraft.	Tarak Gandhi,Mau-Tsuen Yang,Rangachar Kasturi,Octavia I. Camps,Lee D. Coraor,Jeffrey McCandless	2000	Detection of Obstacles in the Flight Path of an Aircraft.
CVPR	Efficient Matching of Pictorial Structures.	Pedro F. Felzenszwalb,Daniel P. Huttenlocher	2000	Efficient Matching of Pictorial Structures.
CVPR	Tomographic Reconstruction Using Curve Evolution.	Haihua Feng,David A. Castañon,William Clement Karl	2000	Tomographic Reconstruction Using Curve Evolution.
CVPR	Error Analysis of Background Adaption.	Xiang Gao,Terrance E. Boult,Frans Coetzee,Visvanathan Ramesh	2000	Error Analysis of Background Adaption.
CVPR	A Comparative Technique and Performance Results on Novel Learned Snakes in Two Dissimilar Medical Domains.	Samuel D. Fenster,John R. Kender	2000	A Comparative Technique and Performance Results on Novel Learned Snakes in Two Dissimilar Medical Domains.
CVPR	Perceptual Grouping and Segmentation by Stochastic Clustering.	Yoram Gdalyahu,Noam Shental,Daphna Weinshall	2000	Perceptual Grouping and Segmentation by Stochastic Clustering.
CVPR	Image Retrieval and Segmentation based on Color Invariants.	Jan-Mark Geusebroek,Dennis Koelma,Arnold W. M. Smeulders,Theo Gevers	2000	Image Retrieval and Segmentation based on Color Invariants.
CVPR	Detecting People in Cluttered Indoor Scenes.	Mi-Suen Lee	2000	Detecting People in Cluttered Indoor Scenes.
CVPR	Measurement of Color Invariants.	Jan-Mark Geusebroek,Arnold W. M. Smeulders,Rein van den Boomgaard	2000	Measurement of Color Invariants.
CVPR	Multi-View 3D Analysis with Applications for Augmented Reality and Enhanced Video Visualization.	Yanlin Guo,Steven C. Hsu,Supun Samarasekera,Harpreet S. Sawhney,Rakesh Kumar	2000	Multi-View 3D Analysis with Applications for Augmented Reality and Enhanced Video Visualization.
CVPR	A Formal Classification of 3D Medial Axis Points and Their Local Geometry.	Peter J. Giblin,Benjamin B. Kimia	2000	Abstract--This paper proposes a novel hypergraph skeletal representation for 3D shape based on a formal derivation ofthe generic structure of its medial axis. By classifying each skeletal point by its order of contact, we show that, generically, the medial axis consists of five types of points, which are then organized into sheets, curves, and points: 1) sheets (manifolds with boundary) which are the locus of bitangent spheres with regular tangency A_1^2 (A_k^n notation means n distinct k{\hbox{-}}{\rm{fold}} tangencies of the sphere of contact, as explained in the text); two types of curves, 2) the intersection curve of three sheets and the locus of centers of tritangent spheres, A_1^3, and 3) the boundary of sheets, which are the locus of centers of spheres whose radius equals the larger principal curvature, i.e., higher order contact A_3 points; and two types of points, 4) centers of quad-tangent spheres, A_1^4, and 5) centers of spheres with one regular tangency and one higher order tangency, A_1A_3. The geometry of the 3D medial axis thus consists of sheets (A_1^2) bounded by one type of curve (A_3) on their free end, which corresponds to ridges on the surface, and attached to two other sheets at another type of curve (A_1^3), which support a generalized cylinder description. The A_3 curves can only end in A_1A_3 points where they must meet an A_1^3 curve. The A_1^3 curves meet together in fours at an A_1^4 point. This formal result leads to a compact representation for 3D shape, referred to as the medial axis hypergraph representation consisting of nodes (A_1^4 and A_1 A_3 points), links between pairs of nodes (A_1^3 and A_3 curves) and hyperlinks between groups of links (A_1^2 sheets). The description of the local geometry at nodes by itself is sufficient to capture qualitative aspects of shapes, in analogy to 2D. We derive a pointwise reconstruction formula to reconstruct a surface from this medial axis hypergraph together with the radius function. Thus, this information completely characterizes 3D shape and lays the theoretical foundation for its use in recognition, morphing, design, and manipulation of shapes.
CVPR	A Boundary Extraction Method Based on Dual-T-Snakes and Dynamic Programming.	Gilson A. Giraldi,Edilberto Strauss,Antonio A. F. Oliveira	2000	A Boundary Extraction Method Based on Dual-T-Snakes and Dynamic Programming.
CVPR	Histogram Preserving Image Transformations.	Efstathios Hadjidemetriou,Michael D. Grossberg,Shree K. Nayar	2000	Histograms are used to analyze and index images. They have been found experimentally to have low sensitivity to certain types of image morphisms, for example, viewpoint changes and object deformations. The precise effect of these image morphisms on the histogram, however, has not been studied. In this work we derive the complete class of local transformations that preserve or scale the magnitude of the histogram of all images. We also derive a more general class of local transformations that preserve the histogram relative to a particular image. To achieve this, the transformations are represented as solutions to families of vector fields acting on the image. The local effect of fixed points of the fields on the histograms is also analyzed. The analytical results are verified with several examples. We also discuss several applications and the significance of these transformations for histogram indexing.
CVPR	Rectified Catadioptric Stereo Sensors.	Joshua Gluckman,Shree K. Nayar	2000	It has been previously shown how mirrors can be used to capture stereo images with a single camera, an approach termed catadioptric stereo. In this paper, we present novel catadioptric sensors that use mirrors to produce rectified stereo images. The scan-line correspondence of these images benefits real-time stereo by avoiding the computational cost and image degradation due to resampling when rectification is performed after image capture. First, we develop a theory which determines the number of mirrors that must be used and the constraints on those mirrors that must be satisfied to obtain rectified stereo images with a single camera. Then, we discuss in detail the use of both one and three mirrors. In addition, we show how the mirrors should be placed in order to minimize sensor size for a given baseline, an important design consideration. In order to understand the feasibility of building these sensors, we analyze rectification errors due to misplacement of the camera with respect to the mirrors.
CVPR	Reconstruction of a Scene with Multiple Linearly Moving Objects.	Mei Han,Takeo Kanade	2000	In this paper we describe an algorithm to recover the scene structure, the trajectories of the moving objects and the camera motion simultaneously given a monocular image sequence. The number of the moving objects is automatically detected without prior motion segmentation. Assuming that the objects are moving linearly with constant speeds, we propose a unified geometrical representation of the static scene and the moving objects. This representation enables the embedding of the motion constraints into the scene structure, which leads to a factorization-based algorithm. We also discuss solutions to the degenerate cases which can be automatically detected by the algorithm. Extension of the algorithm to weak perspective projections is presented as well. Experimental results on synthetic and real images show that the algorithm is reliable under noise.
CVPR	Stereo Based Gesture Recognition Invariant to 3D Pose and Lighting.	Radek Grzeszczuk,Gary R. Bradski,Michael H. Chu,Jean-Yves Bouguet	2000	Stereo Based Gesture Recognition Invariant to 3D Pose and Lighting.
CVPR	Fixed Topology Skeletons.	Polina Golland,W. Eric L. Grimson	2000	Fixed Topology Skeletons.
CVPR	Video Summarization Using Singular Value Decomposition.	Yihong Gong,Xin Liu	2000	Video Summarization Using Singular Value Decomposition.
CVPR	Detecting and Tracking Eyes by Using Their Physiological Properties, Dynamics, and Appearance.	Antonio Haro,Myron Flickner,Irfan A. Essa	2000	Detecting and Tracking Eyes by Using Their Physiological Properties, Dynamics, and Appearance.
CVPR	Reconstruction from Six-Point Sequences.	Richard I. Hartley,Nicolas Y. Dano	2000	Reconstruction from Six-Point Sequences.
CVPR	Stereo without Depth Search and Metric Calibration.	Hiroshi Hattori,Atsuto Maki	2000	Stereo without Depth Search and Metric Calibration.
CVPR	Computing Optical Flow with Physical Models of Brightness Variation.	Horst W. Haussecker,David J. Fleet	2000	Although most optical flow techniques presume brightness constancy, it is well-known that this constraint is often violated, producing poor estimates of image motion. This paper describes a generalized formulation of optical flow estimation based on models of brightness variations that are caused by time-dependent physical processes. These include changing surface orientation with respect to a directional illuminant, motion of the illuminant, and physical models of heat transport in infrared images. With these models, we simultaneously estimate the 2D image motion and the relevant physical parameters of the brightness change model. The estimation problem is formulated using total least squares (TLS), with confidence bounds on the parameters. Experiments in four domains, with both synthetic and natural inputs, show how this formulation produces superior estimates of the 2D image motion.
CVPR	Simultaneous Tracking and Verification via Sequential Posterior Estimation.	Baoxin Li,Rama Chellappa	2000	Simultaneous Tracking and Verification via Sequential Posterior Estimation.
CVPR	Statistical Modeling and Performance Characterization of a Real-Time Dual Camera Surveillance System.	Michael Greiffenhagen,Visvanathan Ramesh,Dorin Comaniciu,Heinrich Niemann	2000	Statistical Modeling and Performance Characterization of a Real-Time Dual Camera Surveillance System.
CVPR	A Robust and Efficient Motion Segmentation Based on Orthogonal Projection Matrix of Shape Space.	Naoyuki Ichimura	2000	A Robust and Efficient Motion Segmentation Based on Orthogonal Projection Matrix of Shape Space.
CVPR	An Energy-Based Framework for Dense 3D Registration of Volumetric Brain Images.	Pierre Hellier,Christian Barillot,Étienne Mémin,Patrick Pérez	2000	An Energy-Based Framework for Dense 3D Registration of Volumetric Brain Images.
CVPR	Catadioptric Sensors that Approximate Wide-Angle Perspective Projections.	R. Andrew Hicks,Ruzena Bajcsy	2000	We present two families of reflective surfaces that are capable of providing a wide field of view, and yet still approximate a perspective projection to a high degree. These surfaces are derived by considering a plane perpendicular to the axis of a surface of revolution and finding the equations governing the distortion of the image of the plane in this surface.We then view this relation as a differential equation and prescribe the distortion term to be linear. By choosing appropriate initial conditions for the differential equation and solving it numerically, we derive the surface shape and obtain a precise estimate as to what degree the resulting sensor can approximate a perspective projection. Thus, these surfaces act as computational sensors, allowing for a wide-angle perspective view of a scene without processing the image in software. The applications of such a sensor should be numerous, including surveillance, robotics and traditional photography.Recently, many researchers in the robotics and vision community have begun to consider visual sensors that are able to obtain wide fields of view. Such devices are the natural solution to various difficulties encountered with conventional imaging systems.The two most common means of obtaining wide fields of view are fish-eye lenses and reflective surfaces, also known as catoptrics. When catoptrics are combined with conventional lens systems, known as dioptrics, the resulting sensors are known as catadioptrics. The possible uses of these systems include applications such as robot control and surveillance. In this paper, we will consider only catadioptri-based sensors. Often such systems consist of a camera pointing at a convex mirror.How to interpret and make use of the visual information obtained by such systems, e.g. how they should be used to control robots is not at all obvious. There are infinitely many different shapes that a mirror can have, and at least two different camera models (perspective and orthographic projection) with which to combine each mirror.The properties of the resulting sensors are very sensitive to these choices. The classic need for wide-angle lenses have, of course been in photography. In particular, underwater and architectural photography are two examples in which having a wide-angle lens is often crucial. The commercially available lens with the widest field of view (without radial distortion) that the authors are aware of is the Nikon 13mm f/5.6 Nikkor AIS, which provides a field of view of 118 degrees at a cost of $(US) 12000. Note that our prototype orthographic sensor provides a field of view of 142 degrees.
CVPR	Representation and Recognition of Complex Human Motion.	Jesse Hoey,James J. Little	2000	Representation and Recognition of Complex Human Motion.
CVPR	Learning Patterns from Images by Combining Soft Decisions and Hard Decisions.	Pengyu Hong,Thomas S. Huang,Roy Wang	2000	Learning Patterns from Images by Combining Soft Decisions and Hard Decisions.
CVPR	Representation and Optimal Recognition of Human Activities.	Somboon Hongeng,François Brémond,Ramakant Nevatia	2000	Representation and Optimal Recognition of Human Activities.
CVPR	Integrating Color, Texture, and Geometry for Image Retrieval.	Nicholas R. Howe,Daniel P. Huttenlocher	2000	Integrating Color, Texture, and Geometry for Image Retrieval.
CVPR	Pose Estimation, Model Refinement, and Enhanced Visualization Using Video.	Stephen C. Hsu,Supun Samarasekera,Rakesh Kumar,Harpreet S. Sawhney	2000	Pose Estimation, Model Refinement, and Enhanced Visualization Using Video.
CVPR	Real-Time 3-D Motion and Structure of Point-Features: A Front-End for Vision-Based Control and Interaction.	Hailin Jin,Paolo Favaro,Stefano Soatto	2000	Real-Time 3-D Motion and Structure of Point-Features: A Front-End for Vision-Based Control and Interaction.
CVPR	Stereoscopic Shading: Integrating Mult1Frame Shape Cues in a Variational Framework.	Hailin Jin,Stefano Soatto,Anthony J. Yezzi	2000	Stereoscopic Shading: Integrating Mult1Frame Shape Cues in a Variational Framework.
CVPR	Surface Landmark Selection and Matching in Natural Terrain.	Andrew E. Johnson	2000	Surface Landmark Selection and Matching in Natural Terrain.
CVPR	Statistics of Range Images.	Jinggang Huang,Ann B. Lee,David Mumford	2000	Statistics of Range Images.
CVPR	Transformed Hidden Markov Models: Estimating Mixture Models of Images and Inferring Spatial Transformations in Video Sequence.	Nebojsa Jojic,Nemanja Petrovic,Thomas S. Huang,Brendan J. Frey	2000	Transformed Hidden Markov Models: Estimating Mixture Models of Images and Inferring Spatial Transformations in Video Sequence.
CVPR	Multisensor Integration for Building Modeling.	Andres Huertas,Zu Whan Kim,Ramakant Nevatia	2000	Multisensor Integration for Building Modeling.
CVPR	Local Appearance for Robust Object Recognition.	Deeptiman Jugessur,Gregory Dudek	2000	Local Appearance for Robust Object Recognition.
CVPR	Are Multifractal Multipermuted Multinomial Measures Good Enough for Unsupervised Image Segmentation?	Lui Kam,Jacques Blanc-Talon	2000	Are Multifractal Multipermuted Multinomial Measures Good Enough for Unsupervised Image Segmentation?
CVPR	Ill-Posed Problems in Surface and Surface Shape Recovery.	George Kamberov,Gerda Kamberova	2000	Ill-Posed Problems in Surface and Surface Shape Recovery.
CVPR	Visual Tunnel Analysis for Visibility Prediction and Camera Planning.	Sing Bing Kang,Peter-Pike J. Sloan,Steven M. Seitz	2000	Visual Tunnel Analysis for Visibility Prediction and Camera Planning.
CVPR	Spatio-Temporal Analysis of Omni Image.	Hiroshi Kawasaki,Katsushi Ikeuchi,Masao Sakauchi	2000	Spatio-Temporal Analysis of Omni Image.
CVPR	Camera Model Selection Based on Geometric AIC.	Keisuke Kinoshita,Michael Lindenbaum	2000	Camera Model Selection Based on Geometric AIC.
CVPR	Statistical cues for Domain Specific Image Segmentation with Performance Analysis.	Scott Konishi,Alan L. Yuille	2000	Statistical cues for Domain Specific Image Segmentation with Performance Analysis.
CVPR	Retinal Thickness Measurements in Optical Coherence Tomography Using a Markov Boundary Model.	Dara Koozekanani,Kim L. Boyer,Cynthia Roberts	2000	Retinal Thickness Measurements in Optical Coherence Tomography Using a Markov Boundary Model.
CVPR	Tracking Segmented Objects Using Tensor Voting.	Pierre Kornprobst,Gérard G. Medioni	2000	Tracking Segmented Objects Using Tensor Voting.
CVPR	Augmented Reality System for Surgical Navigation Using Robust Target Vision.	Akio Kosaka,Akito Saito,Yukihito Furuhashi,Takao Shibasaki	2000	Augmented Reality System for Surgical Navigation Using Robust Target Vision.
CVPR	A Fast and Robust Approach to Recovering Structure and Motion from Live Video Frames.	Takeshi Kurata,Jun Fujiki,Masakatsu Kourogi,Katsuhiko Sakaue	2000	A Fast and Robust Approach to Recovering Structure and Motion from Live Video Frames.
CVPR	Shape Descriptors for Non-Rigid Shapes with a Single Closed Contour.	Longin Jan Latecki,Rolf Lakämper,Ulrich Eckhardt	2000	Shape Descriptors for Non-Rigid Shapes with a Single Closed Contour.
CVPR	Detecting Changes in 3-D Shape Using Self-Consistency.	Yvan G. Leclerc,Quang-Tuan Luong,Pascal Fua,Koji Miyajima	2000	Detecting Changes in 3-D Shape Using Self-Consistency.
CVPR	A Semi-Automatic Method for Resolving Occlusion in Augmented Reality.	Vincent Lepetit,Marie-Odile Berger	2000	A Semi-Automatic Method for Resolving Occlusion in Augmented Reality.
CVPR	Statistical Shape Influence in Geodesic Active Contours.	Michael E. Leventon,W. Eric L. Grimson,Olivier D. Faugeras	2000	Statistical Shape Influence in Geodesic Active Contours.
CVPR	Visual Websearching Using Iconic Queries.	Michael S. Lew,Nicu Sebe	2000	Visual Websearching Using Iconic Queries.
CVPR	Improving Visual Matching.	Michael S. Lew,Nicu Sebe,Thomas S. Huang	2000	Improving Visual Matching.
CVPR	Edge-Constrained Joint View Triangulation for Image Interpolation.	Maxime Lhuillier,Long Quan	2000	Edge-Constrained Joint View Triangulation for Image Interpolation.
CVPR	Codimension - Two Geodesic Active Contours for the Segmentation of Tubular Structures.	Liana M. Lorigo,W. Eric L. Grimson,Olivier D. Faugeras,Renaud Keriven,Ron Kikinis,Arya Nabavi,Carl-Fredrik Westin	2000	Codimension - Two Geodesic Active Contours for the Segmentation of Tubular Structures.
CVPR	Color Channels Decorrelation by ICA Transformation in the Wavelet Domain for Color Texture Analysis and Synthesis.	Yufeng Liang,Eero P. Simoncelli,Zhibin Lei	2000	Color Channels Decorrelation by ICA Transformation in the Wavelet Domain for Color Texture Analysis and Synthesis.
CVPR	An On-Line Handwriting Recognition System Using Fisher Segmental Matching and Hypotheses Propagation Network.	Jong Oh,Davi Geiger	2000	An On-Line Handwriting Recognition System Using Fisher Segmental Matching and Hypotheses Propagation Network.
CVPR	Conditioning Analysis of Missing Data Estimation for Large Sensor Array.	Hairong Qi,Wesley E. Snyder	2000	Conditioning Analysis of Missing Data Estimation for Large Sensor Array.
CVPR	Multidimensional Motion Segmentation and Identification.	Chunmei Lu,Haizhu Liu,Nicola J. Ferrier	2000	Multidimensional Motion Segmentation and Identification.
CVPR	Zebra-Crossing Detection for the Partially Sighted.	Stephen Se	2000	Zebra-Crossing Detection for the Partially Sighted.
CVPR	An Appearance Representation for Multiple Reflection Components.	Stephen Lin,Sang Wook Lee	2000	An Appearance Representation for Multiple Reflection Components.
CVPR	On the Number of Samples Needed in Light Field Rendering with Constant-Depth Assumption.	Zhouchen Lin,Heung-Yeung Shum	2000	On the Number of Samples Needed in Light Field Rendering with Constant-Depth Assumption.
CVPR	Robust Snake Model.	Hui Luo,Qiang Lu,Raj Acharya,R. Gaborski	2000	Robust Snake Model.
CVPR	On Measuring Low-Level Saliency in Photographic Images.	Jiebo Luo,Amit Singhal	2000	On Measuring Low-Level Saliency in Photographic Images.
CVPR	A Novel Algorithm for Rotated Human Face Detection.	Xiao-guang Lv,Jie Zhou,Changshui Zhang	2000	A Novel Algorithm for Rotated Human Face Detection.
CVPR	Region Correspondence by Global Configuration Matching and Progressive Delaunay Triangulation.	Jianbo Ma,Narendra Ahuja	2000	Region Correspondence by Global Configuration Matching and Progressive Delaunay Triangulation.
CVPR	Color Tracking by Transductive Learning.	Ying Wu,Thomas S. Huang	2000	Color Tracking by Transductive Learning.
CVPR	Iterative Projective Reconstruction from Multiple Views.	Shyjan Mahamud,Martial Hebert	2000	Iterative Projective Reconstruction from Multiple Views.
CVPR	A Computational Model for Repeated Pattern Perception Using Frieze and Wallpaper Groups.	Yanxi Liu,Robert T. Collins	2000	A Computational Model for Repeated Pattern Perception Using Frieze and Wallpaper Groups.
CVPR	Assignment Problem in Edge Detection Performance Evaluation.	Gang Liu,Robert M. Haralick	2000	Assignment Problem in Edge Detection Performance Evaluation.
CVPR	Mixture Models and the Segmentation of Multimodal Textures.	Roberto Manduchi	2000	Mixture Models and the Segmentation of Multimodal Textures.
CVPR	Reconstruction of Scene Models from Sparse 3D Structure.	Anastasios Manessis,Adrian Hilton,Phil Palmer,Philip F. McLauchlan,Xinquan Shen	2000	Reconstruction of Scene Models from Sparse 3D Structure.
CVPR	Recognition of Partially Occluded and/or Imprecisely Localized Faces Using a Probabilistic Approach.	Aleix M. Martínez	2000	Recognition of Partially Occluded and/or Imprecisely Localized Faces Using a Probabilistic Approach.
CVPR	A General Method for Errors-in-Variables Problems in Computer Vision.	Bogdan Matei,Peter Meer	2000	A General Method for Errors-in-Variables Problems in Computer Vision.
CVPR	Dynamic Memory: Architecture for Real Time Integration of Visual Perception, Camera Action, and Network Communication.	Takashi Matsuyama,Shinsaku Hiura,Toshikazu Wada,K. Murase,A. Yoshioka	2000	Dynamic Memory: Architecture for Real Time Integration of Visual Perception, Camera Action, and Network Communication.
CVPR	A Batch/Recursive Algorithm for 3D Scene Reconstruction.	Philip F. McLauchlan	2000	A Batch/Recursive Algorithm for 3D Scene Reconstruction.
CVPR	An Image-Based Bayesian Framework for Face Detection.	Lingmin Meng,Truong Q. Nguyen,David A. Castañon	2000	An Image-Based Bayesian Framework for Face Detection.
CVPR	Learning from One Example through Shared Densities on Transforms.	Erik G. Miller,Nicholas E. Matsakis,Paul A. Viola	2000	Learning from One Example through Shared Densities on Transforms.
CVPR	Recovering Projection Geometry: How a Cheap Camera Can Outperform an Expensive Stereo System.	Matthias Mitschke,Nassir Navab	2000	Recovering Projection Geometry: How a Cheap Camera Can Outperform an Expensive Stereo System.
CVPR	Scene Modeling for Wide Area Surveillance and Image Synthesis.	Anurag Mittal,Daniel P. Huttenlocher	2000	Scene Modeling for Wide Area Surveillance and Image Synthesis.
CVPR	High Dynamic Range Imaging: Spatially Varying Pixel Exposures.	Shree K. Nayar,Tomoo Mitsunaga	2000	High Dynamic Range Imaging: Spatially Varying Pixel Exposures.
CVPR	Likelihood Functions and Confidence Bounds for Total-Least-Squares Problems.	Oscar Nestares,David J. Fleet,David J. Heeger	2000	Likelihood Functions and Confidence Bounds for Total-Least-Squares Problems.
CVPR	Solving the Rotation-Estimation Problem by Using the Perspective Three-Point Algorithm.	Toshio Moriya,Haruo Takeda	2000	Solving the Rotation-Estimation Problem by Using the Perspective Three-Point Algorithm.
CVPR	Image-Consistent Surface Triangulation.	Daniel D. Morris,Takeo Kanade	2000	Image-Consistent Surface Triangulation.
CVPR	Multifeature Object Tracking Using a Model-Free Approach.	Hieu Tat Nguyen,Marcel Worring	2000	Multifeature Object Tracking Using a Model-Free Approach.
CVPR	Intelligent Selection Tools.	Eric N. Mortensen,L. Jack Reese,William A. Barrett	2000	Intelligent Selection Tools.
CVPR	Improved Motion Stereo Matching Based on a Modified Dynamic Programming.	Mikhail Mozerov,Vitaly Kober,Tae-Sun Choi	2000	Improved Motion Stereo Matching Based on a Modified Dynamic Programming.
CVPR	Multi-Modality Model-Based Registration in the Cardiac Domain.	"Thomas O'Donnell,Shmuel Aharon,Alok Gupta,Gareth Funka-Lea,Sandra Simon Halliburton,Richard D. White"	2000	Multi-Modality Model-Based Registration in the Cardiac Domain.
CVPR	Face Recognition by Distribution Specific Feature Extraction.	Kenji Nagao	2000	Face Recognition by Distribution Specific Feature Extraction.
CVPR	Chromatic Framework for Vision in Bad Weather.	Srinivasa G. Narasimhan,Shree K. Nayar	2000	Chromatic Framework for Vision in Bad Weather.
CVPR	Visual Servoing for Automatic and Uncalibrated Needle Placement for Percutaneous Procedures.	Nassir Navab,Benedicte Bascle,Michael H. Loser,Bernhard Geiger,Russel Taylor	2000	Visual Servoing for Automatic and Uncalibrated Needle Placement for Percutaneous Procedures.
CVPR	Lines in One Orthographic and Two Perspective Views.	Nassir Navab,Yakup Genc,Mirko Appel	2000	We introduce a linear algorithm to recover the Euclidean motion between an orthographic and two perspective cameras from straight line correspondences filling the gap in the analysis of motion estimation from line correspondences for various projection models. The general relationship between lines in three views is described by the trifocal tensor. Euclidean structure from motion for three perspective views is a special case in which the relationship is defined by a collection of three matrices. Here, we describe the case of two calibrated perspective views and an orthographic view. Similar to the other cases, our linear algorithm requires 13 or more line correspondences to recover 27 coefficients of the trifocal tensor.
CVPR	360 x 360 Mosaics.	Shree K. Nayar,Amruta Karmarkar	2000	360 x 360 Mosaics.
CVPR	Structure from Motion Using Points, Lines, and Intensities.	John Oliensis,Michael Werman	2000	Structure from Motion Using Points, Lines, and Intensities.
CVPR	Maximum-Likelihood Template Matching.	Clark F. Olson	2000	Maximum-Likelihood Template Matching.
CVPR	Robust Stereo Ego-motion for Long Distance Navigation.	Clark F. Olson,Larry Matthies,Marcel Schoppers,Mark W. Maimone	2000	Robust Stereo Ego-motion for Long Distance Navigation.
CVPR	"Active Character Recognition Using 'A*-Like' Algorit."	Jaehwa Park,Venu Govindaraju	2000	"Active Character Recognition Using 'A*-Like' Algorit."
CVPR	Using Lexical Similarity in Handwritten Word Recognition.	Jaehwa Park,Venu Govindaraju	2000	Using Lexical Similarity in Handwritten Word Recognition.
CVPR	Impact of Dynamic Model Learning on Classification of Human Motion.	Vladimir Pavlovic,James M. Rehg	2000	Impact of Dynamic Model Learning on Classification of Human Motion.
CVPR	Gesture, Speech, and Gaze Cues for Discourse Segmentation.	Francis K. H. Quek,David McNeill,Robert K. Bryll,Cemil Kirbas,Hasan Arslan,Karl E. McCullough,Nobuhiro Furuyama	2000	Gesture, Speech, and Gaze Cues for Discourse Segmentation.
CVPR	Multimodal Speaker Detection Using Error Feedback Dynamic Bayesian Networks.	Vladimir Pavlovic,James M. Rehg,Ashutosh Garg,Thomas S. Huang	2000	Multimodal Speaker Detection Using Error Feedback Dynamic Bayesian Networks.
CVPR	Parameterization of Closed Surfaces for Parametric Surface Descriptio.	Michael Quicke,Christian Brechbühler,Johannes Hug,Hans Blattman,Gábor Székely	2000	Parameterization of Closed Surfaces for Parametric Surface Descriptio.
CVPR	Cameras for Stereo Panoramic Imaging.	Shmuel Peleg,Yael Pritch,Moshe Ben-Ezra	2000	Cameras for Stereo Panoramic Imaging.
CVPR	Rectified Mosaicing: Mosaics without the Curl.	Shmuel Peleg,Assaf Zomet,Chetan Arora	2000	Rectified Mosaicing: Mosaics without the Curl.
CVPR	Maintaining Valid Topology with Active Contours: Theory and Application.	A. G. Amitha Perera,Chia-Ling Tsai,Robin Y. Flatland,Charles V. Stewart	2000	Maintaining Valid Topology with Active Contours: Theory and Application.
CVPR	Calibration of Light Sources.	Mark W. Powell,Sudeep Sarkar,Dmitry B. Goldgof	2000	Calibration of Light Sources.
CVPR	Perspective Pose from Spectral Voting.	Eraldo Ribeiro,Edwin R. Hancock	2000	Perspective Pose from Spectral Voting.
CVPR	Invariant Web Defect Detection and Classification System.	Dominik R. Rohrmus	2000	Invariant Web Defect Detection and Classification System.
CVPR	Inferring Body Pose without Tracking Body Parts.	Rómer Rosales,Stan Sclaroff	2000	Abstract A novel approach for estimating articulated body posture and motion from monocular video sequences is proposed. Human pose is defined as the instantaneous two dimensional configuration (i.e., the projection onto the image plane) of a single articulated body in terms of the position of a predetermined set of joints. First, statistical segmentation of the human bodies from the background is performed and low-level visual features are found given the segmented body shape. The goal is to be able to map these, generally low level, visual features to body configurations. The system estimates different mappings, each one with a specific cluster in the visual feature space. Given a set of body motion sequences for training, unsupervised clustering is obtained via the Expectation Maximation algorithm. Then, for each of the clusters, a function is estimated to build the mapping between low-level features to 3D pose. Currently this mapping is modeled by a neural network. Given new visual features, a mapping from each cluster is performed to yield a set of possible poses. From this set, the system selects the most likely pose given the learned probability distribution and the visual feature similarity between hypothesis and input. Performance of the proposed approach is characterized using a new set of known body postures, showing promising results.
CVPR	Learning to Recognize Objects.	Dan Roth,Ming-Hsuan Yang,Narendra Ahuja	2000	Learning to Recognize Objects.
CVPR	Segmenting Visual Actions Based on Spatio-Temporal Motion Patterns.	Yong Rui,P. Anandan	2000	Segmenting Visual Actions Based on Spatio-Temporal Motion Patterns.
CVPR	Optimizing Learning in Image Retrieval.	Yong Rui,Thomas S. Huang	2000	Optimizing Learning in Image Retrieval.
CVPR	Alpha Estimation in Natural Images.	Mark A. Ruzon,Carlo Tomasi	2000	Alpha Estimation in Natural Images.
CVPR	Variable Albedo Surface Reconstruction from Stereo and Shape from Shading.	Dimitris Samaras,Dimitris N. Metaxas,Pascal Fua,Yvan G. Leclerc	2000	Variable Albedo Surface Reconstruction from Stereo and Shape from Shading.
CVPR	Agent-Based Moving Object Correspondence Using Differential Discriminative Diagnosis.	Mahesh Saptharishi,John B. Hampshire II,Pradeep K. Khosla	2000	Agent-Based Moving Object Correspondence Using Differential Discriminative Diagnosis.
CVPR	Blind Recovery of Transparent and Semireflected Scenes.	Yoav Y. Schechner,Joseph Shamir,Nahum Kiryati	2000	Blind Recovery of Transparent and Semireflected Scenes.
CVPR	A Statistical Method for 3D Object Detection Applied to Faces and Cars.	Henry Schneiderman,Takeo Kanade	2000	A Statistical Method for 3D Object Detection Applied to Faces and Cars.
CVPR	Corner Guided Curve Matching and its Application to Scene Reconstruction.	Ying Shan,Zhengyou Zhang	2000	Corner Guided Curve Matching and its Application to Scene Reconstruction.
CVPR	Fast Multiscale Image Segmentation.	Eitan Sharon,Achi Brandt,Ronen Basri	2000	Fast Multiscale Image Segmentation.
CVPR	Estimation and Prediction of Evolving Color Distributions for Skin Segmentation under Varying Illumination.	Leonid Sigal,Stan Sclaroff,Vassilis Athitsos	2000	Estimation and Prediction of Evolving Color Distributions for Skin Segmentation under Varying Illumination.
CVPR	Smooth Region Structure: Folds, Domes, Bowls, Ridges, Valleys and Slopes.	David Sinclair	2000	Smooth Region Structure: Folds, Domes, Bowls, Ridges, Valleys and Slopes.
CVPR	Template Deformation Constrained by Shape Priors.	Mark B. Skouson,Zhi-Pei Liang	2000	Template Deformation Constrained by Shape Priors.
CVPR	Surface Growing from Stereo Images.	Oskar M. Skrinjar,Hemant Tagare,James S. Duncan	2000	Surface Growing from Stereo Images.
CVPR	Bayesian Super-Resolved Surface Reconstruction from Images.	Vadim N. Smelyansky,Peter Cheeseman,David A. Maluf,Robin D. Morris	2000	Bayesian Super-Resolved Surface Reconstruction from Images.
CVPR	A Geometric Approach to Blind Deconvolution with Application to Shape from Defocu.	Stefano Soatto,Paolo Favaro	2000	A Geometric Approach to Blind Deconvolution with Application to Shape from Defocu.
CVPR	Towards Detection of Human Motion.	Yang Song,Xiaolin Feng,Pietro Perona	2000	Towards Detection of Human Motion.
CVPR	Line Net Global Vectorization: an Algorithm and Its Performance Evaluation.	Jiqiang Song,Feng Su,Jibing Chen,Chiew-Lan Tai,Shijie Cai	2000	Line Net Global Vectorization: an Algorithm and Its Performance Evaluation.
CVPR	3-D Model Construction Using Range and Image Data.	Ioannis Stamos,Peter K. Allen	2000	3-D Model Construction Using Range and Image Data.
CVPR	Image-Based Re-Rendering of Faces for Continuous Pose and Illumination Directions.	Arne Stoschek	2000	Image-Based Re-Rendering of Faces for Continuous Pose and Illumination Directions.
CVPR	Algorithms for Plane-Based Pose Estimation.	Peter F. Sturm	2000	Algorithms for Plane-Based Pose Estimation.
CVPR	Indexing for Topics in Videos Using Foils.	Tanveer Fathima Syeda-Mahmood	2000	Indexing for Topics in Videos Using Foils.
CVPR	CueVideo: A System for Cross-Modal Search and Browse of Video Databases.	Tanveer Fathima Syeda-Mahmood,Savitha Srinivasan,Arnon Amir,Dulce B. Ponceleon,Brian Blanchard,Dragutin Petkovic	2000	CueVideo: A System for Cross-Modal Search and Browse of Video Databases.
CVPR	Layer Extraction from Multiple Images Containing Reflections and Transparency.	Richard Szeliski,Shai Avidan,P. Anandan	2000	Layer Extraction from Multiple Images Containing Reflections and Transparency.
CVPR	Arbitrary View Position and Direction Rendering for Large-Scale Scenes.	Takuji Takahashi,Hiroshi Kawasaki,Katsushi Ikeuchi,Masao Sakauchi	2000	Arbitrary View Position and Direction Rendering for Large-Scale Scenes.
CVPR	A Parametric Template Method and Its Application to Robust Matching.	Koichi Tanaka,Mutsuo Sano,Shuichi Ohara,Masashi Okudaira	2000	A Parametric Template Method and Its Application to Robust Matching.
CVPR	Flying a Toy Plane.	Donald Tanguay	2000	Flying a Toy Plane.
CVPR	Dynamic Layer Representation with Applications to Tracking.	Hai Tao,Harpreet S. Sawhney,Rakesh Kumar	2000	Dynamic Layer Representation with Applications to Tracking.
CVPR	Reconstruction of Articulated Objects from Point Correspondences in a Single Uncalibrated Image.	Camillo J. Taylor	2000	Reconstruction of Articulated Objects from Point Correspondences in a Single Uncalibrated Image.
CVPR	Handwritten Digit Recognition with a Novel Vision Model that Extracts Linearly Separable Features.	Loo-Nin Teow,Kia-Fock Loe	2000	Handwritten Digit Recognition with a Novel Vision Model that Extracts Linearly Separable Features.
CVPR	Fusion of Color, Shading and Boundary Information for Factory Pipe Segmentation.	Bertrand Thirion,Benedicte Bascle,Visvanathan Ramesh,Nassir Navab	2000	Fusion of Color, Shading and Boundary Information for Factory Pipe Segmentation.
CVPR	Recognizing Upper Face Action Units for Facial Expression Analysis.	Ying-li Tian,Takeo Kanade,Jeffrey F. Cohn	2000	Recognizing Upper Face Action Units for Facial Expression Analysis.
CVPR	Boosting Image Retrieval.	Kinh Tieu,Paul A. Viola	2000	We present an approach for image retrieval using a very large number of highly selective features and efficient learning of queries. Our approach is predicated on the assumption that each image is generated by a sparse set of visual &ldquo;causes&rdquo; and that images which are visually similar share causes. We propose a mechanism for computing a very large number of highly selective features which capture some aspects of this causal structure (in our implementation there are over 46,000 highly selective features). At query time a user selects a few example images, and the AdaBoost algorithm is used to learn a classification function which depends on a small number of the most appropriate features. This yields a highly efficient classification function. In addition we show that the AdaBoost framework provides a natural mechanism for the incorporation of relevance feedback. Finally we show results on a wide variety of image queries.
CVPR	Tracking of Elongated Structures Using Statistical Snakes.	Ricardo Toledo,Xavier Orriols,Xavier Binefa,Petia Radeva,Jordi Vitrià,Juan José Villanueva	2000	Tracking of Elongated Structures Using Statistical Snakes.
CVPR	A Curve Evolution Approach to Smoothing and Segmentation Using the Mumford-Shah Functional.	Andy Tsai,Anthony J. Yezzi,Alan S. Willsky	2000	A Curve Evolution Approach to Smoothing and Segmentation Using the Mumford-Shah Functional.
CVPR	Multiscale Combination of Physically-Based Registration and Deformation Modeling.	Leonid V. Tsap,Dmitry B. Goldgof,Sudeep Sarkar	2000	Multiscale Combination of Physically-Based Registration and Deformation Modeling.
CVPR	A Probabilistic Architecture for Content-Based Image Retrieval.	Nuno Vasconcelos,Andrew Lippman	2000	A Probabilistic Architecture for Content-Based Image Retrieval.
CVPR	Shape and Motion Carving in 6D.	Sundar Vedula,Simon Baker,Steven M. Seitz,Takeo Kanade	2000	Shape and Motion Carving in 6D.
CVPR	Image Segmentation by Nested Cuts.	Olga Veksler	2000	Image Segmentation by Nested Cuts.
CVPR	Improving Correlation-Based DEMs by Image Warping and Facade Correlation.	Christophe Vestri,Frederic Devernay	2000	Improving Correlation-Based DEMs by Image Warping and Facade Correlation.
CVPR	An Effective Approach to Detect Lesions in Color Retinal Images.	Huan Wang,Wynne Hsu,Kheng Guan Goh,Mong-Li Lee	2000	An Effective Approach to Detect Lesions in Color Retinal Images.
CVPR	Shape-Based 3D Surface Correspondence Using Geodesics and Local Geometry.	Yongmei Wang,Bradley S. Peterson,Lawrence H. Staib	2000	Shape-Based 3D Surface Correspondence Using Geodesics and Local Geometry.
CVPR	Towards Automatic Discovery of Object Categories.	Markus Weber,Max Welling,Pietro Perona	2000	Towards Automatic Discovery of Object Categories.
CVPR	On the Synthesis of Dynamic Scenes from Reference Views.	Yonatan Wexler,Amnon Shashua	2000	On the Synthesis of Dynamic Scenes from Reference Views.
CVPR	Histogram-Based Object Recognition Using Shape-from-Shading.	Philip L. Worthington,Edwin R. Hancock	2000	Histogram-Based Object Recognition Using Shape-from-Shading.
CVPR	View-Independent Recognition of Hand Postures.	Ying Wu,Thomas S. Huang	2000	View-Independent Recognition of Hand Postures.
CVPR	Discriminant-EM Algorithm with Application to Image Retrieval.	Ying Wu,Qi Tian,Thomas S. Huang	2000	Discriminant-EM Algorithm with Application to Image Retrieval.
CVPR	Feature Based Visualization of Geophysical Data.	Qing Yang,Bahram Parvin	2000	Feature Based Visualization of Geophysical Data.
CVPR	Two-Stage Robust Optical Flow Estimation.	Ming Ye,Robert M. Haralick	2000	Two-Stage Robust Optical Flow Estimation.
CVPR	A Linear Algorithm for Camera Self-Calibration, Motion and Structure Recovery for Multi-Planar Scenes from Two Perspective Images.	Gang Xu,Jun-ichi Terai,Heung-Yeung Shum	2000	A Linear Algorithm for Camera Self-Calibration, Motion and Structure Recovery for Multi-Planar Scenes from Two Perspective Images.
CVPR	Detecting and Tracking Human Face and Eye Using Space-Varying Sensor and an Active Vision Head.	Mohammed Yeasin,Yasuo Kuniyoshi	2000	Detecting and Tracking Human Face and Eye Using Space-Varying Sensor and an Active Vision Head.
CVPR	Geodesic Distance Evolution of Surfaces: A New Method for Matching Surfaces.	Hussein M. Yahia,Etienne G. Huot,Isabelle Herlin,Isaac Cohen	2000	Geodesic Distance Evolution of Surfaces: A New Method for Matching Surfaces.
CVPR	Scene Constraints-Aided Tracking of Human Body.	Masanobu Yamamoto,Katsutoshi Yagishita	2000	Scene Constraints-Aided Tracking of Human Body.
CVPR	Physical Panoramic Pyramid and Noise Sensitivity in Pyramids.	Weihong Yin,Terrance E. Boult	2000	Physical Panoramic Pyramid and Noise Sensitivity in Pyramids.
CVPR	A Geometric Approach to Train Support Vector Machines.	Ming-Hsuan Yang,Narendra Ahuja	2000	A Geometric Approach to Train Support Vector Machines.
CVPR	Hierarchical Structure and Nonrigid Motion Recovery from 2D Monocular Views.	Lin Zhou,Chandra Kambhamettu	2000	Hierarchical Structure and Nonrigid Motion Recovery from 2D Monocular Views.
CVPR	Fluid Structure and Motion Analysis from Multi-spectrum 2D Cloud Image Sequences.	Lin Zhou,Chandra Kambhamettu,Dmitry B. Goldgof	2000	Fluid Structure and Motion Analysis from Multi-spectrum 2D Cloud Image Sequences.
CVPR	Order Parameters for Minimax Entropy Distributions: When Does High Level Knowledge Help?	Alan L. Yuille,James M. Coughlan,Song Chun Zhu,Ying Nian Wu	2000	Order Parameters for Minimax Entropy Distributions: When Does High Level Knowledge Help?
CVPR	Learning in Gibbsian Fields: How Accurate and How Fast Can It Be?	Song Chun Zhu,Xiuwen Liu	2000	"Gibbsian fields or Markov random fields are widely used in Bayesian image analysis, but learning Gibbs models is computationally expensive. The computational complexity is pronounced by the recent minimax entropy (FRAME) models which use large neighborhoods and hundreds of parameters. In this paper, we present a common framework for learning Gibbs models. We identify two key factors that determine the accuracy and speed of learning Gibbs models: The efficiency of likelihood functions and the variance in approximating partition functions using Monte Carlo integration. We propose three new algorithms. In particular, we are interested in a maximum satellite likelihood estimator, which makes use of a set of precomputed Gibbs models called ""satellites"" to approximate likelihood functions. This algorithm can approximately estimate the minimax entropy model for textures in seconds in a HP workstation. The performances of various learning algorithms are compared in our experiments."
CVPR	Fast Face Detection Using Subspace Discriminant Wavelet Features.	Ying Zhu,Stuart C. Schwartz,Michael T. Orchard	2000	Fast Face Detection Using Subspace Discriminant Wavelet Features.
CVPR	Integrating Bottom-Up/Top-Down for Object Recognition by Data Driven Markov Chain Monte Carlo.	Song Chun Zhu,Rong Zhang,Zhuowen Tu	2000	Integrating Bottom-Up/Top-Down for Object Recognition by Data Driven Markov Chain Monte Carlo.
CVPR	Integrated 3D Scene Flow and Structure Recovery from Multiview Image Sequences.	Ye Zhang,Chandra Kambhamettu	2000	Integrated 3D Scene Flow and Structure Recovery from Multiview Image Sequences.
CVPR	Illuminant Direction Determination for Multiple Light Sources.	Yufei Zhang,Yee-Hong Yang	2000	Illuminant Direction Determination for Multiple Light Sources.
CVPR	Illumination-Insensitive Face Recognition Using Symmetric Shape-from-Shading.	Wen-Yi Zhao,Rama Chellappa	2000	Illumination-Insensitive Face Recognition Using Symmetric Shape-from-Shading.
CVPR	2000 Conference on Computer Vision and Pattern Recognition (CVPR 2000), 13-15 June 2000, Hilton Head, SC, USA		2000	2000 Conference on Computer Vision and Pattern Recognition (CVPR 2000), 13-15 June 2000, Hilton Head, SC, USA
IEEE Visualization	Real-time visualization of the clear-up of a former U.S. Naval base.	Paul Chapman,Derek Wills,Peter Stevens,Graham Brookes	2000	"This paper describes the effective real-time visualization of theclear-up operation of a former U.S. Nuclear Submarine Base,located in Holy Loch, Scotland. The Whole Field ModellingSystem has provided an extremely accurate real-time visualizationof a large number of varying parameters such as remotelyoperated vehicles, cranes, barges, grabs, magnets, and detailedseabed topography. The system has improved the field staffs'spatial and temporal awareness of the underwater environmentand facilitated decision-making within the complex offshoreworking environment."
IEEE Visualization	Automatic alignment of high-resolution multi-projector display using an un-calibrated camera.	Yuqun Chen,Douglas W. Clark,Adam Finkelstein,Timothy C. Housel,Kai Li	2000	A scalable, high-resolution display may be constructed by tilingmany projected images over a single display surface. One fundamentalchallenge for such a display is to avoid visible seams dueto misalignment among the projectors. Traditional methods foravoiding seams involve sophisticated mechanical devices and expensiveCRT projectors, coupled with extensive human effort forfine-tuning the projectors. This paper describes an automatic alignmentmethod that relies on an inexpensive, uncalibrated camera tomeasure the relative mismatches between neighboring projectors,and then correct the projected imagery to avoid seams without significanthuman effort.
IEEE Visualization	New techniques for topologically correct surface reconstruction.	Udo Adamy,Joachim Giesen,Matthias John	2000	We present a new approach to surface reconstruction based on theDelaunay complex. First we give a simple and fast algorithm thatpicks locally a surface at each vertex. For that, we introduce theconcept of ¿-intervals. It turns out that for smooth regions of thesurface this method works very well and at difficult parts of the surfaceyields an output well-suited for postprocessing. As a postprocessingstep we propose a topological clean up and a new techniquebased on linear programming in order to establish a topologicallycorrect surface. These techniques should be useful also for manyother reconstruction schemes.
IEEE Visualization	CEASAR: a smooth, accurate and robust centerline extraction algorithm.	Ingmar Bitter,Mie Sato,Michael A. Bender,Kevin T. McDonnell,Arie E. Kaufman,Ming Wan	2000	CEASAR: a smooth, accurate and robust centerline extraction algorithm.
IEEE Visualization	Next-generation visualization displays: the research challenges of building tiled displays (panel session).	James P. Ahrens,Kai Li,Daniel A. Reed	2000	Next-generation visualization displays: the research challenges of building tiled displays (panel session).
IEEE Visualization	Polyhedral modeling.	Georges-Pierre Bonneau,Stefanie Hahmann	2000	Polyhedral meshes are used for visualization, computer graphicsor geometric modeling purposes and result from many applicationslike iso-surface extraction, surface reconstruction or CAD/CAM.The present paper introduces a method for constructing smooth surfacesfrom a triangulated polyhedral mesh of arbitrary topology.It presents a new algorithm which generalizes and improves thetriangle 4-split method in the crucial point of boundary curvenetwork construction. This network is then filled-in by a visualsmooth surface from which an explicit closed form parametrizationis given. Furthermore, the method becomes now completely localand can interpolate normal vector input at the mesh vertices.
IEEE Visualization	Constructing material interfaces from data sets with volume-fraction information.	Kathleen S. Bonnell,Kenneth I. Joy,Bernd Hamann,Daniel Schikore,Mark A. Duchaineau	2000	Constructing material interfaces from data sets with volume-fraction information.
IEEE Visualization	Case study: a methodology for plume visualization with application to real-time acquisition and navigation.	Karen G. Bemis,Deborah Silver,Peter A. Rona,Chengwei Feng	2000	Case study: a methodology for plume visualization with application to real-time acquisition and navigation.
IEEE Visualization	Procedural annotation of uncertain information.	Andrej Cedilnik,Penny Rheingans	2000	In many applications of scientific visualization, a large quantity of data is being processed and displayed in order to enable a viewer to make informed and effective decisions. Since little data is perfect, there is almost always some degree of associated uncertainty. This uncertainty is an important part of the data and should be taken into consideration when interpreting the data. Uncertainty, however, should not overshadow the data values. Many methods that address the problem of visualizing data with uncertainty can distort the data and emphasize areas with uncertain values. We have developed a method for showing the uncertainty information together with data with minimal distraction. This method uses procedurally generated annotations which are deformed according to the uncertainty information. As another possible technique we propose distorting glyphs according to the uncertainty information.
IEEE Visualization	Bicubic subdivision-surface wavelets for large-scale isosurface representation and visualization.	Martin Bertram,Mark A. Duchaineau,Bernd Hamann,Kenneth I. Joy	2000	We introduce a new subdivision-surface wavelet transform for arbitrarytwo-manifolds with boundary that is the first to use simplelifting-style filtering operations with bicubic precision. We alsodescribe a conversion process for re-mapping large-scale isosurfacesto have subdivision connectivity and fair parameterizationsso that the new wavelet transform can be used for compressionand visualization. The main idea enabling our wavelet transformis the circular symmetrization of the filters in irregular neighborhoods,which replaces the traditional separation of filters into two1-D passes. Our wavelet transform uses polygonal base meshes torepresent surface topology, from which a Catmull-Clark-style subdivisionhierarchy is generated. The details between these levels ofresolution are quickly computed and compactly stored as waveletcoefficients. The isosurface conversion process begins with a contourtriangulation computed using conventional techniques, whichwe subsequently simplify with a variant edge-collapse procedure,followed by an edge-removal process. This provides a coarse initialbase mesh, which is subsequently refined, relaxed and attractedin phases to converge to the contour. The conversion is designed toproduce smooth, untangled and minimally- skewed parameterizations,which improves the subsequent compression after applyingthe transform. We have demonstrated our conversion and transformfor an isosurface obtained from a high-resolution turbulent-mixinghydrodynamics simulation, showing the potential for compressionand level-of-detail visualization.
IEEE Visualization	Isosurfacing in higher dimensions.	Praveen Bhaniramka,Rephael Wenger,Roger Crawfis	2000	Visualization algorithms have seen substantial improvements inthe past several years. However, very few algorithms have beendeveloped for directly studying data in dimensions higher thanthree. Most algorithms require a sampling in three-dimensionsbefore applying any visualization algorithms. This samplingtypically ignores vital features that may be present whenexamined in oblique cross-sections, and places an undo burden onsystem resources when animation through additional dimensionsis desired. For time-varying data of large data sets, smoothanimation is desired at interactive rates. This paper provides afast Marching Cubes like algorithm for hypercubes of anydimension. To support this, we have developed a new algorithm toautomatically generate the isosurface and triangulation tables forany dimension. This allows the efficient calculation of 4Disosurfaces, which can be interactively sliced to provide smoothanimation or slicing through oblique hyperplanes. The formerallows for smooth animation in a very compressed format. Thelatter provide better tools to study time-evolving features as theymove downstream. We also provide examples in using thistechnique to show interval volumes or the sensitivity of aparticular isovalue threshold.
IEEE Visualization	Toward a compelling sensation of telepresence: demonstrating a portal to a distant (static) office.	Wei-Chao Chen,Herman Towles,Lars S. Nyland,Greg Welch,Henry Fuchs	2000	In 1998 we introduced the idea for a project we call the Office ofthe Future. Our long-term vision is to provide a better everydayworking environment, with high-fidelity scene reconstruction forlife-sized 3D tele-collaboration. In particular, we want a true senseof presence with our remote collaborator and their real surroundings.The challenges related to this vision are enormous and involvemany technical tradeoffs. This is true in particular for scenereconstruction. Researchers have been striving to achieve real-timeapproaches, and while they have made respectable progress, thelimitations of conventional technologies relegate them to relativelylow resolution in a restricted volume.In this paper we present a significant step toward our ultimategoal, via a slightly different path. In lieu of low-fidelity dynamicscene modeling we present an exceedingly high fidelity reconstructionof a real but static office. By assembling the best of availablehardware and software technologies in static scene acquisition,modeling algorithms, rendering, tracking and stereo projective display,we are able to demonstrate a portal to a real office, occupiedtoday by a mannequin, and in the future by a real remote collaborator.We now have both a compelling sense of just how good it couldbe, and a framework into which we will later incorporate dynamicscene modeling, as we continue to head toward our ultimate goal of3D collaborative, telepresence.
IEEE Visualization	Simplification of Tetrahedral meshes with accurate error evaluation.	Paolo Cignoni,D. Constanza,Claudio Montani,Claudio Rocchini,Roberto Scopigno	2000	The techniques for reducing the size of a volume dataset by preserving both the geometrical / topological shape and the information encoded in an attached scalar field areattracting growing interest. Given the framework of incremental 3Dmesh simplification based on edge collapse, the paper proposes an approach for the integrated evaluation of the error introduced by both the modification of the domain and the approximation of the field of the original volume dataset. We present and compare various techniques to evaluate the approximation error or to produce a sound prediction. A flexible simplification tool has beenimplemented, which provides different degree of accuracy and computational efficiency for the selection of the edge to be collapsed. Techniques for preventing a geometric or topological degeneration of the mesh are also presented.
IEEE Visualization	Anisotropic geometric diffusion in surface processing.	Ulrich Clarenz,Udo Diewald,Martin Rumpf	2000	A new multiscale method in surface processing is presented herewhich combines the image processing methodology based on nonlineardiffusion equations and the theory of geometric evolutionproblems. Its aim is to smooth discretized surfaces while simultaneouslyenhancing geometric features such as edges and corners.This is obtained by an anisotropic curvature evolution, where timeis the multiscale parameter. Here, the diffusion tensor depends onthe shape operator of the evolving surface.A spatial finite element discretization on arbitrary unstructuredtriangular meshes and a semi-implicit finite difference discretizationin time are the building blocks of the easy to code algorithmpresented here. The systems of linear equations in each timestepare solved by appropriate, preconditioned iterative solvers. Differentapplications underline the efficiency and flexibility of the presentedtype of surface processing tool.
IEEE Visualization	Multi-user view-dependent rendering.	Jihad El-Sana	2000	"In this paper we are presenting a novel architecture which allowsrendering of large-shared dataset at interactive rates on aninexpensive workstation. The idea is based on view-dependentrendering on a client-server network. The server stores thelarge dataset and manages the selection of the various levelsof detail while the inexpensive clients receive a stream of updateoperations that generate the appropriate level of detail inan incremental fashion. These update operations are based onchanges in the clients' view-parameters. Our approach dramaticallyreduces the amount of memory needed by each clientand the entire computing system since the dataset is stored onlyonce on the server's local memory. In addition, it decreases theload on the network as results of the incremental update contributedby view-dependent rendering."
IEEE Visualization	Combining local and remote visualization techniques for interactive volume rendering in medical applications.	Klaus Engel,Thomas Ertl,Peter Hastreiter,Bernd Tomandl,K. Eberhardt	2000	Combining local and remote visualization techniques for interactive volume rendering in medical applications.
IEEE Visualization	Geometric compression for interactive transmission.	Olivier Devillers,Pierre-Marie Gandoin	2000	The compression of geometric structures is a relatively new fieldof data compression. Since about 1995, several articles have dealtwith the coding of meshes, using for most of them the followingapproach: the vertices of the mesh are coded in an order that partiallycontains the topology of the mesh. In the same time, somesimple rules attempt to predict the position of each vertex from thepositions of its neighbors that have been previously coded.In this article, we describe a compression algorithm whose principleis completely different: the coding order of the vertices is usedto compress their coordinates, and then the topology of the mesh isreconstructed from the vertices. This algorithm achieves compressionratios that are slightly better than those of the currently availablealgorithms, and moreover, it allows progressive and interactivetransmission of the meshes.
IEEE Visualization	An integrated visualization and design toolkit for flexible prosthetic heart valves.	A. J. Fenlon,T. David,J. P. R. B. Walton	2000	We describe a toolkit for the design and visualization of flexible artificial heart valves. The toolkit consists of interlinked modules with a visual programming interface. The user ofthe toolkit can set the initial geometry and material properties of the valve leaflet, solve for the flexing of the leaflet and the flow of blood around it, and display the results using the visualization capabilities of the toolkit. The interactive nature of our environment is highlighted by the fact that changes in leaflet properties are immediately reflected in the flow field and response of the leaflet. Hence the user may, in a single session, investigate a broad range of designs, each one of which provides important information about the blood flow and motion of the valve during the cardiac cycle.
IEEE Visualization	Texturing techniques for terrain visualization.	Jürgen Döllner,Konstantin Baumann,Klaus Hinrichs	2000	We present a new rendering technique for processing multiplemultiresolution textures of LOD terrain models and describe itsapplication to interactive, animated terrain content design. Theapproach is based on a multiresolution model for terrain texturewhich cooperates with a multiresolution model for terrain geometry.For each texture layer, an image pyramid and a texture treeare constructed. Multiple texture layers can be associated with oneterrain model and can be combined in different ways, e.g., byblending and masking. The rendering algorithm traverses simultaneouslythe geometry multiresolution model and the texture multiresolutionmodel, and takes into account geometric and textureapproximation errors. It uses multi-pass rendering and exploitsmultitexturing to achieve real-time performance. Applicationsinclude interactive texture lenses, texture animation, and topographictextures. These techniques offer an enormous potential fordeveloping new visualization applications for presenting, exploringand manipulating spatio-temporal data.
IEEE Visualization	Immersive virtual reality for visualizing flow through an artery.	Andrew S. Forsberg,David H. Laidlaw,Andries van Dam,Robert M. Kirby,George E. Karniadakis,Jonathan L. Elion	2000	Immersive virtual reality for visualizing flow through an artery.
IEEE Visualization	Volume illustration: non-photorealistic rendering of volume models.	David S. Ebert,Penny Rheingans	2000	Accurately and automatically conveying the structure of a volumemodel is a problem not fully solved by existing volume renderingapproaches. Physics-based volume rendering approaches createimages which may match the appearance of translucent materialsin nature, but may not embody important structural details.Transfer function approaches allow flexible design of the volumeappearance, but generally require substantial hand tuning for eachnew data set in order to be effective. We introduce the volumeillustration approach, combining the familiarity of a physics-basedillumination model with the ability to enhance importantfeatures using non-photorealistic rendering techniques. Sincefeatures to be enhanced are defined on the basis of local volumecharacteristics rather than volume sample value, the applicationof volume illustration techniques requires less manual tuning thanthe design of a good transfer function. Volume illustrationprovides a flexible unified framework for enhancing structuralperception of volume models through the amplification offeatures and the addition of illumination effects.
IEEE Visualization	A continuous clustering method for vector fields.	Harald Garcke,Tobias Preußer,Martin Rumpf,Alexandru Telea,Ulrich Weikard,Jarke J. van Wijk	2000	A new method for the simplification of flow fields is presented. It isbased on continuous clustering. A well-known physical clusteringmodel, the Cahn Hillard model which describes phase separation, ismodified to reflect the properties of the data to be visualized. Clustersare defined implicitly as connected components of the positivityset of a density function. An evolution equation for this function isobtained as a suitable gradient flow of an underlying anisotropicenergy functional. Here, time serves as the scale parameter. Theevolution is characterized by a successive coarsening of patterns -the actual clustering - and meanwhile the underlying simulationdata specifies preferable pattern boundaries. Here we discuss theapplicability of this new type of approach mainly for flow fields,where the cluster energy penalizes cross streamline boundaries, butthe method also carries provisions in other fields as well. The clustersare visualized via iconic representations. A skeletonization algorithmis used to find suitable positions for the icons.
IEEE Visualization	Topology preserving and controlled topology simplifying multiresolution isosurface extraction.	Thomas Gerstner,Renato Pajarola	2000	Multiresolution methods are becoming increasingly important toolsfor the interactive visualization of very large data sets. Multiresolutionisosurface visualization allows the user to explore volumedata using simplified and coarse representations of the isosurfacefor overview images, and finer resolution in areas of high interestor when zooming into the data. Ideally, a coarse isosurface shouldhave the same topological structure as the original. The topologicalgenus of the isosurface is one important property which is often neglectedin multiresolution algorithms. This results in uncontrolledtopological changes which can occur whenever the level-of-detailis changed. The scope of this paper is to propose an efficient techniquewhich allows preservation of topology as well as controlledtopology simplification in multiresolution isosurface extraction.
IEEE Visualization	Six degree-of-freedom haptic display of polygonal models.	Arthur D. Gregory,Ajith Mascarenhas,Stephen A. Ehmann,Ming C. Lin,Dinesh Manocha	2000	Six degree-of-freedom haptic display of polygonal models.
IEEE Visualization	WEAVE: a system for visually linking 3-D and statistical visualizations, applied to cardiac simulation and measurement data.	Donna L. Gresh,Bernice E. Rogowitz,Raimond L. Winslow,David F. Scollan,Christina K. Yung	2000	WEAVE (Workbench Environment for Analysis and Visual Exploration)is an environment for creating interactive visualization applications.WEAVE differs from previous systems in that it providestransparent linking between custom 3-D visualizations andmultidimensional statistical representations, and provides interactivecolor brushing betwen all visualizations.In this paper, we demonstrate how WEAVE can be used torapidly prototype a biomedical application, weaving together simulationdata, measurement data, and 3-d anatomical data concerningthe propagation of excitation in the heart. These linked statisticaland custom three-dimensional visualizations of the heart can allowscientists to more effectively study the correspondance of structureand behavior.
IEEE Visualization	Two-level volume rendering &mdash fusing MIP and DVR.	Helwig Hauser,Lukas Mroz,Gian Italo Bischi,Eduard Gröller	2000	Two-level volume rendering &mdash fusing MIP and DVR.
IEEE Visualization	Visualizing geodesics.	Ingrid Hotz,Hans Hagen	2000	"One of the main research topics in scientific visualization is to ""visualizethe appropriate features"" of a certain structure or data set.Geodesics are very important in geometry and physics, but there isone major problem which prevents scientists from using them as avisualization tool: The differential equations for geodesics are verycomplicated and in most cases numerical algorithms must be used.There is always a certain approximation error involved. How canyou be sure to visualize the features and not only the approximationquality. We present here an algorithm to overcome this problem.This paper consists of two parts. In the first, a geometric methodfor the construction of geodesics of arbitrary surfaces is introduced.This method is based on the fundamental property that geodesicsare a generalization of straight lines on plains. In the second partthese geodesics are used to generate local nets on the surfaces."
IEEE Visualization	Visual cues for imminent object contact in realistic virtual environment.	Helen H. Hu,Amy Gooch,William B. Thompson,Brian E. Smits,John J. Rieser,Peter Shirley	2000	Distance judgments are difficult in current virtual environments,limiting their effectiveness in conveying spatial information. Thisproblem is apparent when contact occurs while a user is manipulatingobjects. In particular, the computer graphics used to supportcurrent generation immersive interfaces does a poor job of providingthe visual cues necessary to perceive when contact between objectsis about to occur. This perception of imminent contact is importantin human motor control. Its absence prevents a sense ofnaturalness in interactive displays which allow for object manipulation.This paper reports results from an experiment evaluating theeffectiveness of binocular disparity, cast shadows, and diffuse interreflectionsin signaling imminent contact in a manipulation task.
IEEE Visualization	FastSplats: optimized splatting on rectilinear grids.	Jian Huang,Roger Crawfis,Naeem Shareef,Klaus Mueller	2000	FastSplats: optimized splatting on rectilinear grids.
IEEE Visualization	A visibility determination algorithm for interactive virtual endoscopy.	Rami Hietala,Jarkko Oikarinen	2000	We present a new visibility determination algorithm for interactivevirtual endoscopy. The algorithm uses a modified version oftemplate-based ray casting to extract a view dependent set of potentiallyvisible voxels from volume data. The voxels are triangulatedby Marching Cubes and the triangles are rendered onto thedisplay by a graphics accelerator. Early ray termination and spaceleaping are used to accelerate the ray casting step and a quadtreesubdivision algorithm is used to reduce the number of cast rays.Compared to other recently proposed rendering algorithms for virtualendoscopy, our rendering algorithm does not require a longpreprocessing step or a high-end graphics workstation, but achievesinteractive frame rates on a standard PC equipped with a low-costgraphics accelerator.
IEEE Visualization	Fairing of non-manifolds for visualization.	Andreas Hubeli,Markus H. Gross	2000	The concept of fairing applied to irregular triangular meshes hasbecome more and more important. Previous contributions constructedbetter fairing operators, and applied them both to multiresolutionediting tools and to multiresolution representations ofmeshes. In this paper, we generalize these powerful techniques tohandle non-manifold models. Our framework computes a multi-levelfairing of models by fairing both the two-manifold surfacesthat define the model, the so-called two-features, and all theboundary and intersection curves of the model, the so-called one-features.In addition we introduce two extensions that can be usedin our framework as well as in manifold fairing concepts: an exactlocal volume preservation strategy and a method for feature preservation.Our framework works with any of the manifold fairingoperators for meshes.
IEEE Visualization	Interactive visualization of protein dynamics.	Henk Huitema,Robert van Liere	2000	The study of time dependent characteristics of proteins is importantfor gaining insight into many biological processes. However,visualizing protein dynamics by animating atom trajectories doesnot provide satisfactory results. When the trajectory is sampledwith large times steps, the impression of smooth motion will be destroyeddue to the effects of temporal aliasing. Sampling with smalltime steps will result in the camouflage of interesting motions.In this case study, we discuss techniques for the interactive 3Dvisualization of the dynamics of the Photoactive Yellow Protein.We use essential dynamics methods to filter out uninteresting atommotions from the larger concerted motions. In this way, clear andconcise 3D animations of protein motions can be produced. In addition,we discuss various interactive techniques that allow explorationof the essential subspace of the protein. We discuss the meritsof these techniques when applied to the analysis of the yellowprotein.
IEEE Visualization	Creating reusable visualizations with the relational visualization notation.	Matthew C. Humphrey	2000	"Richly expressive information visualizations are difficult to designand rarely found. Few software tools can generate multi-dimensionalvisualizations at all, let alone incorporate artisticdetail. Although, it is a great efficiency to reuse thesevisualizations with new data, the associated artistic detail is rarelyreusable. The Relational Visualization Notation is a newtechnique and toolkit for specifying highly expressive graphicalrepresentations of data without traditional programming. We seekto discover the accessible power of this notation - both itsgraphical expressiveness and its ease of re-use. Towards this endwe have used the system to reconstruct Minard's visualization ofNapoleon's Russian campaign of 1812. The resulting image isstrikingly similar to the original, and the design is straightforwardto construct. Furthermore, the design permitted by the notationcan be directly reused to visualize Hitler's WWII defeat beforeMoscow. This experience leads us to believe that artisticallyexpressive visualizations can be made to be reusable."
IEEE Visualization	Uniform frequency images: adding geometry to images to produce space-efficient textures.	Adam Hunter,Jonathan D. Cohen	2000	"We discuss the concept of uniform frequency images, whichexhibit uniform local frequency properties. Such images makeoptimal use of space when sampled close to their Nyquist limit.A warping function may be applied to an arbitrary image to redistributeits local frequency content, reducing its highest frequenciesand increasing its lowest frequencies in order to approachthis uniform frequency ideal. The warped image maythen be downsampled according to its new, reduced Nyquistlimit, thereby reducing its storage requirements. To reconstructthe original image, the inverse warp is applied.We present a general, top-down algorithm to automaticallygenerate a piecewise-linear warping function with this frequencybalancing property for a given input image. The image sizeis reduced by applying the warp and then downsampling. Westore this warped, downsampled image plus a small number ofpolygons with texture coordinates to describe the inverse warp.The original image is later reconstructed by rendering the associatedpolygons with the warped image applied as a texture map,a process which is easily accelerated by current graphics hardware.As compared to previous image compression techniques,we generate a similar graceful space-quality tradeoff with theadvantage of being able to ""uncompress"" images during rendering.We report results for several images with sizes ranging from15,000 to 300,000 pixels, achieving reduction rates of 70-90%with improved quality over downsampling alone."
IEEE Visualization	A spreadsheet interface for visualization exploration.	T. J. Jankun-Kelly,Kwan-Liu Ma	2000	"As the size and complexity of data sets continues to increase, thedevelopment of user interfaces and interaction techniques that expeditethe process of exploring that data must receive new attention.Regardless of the speed of rendering, it is important to coherentlyorganize the visual process of exploration: this information bothgrants insights about the data to a user and can be used by collaboratorsto understand the results. To fulfill these needs, we presenta spreadsheet-like interface to data exploration. The interface displaysa 2-dimensional window into visualization parameter spacewhich users manipulate as they search for desired results. Throughtabular organization and a clear correspondence between parametersand results, the interface eases the discovery, comparison andanalysis of the underlying data. Users can utilize operators and theintegrated interpreter to further explore and automate the visualizationprocess; using a method introduced in this paper, these operationscan be applied to ""cells"" in different ""stacks"" of the interface.Via illustrations using a variety of data sets, we demonstrate theefficacy of this novel interface."
IEEE Visualization	Hardware-accelerated texture advection for unsteady flow visualization.	Bruno Jobard,Gordon Erlebacher,M. Yousuff Hussaini	2000	We present a novel hardware-accelerated texture advectionalgorithm to visualize the motion of two-dimensional unsteadyflows. Making use of several proposed extensions to theOpenGL-1.2 specification, we demonstrate animations of over65,000 particles at 2 frames/sec on an SGI Octane with EMXIgraphics. High image quality is achieved by careful attention toedge effects, noise frequency, and image enhancement. Weprovide a detailed description of the hardware implementation,including temporal and spatial coherence techniques, dyeadvection techniques, and feature extraction.
IEEE Visualization	Visualization of multi-dimensional data with vector-fusion.	R. R. Johnson	2000	"Multi-dimensional entities are modeled, displayed, andunderstood with a new algorithm vectorizing data of anydimensionality. This algorithm is called SBP; it is a vectorizedgeneralization of parallel coordinates. Classic geometries of anydimensionality can be demonstrated to facilitate perception andunderstanding of the shapes generated by this algorithm. SBPimages of a 4D line, a circle, and 3D and 4D spherical helicesare shown. A strategy for synthesizing multi-dimensionalmodels matching multi-dimensional data is presented. Currrentapplications include data mining; modeling data-definedstructures of scientific interest such as protein structure andCalabi-Yau figures as multi-dimensional geometric entities;generating vector-fused data signature ""finger prints"" of classicfrequency spectra that identify substances; and treating complextargets as multi-dimensional entities for automatic targetrecognition. SBP Vector Data Signatures apply to all patternrecognition problems."
IEEE Visualization	3D digital cleansing using segmentation rays.	Sarang Lakare,Ming Wan,Mie Sato,Arie E. Kaufman	2000	We propose a novel approach for segmentation and digital cleansingof endoscopic organs. Our method can be used for a varietyof segmentation needs with little or no modification. It aims atfulfilling the dual and often conflicting requirements of a fast andaccurate segmentation and also eliminates the undesirable partialvolume effect which contemporary approaches cannot. For segmentationand digital cleansing, we use the peculiar characteristics exhibitedby the intersection of any two distinct-intensity regions. Todetect these intersections we cast rays through the volume, whichwe call the segmentation rays as they assist in the segmentation.We then associate a certain task of reconstruction and classificationwith each intersection the ray detects. We further use volumetriccontrast enhancement to reconstruct surface lost by segmentation(if any), which aids in improving the quality of the volume rendering.
IEEE Visualization	Shock and vortex visualization using a combined visual/Haptic interface.	Dale A. Lawrence,Christopher D. Lee,Lucy Y. Pao,Roman Y. Novoselov	2000	Specific rendering modes are developed for a combined visual/haptic interface to allow exploration and understanding of fluiddynamics data. The focus is on visualization of shock surfacesand vortex cores. Advantages provided by augmenting traditionalgraphical rendering modes with haptic rendering modes are discussed.Particular emphasis is placed on synergistic combinationsof visual and haptic modes which enable rapid, exploratory interationwith the data. Implementation issues are also discussed.
IEEE Visualization	Visualization of time dependent confocal microscopy data.	Wim C. de Leeuw,Robert van Liere,Pernette J. Verschure,Roel van Driel,Astrid E. Visser,Erik M. M. Manders	2000	Visualization of time dependent confocal microscopy data.
IEEE Visualization	Interactive visualization of particle-in-cell simulations.	Patric Ljung,Mark Dieckmann,Niclas Andersson,Anders Ynnerman	2000	In this work we present a visualization system for interactive real-timeanimation and visualization of simulation results from a par-allelParticle-In-Cell code. The system was designed and imple-mentedfor the Onyx2 InifiniteReality hardware. A number of differentvisual objects, such as volume rendered particle density functionalswere implemented. To provide sufficient frame rates forinteractive visualization the system was designed to provide performanceclose to the hardware specifications both in terms of the I/Oand graphics subsystems. The presented case study applies the developed system to theevolution of an instability that gives rise to a plasma surfatron, amechanism which rapidly can accelerate particles to very high velocitiesand thus be of great importance in the context of electronacceleration in astrophysical shocks, in the solar corona and in particleaccelerators. The produced visualizations have allowed us toidentify a previously unknown saturation mechanism for the surfatronand direct research efforts into new areas of interest.
IEEE Visualization	Topology preserving compression of 2D vector fields.	Suresh K. Lodha,Jose C. Renteria,Krishna M. Roskin	2000	"We present an algorithm for compressing 2D vector fields thatpreserves topology. Our approach is to simplify the given dataset using constrained clustering. We employ different types ofglobal and local error metrics including the earth mover's distancemetric to measure the degradation in topology as well as weightedmagnitude and angular errors. As a result, we obtain precise errorbounds in the compressed vector fields. Experiments with bothanalytic and simulated data sets are presented. Results indicatethat one can obtain significant compression with low errors withoutlosing topology information."
IEEE Visualization	Mastering interactive surface rendering for Java-based diagnostic applications.	Lukas Mroz,Rainer Wegenkittl,Eduard Gröller	2000	The display of iso-surfaces in medical data sets is an important visualizationtechnique used by radiologists for the diagnosis of volumetricdensity data sets. The demands put by radiologists on sucha display technique are interactivity, multiple stacked transparentsurfaces and cutting planes that allow an interactive clipping of thesurfaces. This paper presents a Java based, platform independentimplementation of a very fast surface rendering algorithm whichcombines the advantages of explicit surface representation, splatting,and shear-warp projection to fulfill all these requirements. Thealgorithm is implemented within the context of J-Vision, an applicationfor viewing and diagnosing medical images which is currentlyin use at various hospitals.
IEEE Visualization	Achieving color uniformity across multi-projector displays.	Aditi Majumder,Zhu He,Herman Towles,Greg Welch	2000	Large area tiled displays are gaining popularity for use in collaborativeimmersive virtual environments and scientific visualization.While recent work has addressed the issues of geometric registration,rendering architectures, and human interfaces, there hasbeen relatively little work on photometric calibration in general,and photometric non-uniformity in particular. For example, as a resultof differences in the photometric characteristics of projectors,the color and intensity of a large area display varies from place toplace. Further, the imagery typically appears brighter at the regionsof overlap between adjacent projectors. In this paper we analyze and classify the causes of photometricnon-uniformity in a tiled display. We then propose a methodologyfor determining corrections designed to achieve uniformity, that cancorrect for the photometric variations across a tiled projector displayin real time using per channel color look-up-tables (LUT).
IEEE Visualization	Basic research for coloring multichannel MRI data.	Shigeru Muraki,Toshiharu Nakai,Yasuyo Kita	2000	This is basic research for assigning color values to voxels ofmultichannel MRI volume data. The MRI volume datasetsobtained under different scanning conditions are transformed tothe components by independent component analysis (ICA), whichenhances physical characteristics of the tissue. The transferfunctions for generating color values from independentcomponents are obtained using the radial basis function network,a kind of neural net, by training the network with sample datachosen from visible female data set. The resultant color volumedatasets correspond well with the full-color cross-sections of thevisible human datasets.
IEEE Visualization	Visualizing volume data using physical models.	David R. Nadeau,Michael J. Bailey	2000	Visualization techniques enable scientists to interactively explore3D data sets, segmenting and cutting them to reveal innerstructure. While powerful, these techniques suffer from oneserious flaw - the images they create are displayed on a flat pieceof glass or paper. It is not really 3D - it only can be made toappear 3D. In this paper we describe the construction of 3Dphysical models from volumetric data. Using solid freeformfabrication equipment, these models are built as separateinterlocking pieces that express in physical form the segmentationand cutting operations common in display-based visualization.
IEEE Visualization	Interior/exterior classification of polygonal models.	Fakir S. Nooruddin,Greg Turk	2000	We present an algorithm for automatically classifying theinterior and exterior parts of a polygonal model. The needfor visualizing the interiors of objects frequently arises inmedical visualization and CAD modeling. The goal of suchvisualizations is to display the model in a way that the humanobserver can easily understand the relationship betweenthe different parts of the surface. While there exist excellentmethods for visualizing surfaces that are inside one another(nested surfaces), the determination of which parts of thesurface are interior is currently done manually.Our automatic method for interior classification takes asampling approach using a collection of direction vectors.Polygons are said to be interior to the model if they are notvisible in any of these viewing directions from a point outsidethe model. Once we have identified polygons as beinginside or outside the model, these can be textured or havedifferent opacities applied to them so that the whole modelcan be rendered in a more comprehensible manner. An additionalconsideration for some models is that they may haveholes or tunnels running through them that are connectedto the exterior surface. Although an external observer cansee into these holes, it is often desirable to mark the walls ofsuch tunnels as being part of the interior of a model. In orderto allow this modified classification of the interior, we usemorphological operators to close all the holes of the model.An input model is used together with its closed version toprovide a better classification of the portions of the originalmodel.
IEEE Visualization	The transfer function bake-off (panel session).	Hanspeter Pfister,William E. Lorensen,William J. Schroeder,Chandrajit L. Bajaj,Gordon L. Kindlmann	2000	The transfer function bake-off (panel session).
IEEE Visualization	Image based rendering with stable frame rates.	Huamin Qu,Ming Wan,Jiafa Qin,Arie E. Kaufman	2000	This paper presents an efficient keyframeless image-based renderingtechnique. An intermediate image is used to exploit the coherencesamong neighboring frames. The pixels in the intermediateimage are first rendered by a ray-casting method and then warpedto the intermediate image at the current viewpoint and view direction.We use an offset buffer to record the precise positions of thesepixels in the intermediate image. Every frame is generated in threesteps: warping the intermediate image onto the frame, filling inholes, and selectively rendering a group of old pixels. By dynamicallyadjusting the number of those old pixels in the last step, theworkload at every frame can be balanced. The pixels generated bythe last two steps make contributions to the new intermediate image.Unlike occasional keyframes in conventional image-based renderingwhich need to be totally rerendered, intermediate images onlyneed to be partially updated at every frame. In this way, we guaranteemore stable frame rates and more uniform image qualities. Theintermediate image can be warped efficiently by a modified incremental3D warp algorithm. As a specific application, we demonstrateour technique with a voxel-based terrain rendering system.
IEEE Visualization	Visualizing high-dimensional predicitive model quality.	Penny Rheingans,Marie desJardins	2000	Visualizing high-dimensional predicitive model quality.
IEEE Visualization	"The impact of computer games on scientific & information visualization: ""if you can't beat them, join them"" (panel session)."	Theresa-Marie Rhyne,Peter Doenges,William L. Hibbard,Hanspeter Pfister,Nate Robins	2000	"The impact of computer games on scientific & information visualization: ""if you can't beat them, join them"" (panel session)."
IEEE Visualization	Fast visualization methods for comparing dynamics: a case study in combustion.	Kay A. Robbins,Michael Gorman	2000	Visualization can be an important tool for displaying, categorizingand digesting large quantities of inter-related information duringlaboratory and simulation experiments. Summary visualizationsthat compare and represent data sets in the context of a collectionare particularly valuable. Applicable visualizations used in thesesettings must be fast (near real time) and should allow the additionof data sets as they are acquired without requiring rerenderingof the visualization. This paper examines several visualizationtechniques for representing collections of data sets in a combustionexperiment including spectral displays, tiling and geometric mappingsof symmetry. The application provides insight into how suchvisualizations might be used in practical real-time settings to assistin exploration and in conducting parameter space surveys.
IEEE Visualization	Hardware-accelerated volume and isosurface rendering based on cell-projection.	Stefan Röttger,Martin Kraus,Thomas Ertl	2000	We present two beneficial rendering extensions to the ProjectedTetrahedra (PT) algorithm by Shirley and Tuchman. These extensionsare compatible with any cell sorting technique, for examplethe BSP-XMPVO sorting algorithm for unstructured meshes.Using 3D texture mapping our first extension solves the long-standingproblem of hardware-accelerated but accurate renderingof tetrahedral volume cells with arbitrary transfer functions.By employing 2D texture mapping our second extension realizesthe hardware-accelerated rendering of multiple shaded isosurfaceswithin the PT algorithm without reconstructing the isosurfaces.Additionally, two methods are presented to combine projectedtetrahedral volumes with isosurfaces. The time complexity of allour algorithms is linear in the number of tetrahedra and does neitherdepend on the number of isosurfaces nor on the employed transferfunctions.
IEEE Visualization	Tetrahedron based, least squares, progressive volume models with application to freehand ultrasound data.	Tom Roxborough,Gregory M. Nielson	2000	In this paper we present a new method for the modeling offreehand collected three-dimensional ultrasound data. The modelis piece-wise linear and based upon progressive tetrahedraldomains created by a subdivision scheme which splits atetrahedron on its longest edge and guarantees a validtetrahedrization. Least squares error is used to characterize themodel and an effective iterative technique is used to compute thevalues of the model at the vertices of the tetrahedral grid. Sincethe subdivision strategy is adaptive, the complexity of the modelconforms to the complexity of the data leading to an extremelyefficient and highly compressed volume model. The model isevaluated in real time using piece-wise linear interpolation, andgives a medical professional the chance to see images whichwould not be possible using conventional ultrasound techniques.
IEEE Visualization	Visualizing DIII-D Tokamak magnetic field lines.	Gregory L. Schussman,Kwan-Liu Ma,David P. Schissel,Todd Evans	2000	"We demonstrate the use of a combination of perceptually effective techniques for visualizing magnetic field data from the DIII-D Tokamak.These techniques can be implemented to run very efficiently on machines with hardware support for OpenGL. Interactive speeds facilitate clear communication of magnetic field structure, enhancing fusion scientists ' understanding of their data, and thereby accelerating their research."
IEEE Visualization	Enabling level-of-detail matching for exterior scene synthesis.	Randy K. Scoggins,Robert J. Moorhead,Raghu Machiraju	2000	Enabling level-of-detail matching for exterior scene synthesis.
IEEE Visualization	Multi-resolution dynamic meshes with arbitrary deformations.	Ariel Shamir,Chandrajit L. Bajaj,Valerio Pascucci	2000	Multi-resolution dynamic meshes with arbitrary deformations.
IEEE Visualization	H-BLOB: a hierarchical visual clustering method using implicit surfaces.	T. C. Sprenger,R. Brunella,Markus H. Gross	2000	In this paper, we present a new hierarchical clustering and visualizationalgorithm called H-BLOB, which groups and visualizescluster hierarchies at multiple levels-of-detail. Our method is fundamentallydifferent to conventional clustering algorithms, such asC-means, K-means, or linkage methods that are primarily designedto partition a collection of objects into subsets sharing similarattributes. These approaches usually lack an efficient level-of-detailstrategy that breaks down the visual complexity of very largedatasets for visualization. In contrast, our method combines groupingand visualization in a two stage process constructing a hierarchicalsetting. In the first stage a cluster tree is computed makinguse of an edge contraction operator. Exploiting the inherent hierarchicalstructure of this tree, a second stage visualizes the clustersby computing a hierarchy of implicit surfaces. We believe that H-BLOBis especially suited for the visualization of very largedatasets and for visual decision making in information visualization.The versatility of the algorithm is demonstrated using examplesfrom visual data mining.
IEEE Visualization	Scientific visualization of water quality in the Chesapeake Bay.	Robert Stein,Alan M. Shih,M. Pauline Baker,Carl F. Cerco,Mark R. Noel	2000	This paper describes our experience in designing andbuilding a tool for visualizing the results of the CE-QUAL-ICMThree-Dimensional Eutrophication Model, as applied to waterquality in the Chesapeake Bay. This model outputs a highlymultidimensional dataset over very many timesteps - outstrippingthe capabilities of the visualization tools available to the researchteam. As part of the Army Engineer Research and DevelopmentCenter (ERDC) Programming Environment and Training (PET)project, a special visualization tool was developed. This paperincludes discussions on how the simulation data are handledefficiently, as well as how the issues of usability, flexibility andcollaboration are addressed.
IEEE Visualization	Extracting regions of interest applying a local watershed transformation.	Stanislav L. Stoev,Wolfgang Straßer	2000	In this paper, we present a new technique for extracting regions ofinterest (ROI) applying a local watershed transformation. The proposedstrategy for computing catchment basins in a given region ofinterest is based on a rain-falling simulation. Unlike the standardwatershed algorithms, which flood the complete (gradient magnitudeof an) image, the proposed approach allows us to perform thistask locally. Thus, a controlled region growth is performed, savingtime and reducing the memory requirement especially when appliedon volume data.A second problem arising from the standard watershed transformationis the over-segmented result and the lack of sound criteriafor merging the computed basins. For overcoming this drawback,we present a basin-merging strategy introducing four criteria formerging adjacent basins. The threshold values applied in this strategyare derived from the user input and match rather the attributesof the selected object than of all objects in the image. In doing so,the user is not required to adjust abstract numbers, but to simply selecta coarse region of interest. Moreover, the proposed algorithmis not limited to the 2D case. As we show in this work, it is suitablefor volume data processing as well. Finally, we present the resultsof applying the proposed approach on several example images andvolume data sets.
IEEE Visualization	Simplification of surface annotations.	Frank Suits,James T. Klosowski,William Horn,Gérard Lecina	2000	Geometric models are often annotated to provide additional informationduring visualization. Maps may be marked with rivers,roads, or topographical information, and CAD data models mayhighlight the underlying mesh structure. While this additional informationmay be extremely useful, there is a rendering cost associatedwith it. Texture maps have often been used to convey thisinformation at relatively low cost, but they suffer from blurring andpixelization at high magnification.We present a technique for simplifying surface annotations basedon directed, asymmetric tolerance. By maintaining the annotationsas geometry, as opposed to textures, we are able to simplify themwhile still maintaining the overall appearance of the model over awide range of magnifications. Texture maps may still be used toprovide low-resolution surface detail, such as color.We demonstrate a significant gain in rendering performancewhile retaining the original appearance of objects from many applicationdomains.
IEEE Visualization	A computational steering system for studying microwave interactions with missile bodies.	J. Edward Swan II,Marco Lanzagorta,Doug Maxwell,Eddy Kuo,Jeffrey K. Uhlmann,Wendell Anderson,Haw-Jye Shyu,William Smith	2000	A computational steering system for studying microwave interactions with missile bodies.
IEEE Visualization	Pen-and-Ink rendering in volume visualisation.	Steve M. F. Treavett,Min Chen	2000	This paper is concerned with the development of non-photorealisticrendering techniques for volume visualisation.In particular, we present two pen-and-ink renderingmethods, a 3D method based on non-photorealistic solidtextures, and a 2D method that involves two renderingphases in the object space and image space respectively.As both techniques utilize volume- and image-based datarepresentations, they can be built upon a traditional volumerendering pipeline, and be integrated with photorealisticmethods available in such a pipeline. We demonstratethat such an integration facilitates an effective mechanismfor enhancing visualisation and its interpretation.
IEEE Visualization	Visual data fusion for applications of high-resolution numerical weather prediction.	Lloyd Treinish	2000	Visual data fusion for applications of high-resolution numerical weather prediction.
IEEE Visualization	Multi-resolution visualization techniques for nested weather models.	Lloyd Treinish	2000	Scaling of simulations challenge the effectiveness of conventionalvisualization methods. This problem becomes two-fold formesoscale weather models that operate in near-real-time atcloud-scale resolution. For example, typical approaches to vectorfield visualization (e.g., wind) are based upon global methods,which may not illustrate detailed structure. In addition,such computations employ multi-resolution meshes to capturesmall-scale phenomena, which are not properly reflected in bothvector and scalar realizations. To address the former, criticalpoint analysis and simple bandpass filtering of wind fields isemployed for better seed point identification for streamline calculations.For the latter, an encapsulation of nested computationalmeshes is developed for general realization. It is thencombined with the seed point calculation for an improved vectorvisualization of multi-resolution weather forecasting data.
IEEE Visualization	A topology simplification method for 2D vector fields.	Xavier Tricoche,Gerik Scheuermann,Hans Hagen	2000	Topology analysis of plane, turbulent vector fields results in visualclutter caused by critical points indicating vortices of finer and finerscales. A simplification can be achieved by merging critical pointswithin a prescribed radius into higher order critical points. Afterbuilding clusters containing the singularities to merge, the methodgenerates a piecewise linear representation of the vector field ineach cluster containing only one (higher order) singularity. Anyvisualization method can be applied to the result after this process.Using different maximal distances for the critical points tobe merged results in a hierarchy of simplified vector fields that canbe used for analysis on different scales.
IEEE Visualization	A flow-guided streamline seeding strategy.	Vivek Verma,David T. Kao,Alex Pang	2000	This paper presents a seed placement strategy for streamlines basedon flow features in the dataset. The primary goal of our seedingstrategy is to capture flow patterns in the vicinity of critical pointsin the flow field, even as the density of streamlines is reduced. Secondarygoals are to place streamlines such that there is sufficientcoverage in non-critical regions, and to vary the streamline placementsand lengths so that the overall presentation is aestheticallypleasing (avoid clustering of streamlines, avoid sharp discontinuitiesacross several streamlines, etc.). The procedure is straight forwardand non-iterative. First, critical points are identified. Next,the flow field is segmented into regions, each containing a singlecritical point. The critical point in each region is then seeded with atemplate depending on the type of critical point. Finally, additionalseed points are randomly distributed around the field using a Poissondisk distribution to minimize closely spaced seed points. Themain advantage of this approach is that it does not miss the featuresaround critical points. Since the strategy is not image-guided, andhence not view dependent, significant savings are possible whenexamining flow fields from different viewpoints, especially for 3Dflow fields.
IEEE Visualization	Mastering interactive virtual Bronchioscopy on a Low-end PC.	Anna Vilanova,Balint Hegedüs,Eduard Gröller,Daniel Wagner,Rainer Wegenkittl,Martin C. Freund	2000	Mastering interactive virtual Bronchioscopy on a Low-end PC.
IEEE Visualization	Scanline surfacing: building separating surfaces from planar contours.	David M. Weinstein	2000	A standard way to segment medical imaging datasets is by tracingcontours around regions of interest in parallel planar slices. Unfortunately,the standard methods for reconstructing three dimensionalsurfaces from those planar contours tend to be either complicatedor not very robust. Furthermore, they fail to consistently mesh abuttingstructures which share portions of contours. In this paper wepresent a novel, straight-forward algorithm for accurately and automaticallyreconstructing surfaces from planar contours. Our algorithmis based on scanline rendering and separating surface extraction.By rendering the contours as distinctly colored polygonsand reading back each rendered slice into a segmented volume, wereduce the complex problem of building a surface from planar contoursto the much simpler problem of extracting separating surfacesfrom a classified volume. Our scanline surfacing algorithmrobustly handles complex surface topologies such as bifurcations,embedded features, and abutting surfaces.
IEEE Visualization	Four-dimensional non-linear ray tracing as a visualization tool for gravitational physics.	Daniel Weiskopf	2000	In this paper, general relativistic ray tracing is presented as a tool forgravitational physics. It is shown how standard three-dimensionalray tracing can be extended to allow for general relativistic visualization.This visualization technique provides images as seen byan observer under the influence of a gravitational field and allowsto probe spacetime by null geodesics. Moreover, a technique isproposed for visualizing the caustic surfaces generated by a gravitationallens. The suitability of general relativistic ray tracing isdemonstrated by means of two examples, namely the visualizationof the rigidly rotating disk of dust and the warp drive metric.
IEEE Visualization	Real-world relativity: image-based special relativistic visualization.	Daniel Weiskopf,Daniel Kobras,Hanns Ruder	2000	This paper describes a novel rendering technique for special relativisticvisualization. It is an image-based method which allowsto render high speed flights through real-world scenes filmed bya standard camera. The relativistic effects on image generation aredetermined by the relativistic aberration of light, the Doppler effect,and the searchlight effect. These account for changes of apparentgeometry, color, and brightness of the objects. It is shown howthe relativistic effects can be taken into account by a modificationof the plenoptic function. Therefore, all known image-based non-relativisticrendering methods can easily be extended to incorporaterelativistic rendering. Our implementation allows interactive viewingof relativistic panoramas and the production of movies whichshow super-fast travel. Examples in the form of snapshots and filmsequences are included.
IEEE Visualization	A level-set method for flow visualization.	Rüdiger Westermann,Christopher Johnson,Thomas Ertl	2000	In this paper we propose a technique for visualizing steady flow.Using this technique, we first convert the vector field data into ascalar level-set representation. We then analyze the dynamic behaviorand subsequent distortion of level-sets and interactively monitorthe evolving structures by means of texture-based surface rendering.Next, we combine geometrical and topological considerationsto derive a multiscale representation and to implement a methodfor the automatic placement of a sparse set of graphical primitivesdepicting homogeneous streams in the fields. Using the resulting algorithms,we have built a visualization system that enables us to effectivelydisplay the flow direction and its dynamics even for dense3D fields.
IEEE Visualization	Vector fields simplification - a case study of visualizing climate modeling and simulation data sets.	Pak Chung Wong,Harlan Foote,L. Ruby Leung,Elizabeth Jurrus,Dan Adams,Jim Thomas	2000	In our study of regional climate modeling and simulation, wefrequently encounter vector fields that are crowded with largenumbers of critical points. A critical point in a flow is where thevector field vanishes. While these critical points accurately reflectthe topology of the vector fields, in our study only a subset ofthem is worth further investigation. We present a filtering techniquebased on the vorticity of the vector fields to eliminate theless interesting and sometimes sporadic critical points in a multiresolutionfashion. The neighboring regions of the preservedfeatures, which are characterized by strong shear and circulation,are potential locations of weather instability. We apply our feature-filtering technique to a regional climate modeling data setcovering East Asia in the summer of 1991.
IEEE Visualization	Semi-regular mesh extraction from volumes.	Zoë J. Wood,Peter Schröder,David E. Breen,Mathieu Desbrun	2000	Semi-regular mesh extraction from volumes.
IEEE Visualization	Navigating high-dimensional spaces to support design steering.	Helen Wright,Ken Brodlie,Tim David	2000	Throughout the design cycle, visualization, whether a sketchscribbled on the back of a spare piece of paper or a fully detaileddrawing, has been the mainstay of design: we need to see theproduct. One of the most important stages of the design cycle isthe initial, or concept, stage and it is here that design variantsoccur in large numbers to be vetted quickly. At this initial stagethe human element - the designer - is crucial to the success ofthe product.In this paper we describe an interactive environment for conceptdesign which recognises the needs of the designer, not only tosee the product and make rapid modifications, but also tomonitor the progress of their design towards some preferredsolution. This leads to the notion of a design parameter space,typically high-dimensional, which must also be visualized inaddition to the product itself. Using a module developed for IRISExplorer (TM) , design steering is presented as a navigation of thisspace in order to search for optimal designs, either manually orby local optimisation.
IEEE Visualization	On-the-Fly rendering of losslessly compressed irregular volume data.	Chuan-Kai Yang,Tulika Mitra,Tzi-cker Chiueh	2000	On-the-Fly rendering of losslessly compressed irregular volume data.
IEEE Visualization	IEEE Visualization 2000, October 8-13, 2000, Hilton Hotel, Salt Lake City, Utah, USA, Proceedings.		2000	IEEE Visualization 2000, October 8-13, 2000, Hilton Hotel, Salt Lake City, Utah, USA, Proceedings.
IEEE Trans. Vis. Comput. Graph.	Anisotropic Diffusion in Vector Field Visualization on Euclidean Domains and Surfaces.	Udo Diewald,Tobias Preußer,Martin Rumpf	2000	Vector field visualization is an important topic in scientific visualization. Its aim is to graphically represent field data on two and three-dimensional domains and on surfaces in an intuitively understandable way. Here, a new approach based on anisotropic nonlinear diffusion is introduced. It enables an easy perception of vector field data and serves as an appropriate scale space method for the visualization of complicated flow pattern. The approach is closely related to nonlinear diffusion methods in image analysis where images are smoothed while still retaining and enhancing edges. Here, an initial noisy image intensity is smoothed along integral lines, whereas the image is sharpened in the orthogonal direction. The method is based on a continuous model and requires the solution of a parabolic PDE problem. It is discretized only in the final implementational step. Therefore, many important qualitative aspects can already be discussed on a continuous level. Applications are shown for flow fields in 2D and 3D, as well as for principal directions of curvature on general triangulated surfaces. Furthermore, the provisions for flow segmentation are outlined.
IEEE Trans. Vis. Comput. Graph.	"Guest Editor's Introduction: Special Section on Visualization."	David S. Ebert	2000	"Guest Editor's Introduction: Special Section on Visualization."
IEEE Trans. Vis. Comput. Graph.	Visual Discovery and Analysis.	Stephen G. Eick	2000	"We have developed a flexible software environment called ADVIZOR for visual information discovery. ADVIZOR complements existing assumptive-based analyses by providing a discovery-based approach. ADVIZOR consists of five parts: a rich set of flexible visual components, strategies for arranging the components for particular analyses, an in-memory data pool, data manipulation components, and container applications. Working together, ADVIZOR's architecture provides a powerful production platform for creating innovative visual query and analysis applications."
IEEE Trans. Vis. Comput. Graph.	A Functional Framework for Web-Based Information Visualization Systems.	Michael Bender,Ralf Klein,Andreas Disch,Achim Ebert	2000	"The accelerating evolution in information visualization research in the last few years led to several specific system implementations. The obvious drawbacks of this development are highly dependent software systems, which are only available for a restricted number of users. Today, due to the remarkable advances in hardware and software technologies, not only very expensive graphics workstations, but also low-cost PCs are capable of running computational demanding visualization systems. Furthermore, the rapid development of the medium World Wide Web coming along with state-of-the-art internet programming techniques led to a trend toward more generally usable visualization systems. In this paper, we propose a functional developer's framework for general Web-based visualization systems which makes intelligent use of application specific software and hardware components on the server side, as well as of Java's benefits on the client side. To demonstrate the framework's abilities, we have applied it to two practical visualization tasks and we will report on our experience concerning practicability and pitfalls."
IEEE Trans. Vis. Comput. Graph.	Structure-Based Brushes: A Mechanism for Navigating Hierarchically Organized Data and Information Spaces.	Ying-Huey Fua,Matthew O. Ward,Elke A. Rundensteiner	2000	Interactive selection is a critical component in exploratory visualization, allowing users to isolate subsets of the displayed information for highlighting, deleting, analysis, or focused investigation. Brushing, a popular method for implementing the selection process, has traditionally been performed in either screen space or data space. In this paper, we introduce an alternate, and potentially powerful, mode of selection that we term structure-based brushing, for selection in data sets with natural or imposed structure. Our initial implementation has focused on hierarchically structured data, specifically very large multivariate data sets structured via hierarchical clustering and partitioning algorithms. The structure-based brush allows users to navigate hierarchies by specifying focal extents and level-of-detail on a visual representation of the structure. Proximity-based coloring, which maps similar colors to data that are closely related within the structure, helps convey both structural relationships and anomalies. We describe the design and implementation of our structure-based brushing tool. We also validate its usefulness using two distinct hierarchical visualization techniques, namely hierarchical parallel coordinates and tree-maps. Finally, we discuss relationships between different classes of brushes and identify methods by which structure-based brushing could be extended to alternate data structures.
IEEE Trans. Vis. Comput. Graph.	Focusing in Algorithm Explanation.	Beatrix Braune,Reinhard Wilhelm	2000	Algorithm Animation attempts to explain an algorithm by visualizing interesting events of the execution of the implemented algorithm on some sample input. Algorithm Explanation describes the algorithm on some adequate level of abstraction, states invariants, explains how important steps of the algorithm preserve the invariants, and abstracts from the input data up to the relevant properties. It uses a small focus onto the execution state. This paper is concerned with the explanation of algorithms on linked data structures. The thesis of the paper is that shape analysis of such algorithms produces abstract representations of such data structures, which focus on the ¿active¿ parts, i.e., the parts of the data structures, which the algorithm can access during its next steps. The paper presents a concept of visually executing an algorithm on these abstract representations of data.
IEEE Trans. Vis. Comput. Graph.	A Perceptually-Driven Parallel Algorithm for Efficient Radiosity Simulation.	Simon Gibson,Roger J. Hubbold	2000	"We describe a novel algorithm for computing view-independent finite-element radiosity solutions on distributed shared-memory parallel architectures. Our approach is based on the notion of a subiteration being the transfer of energy from a single source to a subset of the scene's receiver patches. By using an efficient queue-based scheduling system to process these subiterations, we show how radiosity solutions can be generated without the need for processor synchronization between iterations of the progressive refinement algorithm. The only significant source of interprocessor communication required by our method is for visibility calculations. We also describe a perceptually-driven approach to visibility estimation, which employs an efficient volumetric grid structure and attempts to reduce the amount of interprocessor communication by approximating visibility queries between distant patches. Our algorithm also eliminates the need for dynamic load-balancing until the end of the solution process and is shown to achieve a super-linear speedup in many situations."
IEEE Trans. Vis. Comput. Graph.	Perturbation Methods for Interactive Specular Reflections.	Min Chen,James Arvo	2000	We describe a new approach for interactively approximating specular reflections in arbitrary curved surfaces. The technique is applicable to any smooth implicitly defined reflecting surface that is equipped with a ray intersection procedure; it is also extremely efficient as it employs local perturbations to interpolate point samples analytically. After ray tracing a sparse set of reflection paths with respect to a given vantage point and static reflecting surfaces, the algorithm rapidly approximates reflections of arbitrary points in 3-space by expressing them as perturbations of nearby points with known reflections. The reflection of each new point is approximated to second-order accuracy by applying a closed-form perturbation formula to one or more nearby reflection paths. This formula is derived from the Taylor expansion of a reflection path and is based on first and second-order path derivatives. After preprocessing, the approach is fast enough to compute reflections of tessellated diffuse objects in arbitrary curved surfaces at interactive rates using standard graphics hardware. The resulting images are nearly indistinguishable from ray traced images that take several orders of magnitude longer to generate.
IEEE Trans. Vis. Comput. Graph.	An Order of Magnitude Faster Isosurface Rendering in Software on a PC than Using Dedicated, General Purpose Rendering Hardware.	George J. Grevera,Jayaram K. Udupa,Dewey Odhner	2000	The purpose of this work is to compare the speed of isosurface rendering in software with that using dedicated hardware. Input data consist of 10 different objects from various parts of the body and various modalities (CT, MR, and MRA) with a variety of surface sizes (up to 1 million voxels/2 million triangles) and shapes. The software rendering technique consists of a particular method of voxel-based surface rendering, called shell rendering. The hardware method is OpenGL-based and uses the surfaces constructed from our implementation of the ¿Marching Cubes¿ algorithm. The hardware environment consists of a variety of platforms, including a Sun Ultra I with a Creator3D graphics card and a Silicon Graphics Reality Engine II, both with polygon rendering hardware, and a 300Mhz Pentium PC. The results indicate that the software method (shell rendering) was 18 to 31 times faster than any hardware rendering methods. This work demonstrates that a software implementation of a particular rendering algorithm (shell rendering) can outperform dedicated hardware. We conclude that, for medical surface visualization, expensive dedicated hardware engines are not required. More importantly, available software algorithms (shell rendering) on a 300Mhz Pentium PC outperform the speed of rendering via hardware engines by a factor of 18 to 31.
IEEE Trans. Vis. Comput. Graph.	Strategies for Direct Volume Rendering of Diffusion Tensor Fields.	Gordon L. Kindlmann,David M. Weinstein,David Hart	2000	Diffusion-weighted magnetic resonance imaging is a relatively new modality capable of elucidating the fibrous structure of certain types of tissue, such as the white matter within the brain. One tool for interpreting this data is volume rendering because it permits the visualization of three dimensional structure without a prior segmentation process. In order to use volume rendering, however, we must develop methods for assigning opacity and color to the data, and create a method to shade the data to improve the legibility of the rendering. Previous work introduced three such methods: barycentric opacity maps, hue-balls (for color), and lit-tensors (for shading). The current paper expands on and generalizes these methods, describing and demonstrating further means of generating opacity, color, and shading from the tensor information. We also propose anisotropic reaction-diffusion volume textures as an additional tool for visualizing the structure of diffusion data. The patterns generated by this process can be visualized on their own or they can be used to supplement the volume rendering strategies described in the rest of the paper. Finally, because interpolation between data points is a fundamental issue in volume rendering, we conclude with a discussion and evaluation of three distinct interpolation methods suitable for diffusion tensor MRI data.
IEEE Trans. Vis. Comput. Graph.	The Prioritized-Layered Projection Algorithm for Visible Set Estimation.	James T. Klosowski,Cláudio T. Silva	2000	Prioritized-Layered Projection (PLP) is a technique for fast rendering of high depth complexity scenes. It works by estimating the visible polygons of a scene from a given viewpoint incrementally, one primitive at a time. It is not a conservative technique, instead PLP is suitable for the computation of partially correct images for use as part of time-critical rendering systems. From a very high level, PLP amounts to a modification of a simple view-frustum culling algorithm, however, it requires the computation of a special occupancy-based tessellation and the assignment to each cell of the tessellation a solidity value, which is used to compute a special ordering on how primitives get projected. In this paper, we detail the PLP algorithm, its main components, and implementation. We also provide experimental evidence of its performance, including results on two types of spatial tessellation (using octree- and Delaunay-based tessellations), and several datasets. We also discuss several extensions of our technique.
IEEE Trans. Vis. Comput. Graph.	"Corrections to 'The Prioritized-Layered Projection Algorithm for Visible Set Estimation'."	James T. Klosowski,Cláudio T. Silva	2000	"Corrections to 'The Prioritized-Layered Projection Algorithm for Visible Set Estimation'."
IEEE Trans. Vis. Comput. Graph.	"Editor's Note."	Hans Hagen	2000	"Editor's Note."
IEEE Trans. Vis. Comput. Graph.	Conformal Surface Parameterization for Texture Mapping.	Steven Haker,Sigurd Angenent,Allen Tannenbaum,Ron Kikinis,Guillermo Sapiro,Michael Halle	2000	In this paper, we give an explicit method for mapping any simply connected surface onto the sphere in a manner which preserves angles. This technique relies on certain conformal mappings from differential geometry. Our method provides a new way to automatically assign texture coordinates to complex undulating surfaces. We demonstrate a finite element method that can be used to apply our mapping technique to a triangulated geometric description of a surface.
IEEE Trans. Vis. Comput. Graph.	Graph Visualization and Navigation in Information Visualization: A Survey.	Ivan Herman,Guy Melançon,M. Scott Marshall	2000	This is a survey on graph visualization and navigation techniques, as used in information visualization. Graphs appear in numerous applications such as web browsing, state-transition diagrams, and data structures. The ability to visualize and to navigate in these potentially large, abstract graphs is often a crucial part of an application. Information visualization has specific requirements, which means that this survey approaches the results of traditional graph drawing from a different perspective.
IEEE Trans. Vis. Comput. Graph.	Analysis of Head Pose Accuracy in Augmented Reality.	William A. Hoff,Tyrone Vincent	2000	A method is developed to analyze the accuracy of the relative head-to-object position and orientation (pose) in augmented reality systems with head-mounted displays. From probabilistic estimates of the errors in optical tracking sensors, the uncertainty in head-to-object pose can be computed in the form of a covariance matrix. The positional uncertainty can be visualized as a 3D ellipsoid. One useful benefit of having an explicit representation of uncertainty is that we can fuse sensor data from a combination of fixed and head-mounted sensors in order to improve the overall registration accuracy. The method was applied to the analysis of an experimental augmented reality system, incorporating an optical see-through head-mounted display, a head-mounted CCD camera, and a fixed optical tracking sensor. The uncertainty of the pose of a movable object with respect to the head-mounted display was analyzed. By using both fixed and head mounted sensors, we produced a pose estimate that is significantly more accurate than that produced by either sensor acting alone.
IEEE Trans. Vis. Comput. Graph.	Interactive Virtual Relighting of Real Scenes.	Céline Loscos,George Drettakis,Luc Robert	2000	Computer augmented reality (CAR) is a rapidly emerging field which enables users to mix real and virtual worlds. Our goal is to provide interactive tools to perform common illumination, i.e., light interactions between real and virtual objects, including shadows and relighting (real and virtual light source modification). In particular, we concentrate on virtually modifying real light source intensities and inserting virtual lights and objects into a real scene; such changes can be very useful for virtual lighting design and prototyping. To achieve this, we present a three-step method. We first reconstruct a simplified representation of real scene geometry using semiautomatic vision-based techniques. With the simplified geometry, and by adapting recent hierarchical radiosity algorithms, we construct an approximation of real scene light exchanges. We next perform a preprocessing step, based on the radiosity system, to create unoccluded illumination textures. These replace the original scene textures which contained real light effects such as shadows from real lights. This texture is then modulated by a ratio of the radiosity (which can be changed) over a display factor which corresponds to the radiosity for which occlusion has been ignored. Since our goal is to achieve a convincing relighting effect, rather than an accurate solution, we present a heuristic correction process which results in visually plausible renderings. Finally, we perform an interactive process to compute new illumination with modified real and virtual light intensities. Our results show that we are able to virtually relight real scenes interactively, including modifications and additions of virtual light sources and objects.
IEEE Trans. Vis. Comput. Graph.	Designing Pixel-Oriented Visualization Techniques: Theory and Applications.	Daniel A. Keim	2000	Visualization techniques are of increasing importance in exploring and analyzing large amounts of multidimensional information. One important class of visualization techniques which is particularly interesting for visualizing very large multidimensional data sets is the class of pixel-oriented techniques. The basic idea of pixel-oriented visualization techniques is to represent as many data objects as possible on the screen at the same time by mapping each data value to a pixel of the screen and arranging the pixels adequately. A number of different pixel-oriented visualization techniques have been proposed in recent years and it has been shown that the techniques are useful for visual data exploration in a number of different application contexts. In this paper, we discuss a number of issues which are of high importance in developing pixel-oriented visualization techniques. The major goal of this article is to provide a formal basis of pixel-oriented visualization techniques and show that the design decisions in developing them can be seen as solutions of well-defined optimization problems. This is true for the mapping of the data values to colors, the arrangement of pixels inside the subwindows, the shape of the subwindows, and the ordering of the dimension subwindows. The paper also discusses the design issues of special variants of pixel-oriented techniques for visualizing large spatial data sets. The optimization functions for the mentioned design decisions are important for the effectiveness of the resulting visualizations. We show this by evaluating the optimization functions and comparing the results to the visualizations obtained in a number of different application.
IEEE Trans. Vis. Comput. Graph.	Dynamic Modeling of Butterfly Subdivision Surfaces.	Chhandomay Mandal,Hong Qin,Baba C. Vemuri	2000	In this paper, we develop integrated techniques that unify physics-based modeling with geometric subdivision methodology and present a scheme for dynamic manipulation of the smooth limit surface generated by the (modified) butterfly scheme using physics-based ¿force¿ tools. This procedure-based surface model obtained through butterfly subdivision does not have a closed-form analytic formulation (unlike other well-known spline-based models) and, hence, poses challenging problems to incorporate mass and damping distributions, internal deformation energy, forces, and other physical quantities required to develop a physics-based model. Our primary contributions to computer graphics and geometric modeling include: 1) a new hierarchical formulation for locally parameterizing the butterfly subdivision surface over its initial control polyhedron, 2) formulation of dynamic butterfly subdivision surface as a set of novel finite elements, and 3) approximation of this new type of finite elements by a collection of existing finite elements subject to implicit geometric constraints. Our new physics-based model can be sculpted directly by applying synthesized forces and its equilibrium is characterized by the minimum of a deformation energy subject to the imposed constraints. We demonstrate that this novel dynamic framework not only provides a direct and natural means of manipulating geometric shapes, but also facilitates hierarchical shape and nonrigid motion estimation from large range and volumetric data sets using very few degrees of freedom (control vertices that define the initial polyhedron).
IEEE Trans. Vis. Comput. Graph.	A BRDF Postprocess to Integrate Porosity on Rendered Surfaces.	Stéphane Mérillou,Jean-Michel Dischler,Djamchid Ghazanfarpour	2000	The behavior of light interacting with materials is a crucial factor in achieving a high degree of realism in image synthesis. Local illumination processes, describing the interactions between a point of the surface and a shading ray, are evaluated by Bidirectional Reflectance Distribution Functions (BRDFs). Current theoretical BRDFs use surface models restricted to roughness only, sometimes at different scales. In this paper, we present a more complete surface micro-geometry description, suitable for some common surface defects, including porosity and micro-cracks; both of them are crucial surface features since they strongly influence light reflection properties. These new features are modeled by holes inserted in the surface profile, depending on two parameters: the proportion of surface covered by the defects and the mean geometric characteristic of these defects. In order to preserve the advantages and characteristics of existing BRDFs, a postprocessing method is adopted (we integrate our technique into existing models, instead of defining a completely new one). Beyond providing graphical results closely matching real behaviors, this method moreover opens the way to various important new considerations in computer graphics (for example, changes of appearance due to the degree of humidity).
IEEE Trans. Vis. Comput. Graph.	Perception-Based Fast Rendering and Antialiasing of Walkthrough Sequences.	Karol Myszkowski,Przemyslaw Rokita,Takehiro Tawara	2000	In this paper, we consider accelerated rendering of high quality walkthrough animation sequences along predefined paths. To improve rendering performance, we use a combination of a hybrid ray tracing and Image-Based Rendering (IBR) technique and a novel perception-based antialiasing technique. In our rendering solution, we derive as many pixels as possible using inexpensive IBR techniques without affecting the animation quality. A perception-based spatiotemporal Animation Quality Metric (AQM) is used to automatically guide such a hybrid rendering. The Image Flow (IF) obtained as a byproduct of the IBR computation is an integral part of the AQM. The final animation quality is enhanced by an efficient spatiotemporal antialiasing which utilizes the IF to perform a motion-compensated filtering. The filter parameters have been tuned using the AQM predictions of animation quality as perceived by the human observer. These parameters adapt locally to the visual pattern velocity.
IEEE Trans. Vis. Comput. Graph.	Spectral Volume Rendering.	Herke Jan Noordmans,Hans T. M. van der Voort,Arnold W. M. Smeulders	2000	Volume renderers for interactive analysis must be sufficiently versatile to render a broad range of volume images: unsegmented ¿raw¿ images as recorded by a 3D scanner, labeled segmented images, multimodality images, or any combination of these. The usual strategy is to assign to each voxel a three component RGB color and an opacity value $\alpha$. This so-called RGB$\alpha$ approach offers the possibility of distinguishing volume objects by color. However, these colors are connected to the objects themselves, thereby bypassing the idea that in reality the color of an object is also determined by the light source and light detectors c.q. human eyes. The physically realistic approach presented here models light interacting with the materials inside a voxel causing spectral changes in the light. The radiated spectrum falls upon a set of RGB detectors. The spectral approach is investigated to see whether it could enhance the visualization of volume data and interactive tools. For that purpose, a material is split into an absorbing part (the medium) and a scattering part (small particles). The medium is considered to be either achromatic or chromatic, while the particles are considered to scatter the light achromatically, elastically, or inelastically. It appears that inelastic scattering particles combined with an achromatic absorbing medium offer additional visual features: Objects are made visible through the surface structure of a surrounding volume object and volume and surface structures can be made visible at the same time. With one or two materials the method is faster than the RGB$\alpha$ approach, with three materials the performance is equal. The spectral approach can be considered as an extension of the RGB$\alpha$ approach with a greater visual flexibility and a better balance between quality and speed.
IEEE Trans. Vis. Comput. Graph.	Compressed Progressive Meshes.	Renato Pajarola,Jarek Rossignac	2000	Most systems that support visual interaction with 3D models use shape representations based on triangle meshes. The size of these representations imposes limits on applications for which complex 3D models must be accessed remotely. Techniques for simplifying and compressing 3D models reduce the transmission time. Multiresolution formats provide quick access to a crude model and then refine it progressively. Unfortunately, compared to the best nonprogressive compression methods, previously proposed progressive refinement techniques impose a significant overhead when the full resolution model must be downloaded. The CPM (Compressed Progressive Meshes) approach proposed here eliminates this overhead. It uses a new technique, which refines the topology of the mesh in batches, which each increase the number of vertices by up to 50 percent.Less than an amortized total of 4 bits per triangle encode where and how the topological refinements should be applied. We estimate the position of new vertices from the positions of their topological neighbors in the less refined mesh using a new estimator that leads to representations of vertex coordinates that are 50 percent more compact than previously reported progressive geometry compression techniques.
IEEE Trans. Vis. Comput. Graph.	"Corrections to 'Compressed Progressive Meshes'."	Renato Pajarola,Jarek Rossignac	2000	"Corrections to 'Compressed Progressive Meshes'."
IEEE Trans. Vis. Comput. Graph.	Tissue Classification Based on 3D Local Intensity Structures for Volume Rendering.	Yoshinobu Sato,Carl-Fredrik Westin,Abhir Bhalerao,Shin Nakajima,Nobuyuki Shiraga,Shinichi Tamura,Ron Kikinis	2000	This paper describes a novel approach to tissue classification using three-dimensional (3D) derivative features in the volume rendering pipeline. In conventional tissue classification for a scalar volume, tissues of interest are characterized by an opacity transfer function defined as a one-dimensional (1D) function of the original volume intensity. To overcome the limitations inherent in conventional 1D opacity functions, we propose a tissue classification method that employs a multidimensional opacity function, which is a function of the 3D derivative features calculated from a scalar volume as well as the volume intensity. Tissues of interest are characterized by explicitly defined classification rules based on 3D filter responses highlighting local structures, such as edge, sheet, line, and blob, which typically correspond to tissue boundaries, cortices, vessels, and nodules, respectively, in medical volume data. The 3D local structure filters are formulated using the gradient vector and Hessian matrix of the volume intensity function combined with isotropic Gaussian blurring. These filter responses and the original intensity define a multidimensional feature space in which multichannel tissue classification strategies are designed. The usefulness of the proposed method is demonstrated by comparisons with conventional single-channel classification using both synthesized data and clinical data acquired with CT (computed tomography) and MRI (magnetic resonance imaging) scanners. The improvement in image quality obtained using multichannel classification is confirmed by evaluating the contrast and contrast-to-noise ratio in the resultant volume-rendered images with variable opacity values.
IEEE Trans. Vis. Comput. Graph.	Calibration-Free Augmented Reality in Perspective.	Yongduek Seo,Ki-Sang Hong	2000	This paper deals with video-based augmented reality and proposes an algorithm for augmenting a real video sequence with views of graphics objects without metric calibration of the video camera by representing the motion of the video camera in projective space. A virtual camera, by which views of graphics objects are generated, is attached to a real camera by specifying image locations of the world coordinate system of the virtual world. The virtual camera is decomposed into calibration and motion components in order to make full use of graphics tools. The projective motion of the real camera recovered from image matches has the function of transferring the virtual camera and makes the virtual camera move according to the motion of the real camera. The virtual camera also follows the change of the internal parameters of the real camera. This paper shows the theoretical and experimental results of our application of nonmetric vision to augmented reality.
IEEE Trans. Vis. Comput. Graph.	Accessibility Analysis Using Computer Graphics Hardware.	Steven N. Spitz,Aristides A. G. Requicha	2000	"Analyzing the accessibility of an object's surface to probes or tools is important for many planning and programming tasks that involve spatial reasoning and arise in robotics and automation. This paper presents novel and efficient algorithms for computing accessible directions for tactile probes used in 3D digitization with Coordinate Measuring Machines. The algorithms are executed in standard computer graphics hardware. They are a nonobvious application of rendering hardware to scientific and technological areas beyond computer graphics."
IEEE Trans. Vis. Comput. Graph.	Fast Ray-Tracing of Rectilinear Volume Data Using Distance Transforms.	Milos Srámek,Arie E. Kaufman	2000	This paper discusses and experimentally compares distance-based acceleration algorithms for ray-tracing of volumetric data with an emphasis on the Chessboard Distance (CD) voxel traversal. The acceleration of this class of algorithms is achieved by skipping empty macro regions, which are defined for each background voxel of the volume. Background voxels are labeled in a preprocessing phase by a value, defining the macro region size, which is equal to the voxel distance to the nearest foreground voxel. The CD algorithm exploits the chessboard distance and defines the ray as a nonuniform sequence of samples positioned at voxel faces. This feature assures that no foreground voxels are missed during the scene traversal. Further, due to parallelepipedal shape of the macro region, it supports accelerated visualization of cubic, regular, and rectilinear grids. The CD algorithm is suitable for all modifications of the ray tracing/ray casting techniques being used in volume visualization and volume graphics. However, when used for rendering based on local surface interpolation, it also enables fast search of intersections between rays and the interpolated surface, further improving speed of the process.
IEEE Trans. Vis. Comput. Graph.	Accelerated Isosurface Extraction in Time-Varying Fields.	Philip M. Sutton,Charles D. Hansen	2000	For large time-varying data sets, memory and disk limitations can lower the performace of visualization applications. Algorithms and data structures must be explicitly designed to handle these data sets in order to achieve more interactive rates. The Temporal Branch-on-Need Octree (T-BON) extends the three-dimensional branch-on-need octree for time-varying isosurface extraction. This data structure minimizes the impact of the I/O bottleneck by reading from disk only those portions of the search structure and data necessary to construct the current isosurface. By performing a minimum of I/O and exploiting the hierarchical memory found in modern CPUs, the T-BON algorithm achieves high performance isosurface extraction in time-varying fields. This paper extends earlier work on the T-BON data structure by including techniques for better memory utilization, out-of-core isosurface extraction, and support for nonrectilinear grids. Results from testing the T-BON algorithm on large data sets show that its performance is similar to that of the three-dimensional branch-on-need octree for static data sets while providing substantial advantages for time-varying fields.
CVPR	2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2001), with CD-ROM, 8-14 December 2001, Kauai, HI, USA		2001	2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2001), with CD-ROM, 8-14 December 2001, Kauai, HI, USA
ICCV	High Dynamic Range Panoramic Imaging.	Manoj Aggarwal,Narendra Ahuja	2001	High Dynamic Range Panoramic Imaging.
ICCV	A New Imaging Model.	Manoj Aggarwal,Narendra Ahuja	2001	A New Imaging Model.
ICCV	Illumination Insensitive Eigenspaces.	Horst Bischof,Horst Wildenauer,Ales Leonardis	2001	Illumination Insensitive Eigenspaces.
ICCV	Split Aperture Imaging for High Dynamic Range.	Manoj Aggarwal,Narendra Ahuja	2001	Most imaging sensors have limited dynamic range and hence are sensitive to only a part of the illumination range present in a natural scene. The dynamic range can be improved by acquiring multiple images of the same scene under different exposure settings and then combining them. In this paper, we describe a camera design for simultaneously acquiring multiple images. The cross-section of the incoming beam from a scene point is partitioned into as many parts as the required number of images. This is done by splitting the aperture into multiple parts and directing the beam exiting from each in a different direction using an assembly of mirrors. A sensor is placed in the path of each beam and exposure of each sensor is controlled either by appropriately setting its exposure parameter, or by splitting the incoming beam unevenly. The resulting multiple exposure images are used to construct a high dynamic range image. We have implemented a video-rate camera based on this design and the results obtained are presented.
ICCV	On Cosine-Fourth and Vignetting Effects in Real Lenses.	Manoj Aggarwal,Hong Hua,Narendra Ahuja	2001	On Cosine-Fourth and Vignetting Effects in Real Lenses.
ICCV	Using Scene Constraints during the Calibration Procedure.	Didier Bondyfalat,Théodore Papadopoulo,Bernard Mourrain	2001	Using Scene Constraints during the Calibration Procedure.
ICCV	Accurate Catadioptric Calibration for Real-time Pose Estimation of Room-size Environments.	Daniel G. Aliaga	2001	Accurate Catadioptric Calibration for Real-time Pose Estimation of Room-size Environments.
ICCV	Matching Shapes.	Serge Belongie,Jitendra Malik,Jan Puzicha	2001	Matching Shapes.
ICCV	Video Phase-Locked Loops in Gait Recognition.	Jeffrey E. Boyd	2001	Video Phase-Locked Loops in Gait Recognition.
ICCV	Real-Time Video Phase-Locked Loops.	Jeffrey E. Boyd,Maxwell Sayles	2001	Real-Time Video Phase-Locked Loops.
ICCV	Interactive Graph Cuts for Optimal Boundary and Region Segmentation of Objects in N-D Images.	Yuri Boykov,Marie-Pierre Jolly	2001	Interactive Graph Cuts for Optimal Boundary and Region Segmentation of Objects in N-D Images.
ICCV	Demonstration of Segmentation with Interactive Graph Cuts.	Yuri Boykov,Marie-Pierre Jolly	2001	Demonstration of Segmentation with Interactive Graph Cuts.
ICCV	A Characterization of Inherent Stereo Ambiguities.	Simon Baker,Terence Sim,Takeo Kanade	2001	A Characterization of Inherent Stereo Ambiguities.
ICCV	Confidence and Curvature Estimation of Curvilinear Structures in 3-D.	Peter Bakker,Lucas J. van Vliet,Piet W. Verbeek	2001	Confidence and Curvature Estimation of Curvilinear Structures in 3-D.
ICCV	A Variational Model for Filling-In Gray Level and Color Images.	Coloma Ballester,Vicent Caselles,Joan Verdera,Marcelo Bertalmío,Guillermo Sapiro	2001	A Variational Model for Filling-In Gray Level and Color Images.
ICCV	Optimal Method for the Affine F-Matrix and Its Uncertainty Estimation in the Sense of both Noise and Outliers.	Sami S. Brandt,Jukka Heikkonen	2001	Optimal Method for the Affine F-Matrix and Its Uncertainty Estimation in the Sense of both Noise and Outliers.
ICCV	Incorporating Process Knowledge into Object Recognition for Assemblies.	Elke Braun,Jannik Fritsch,Gerhard Sagerer	2001	Incorporating Process Knowledge into Object Recognition for Assemblies.
ICCV	A Probabilistic Framework for Space Carving.	Adrian Broadhurst,Tom Drummond,Roberto Cipolla	2001	A Probabilistic Framework for Space Carving.
ICCV	Learning the Semantics of Words and Pictures.	Kobus Barnard,David A. Forsyth	2001	Learning the Semantics of Words and Pictures.
ICCV	What Value Covariance Information in Estimating Vision Parameters?	Michael J. Brooks,Wojciech Chojnacki,Darren Gawley,Anton van den Hengel	2001	What Value Covariance Information in Estimating Vision Parameters?
ICCV	Colour Photometric Stereo: Simultaneous Reconstruction of Local Gradient and Colour of Rough Textured Surfaces.	Svetlana Barsky,Maria Petrou	2001	Colour Photometric Stereo: Simultaneous Reconstruction of Local Gradient and Colour of Rough Textured Surfaces.
ICCV	Document Restoration Using 3D Shape: A General Deskewing Algorithm for Arbitrarily Warped Documents.	Michael S. Brown,W. Brent Seales	2001	Document Restoration Using 3D Shape: A General Deskewing Algorithm for Arbitrarily Warped Documents.
ICCV	Projective Structure and Motion from Two Views of a Piecewise Planar Scene.	Adrien Bartoli,Peter F. Sturm,Radu Horaud	2001	Projective Structure and Motion from Two Views of a Piecewise Planar Scene.
ICCV	Lambertian Reflectance and Linear Subspaces.	Ronen Basri,David W. Jacobs	2001	We prove that the set of all reflectance functions (the mapping from surface normals to intensities) produced by Lambertian objects under distant, isotropic lighting lies close to a 9D linear subspace. This implies that, in general, the set of images of a convex Lambertian object obtained under a wide variety of lighting conditions can be approximated accurately by a low-dimensional linear subspace, explaining prior empirical results. We also provide a simple analytic characterization of this linear space. We obtain these results by representing lighting using spherical harmonics and describing the effects of Lambertian materials as the analog of a convolution. These results allow us to construct algorithms for object recognition based on linear methods as well as algorithms that use convex optimization to enforce non-negative lighting functions. Finally, we show a simple way to enforce non-negative lighting when the images of an object lie near a 4D linear space.
ICCV	Multi-View Scene Capture by Surfel Sampling: From Video Streams to Non-Rigid 3D MotionShape & Reflectance.	Rodrigo L. Carceroni,Kiriakos N. Kutulakos	2001	Multi-View Scene Capture by Surfel Sampling: From Video Streams to Non-Rigid 3D MotionShape & Reflectance.
ICCV	Learning Local Evidence for Shading and Reflectance.	Matt Bell,William T. Freeman	2001	Learning Local Evidence for Shading and Reflectance.
ICCV	Feature Selection from Huge Feature Sets.	José Bins,Bruce A. Draper	2001	Feature Selection from Huge Feature Sets.
ICCV	Learning Spectral Calibration Parameters For Color Inspection.	Paulo D. Carvalho,Amâncio Santos,António Dourado,Bernardete Ribeiro	2001	Learning Spectral Calibration Parameters For Color Inspection.
ICCV	Affine Invariant Erosion of 3D Shapes.	Santiago Betelu,Guillermo Sapiro,Allen Tannenbaum	2001	Affine Invariant Erosion of 3D Shapes.
ICCV	Alignment of Non-Overlapping Sequences.	Yaron Caspi,Michal Irani	2001	Alignment of Non-Overlapping Sequences.
ICCV	Depth From Defocus in Presence of Partial Self Occlusion.	Sundeep Singh Bhasin,Subhasis Chaudhuri	2001	Depth From Defocus in Presence of Partial Self Occlusion.
ICCV	Fast Algorithm for Nearest Neighbor Search Based on a Lower Bound Tree.	Yong-Sheng Chen,Yi-Ping Hung,Chiou-Shann Fuh	2001	Fast Algorithm for Nearest Neighbor Search Based on a Lower Bound Tree.
ICCV	Trust-Region Methods for Real-Time Tracking.	Hwann-Tzong Chen,Tyng-Luh Liu	2001	Trust-Region Methods for Real-Time Tracking.
ICCV	Example-Based Facial Sketch Generation with Non-parametric Sampling.	Hong Chen,Ying-Qing Xu,Heung-Yeung Shum,Song Chun Zhu,Nanning Zheng	2001	Example-Based Facial Sketch Generation with Non-parametric Sampling.
ICCV	Sparse PCA: Extracting Multi-scale Structure from Data.	Chakra Chennubhotla,Allan D. Jepson	2001	Sparse PCA: Extracting Multi-scale Structure from Data.
ICCV	People Tracking Using Hybrid Monte Carlo Filtering.	Kiam Choo,David J. Fleet	2001	People Tracking Using Hybrid Monte Carlo Filtering.
ICCV	The Variable Bandwidth Mean Shift and Data-Driven Scale Selection.	Dorin Comaniciu,Visvanathan Ramesh,Peter Meer	2001	The Variable Bandwidth Mean Shift and Data-Driven Scale Selection.
ICCV	Constrained Active Appearance Models.	Timothy F. Cootes,Christopher J. Taylor	2001	Constrained Active Appearance Models.
ICCV	3D Object Recognition Using Shape Similarity-Based Aspect Graph.	Christopher M. Cyr,Benjamin B. Kimia	2001	3D Object Recognition Using Shape Similarity-Based Aspect Graph.
ICCV	BRDF/BTF Measurement Device.	Kristin J. Dana	2001	BRDF/BTF Measurement Device.
ICCV	Plan-View Trajectory Estimation with Dense Stereo Background Models.	Trevor Darrell,David Demirdjian,Neal Checka,Pedro F. Felzenszwalb	2001	Plan-View Trajectory Estimation with Dense Stereo Background Models.
ICCV	Automatic Segmentation and Indexing in a Database of Bird Images.	Madirakshi Das,R. Manmatha	2001	Automatic Segmentation and Indexing in a Database of Bird Images.
ICCV	Markov Face Models.	Sarat C. Dass,Anil K. Jain	2001	Markov Face Models.
ICCV	Motion Estimation from Disparity Images.	David Demirdjian,Trevor Darrell	2001	Motion Estimation from Disparity Images.
ICCV	Combining Single View Recognition and Multiple View Stereo for Architectural Scenes.	Anthony R. Dick,Philip H. S. Torr,Simon J. Ruffle,Roberto Cipolla	2001	Combining Single View Recognition and Multiple View Stereo for Architectural Scenes.
ICCV	Reconstructing Surfaces Using Anisotropic Basis Functions.	Huong Quynh Dinh,Greg Turk,Gregory G. Slabaugh	2001	Reconstructing Surfaces Using Anisotropic Basis Functions.
ICCV	Self-Calibration of a Stereo Rig Using Monocular Epipolar Geometry.	Fadi Dornaika	2001	This paper addresses the problem of self-calibration from one unknown motion of an uncalibrated stereo rig. Unlike the existing methods for stereo rig self-calibration, which have been focused on applying the autocalibration paradigm using both motion and stereo correspondences, our method does not require the recovery of stereo correspondences. Our method combines purely algebraic constraints with implicit geometric constraints. Assuming that the rotational part of the stereo geometry has two unknown degrees of freedom (i.e., the third dof is roughly known), and that the principle point of each camera is known, we first show that the computation of the intrinsic and extrinsic parameters of the stereo rig can be recovered from the motion correspondences only, i.e., the monocular fundamental matrices. We then provide an initialization procedure for the proposed non-linear method. We provide an extensive performance study for the method in the presence of image noise. In addition, we study some of the aspects related to the 3D motion that govern the accuracy of the proposed self-calibration method. Experiments conducted on synthetic and real data/images demonstrate the effectiveness and efficiency of the proposed method.
ICCV	Unambigous Determination of Shape from Photometric Stereo with Unknown Light Sources.	Ondrej Drbohlav,Radim Sára	2001	Unambigous Determination of Shape from Photometric Stereo with Unknown Light Sources.
ICCV	Real-Time Tracking of Highly Articulated Structures in the Presence of Noisy Measurements.	Tom Drummond,Roberto Cipolla	2001	Real-Time Tracking of Highly Articulated Structures in the Presence of Noisy Measurements.
ICCV	A Novel Modeling Algorithm for Shape Recovery of Unknown Topology.	Ye Duan,Hong Qin	2001	A Novel Modeling Algorithm for Shape Recovery of Unknown Topology.
ICCV	Biologically MotivatedPrecise and Simple Calibration and Reconstruction Using a Stereo Light Microscope.	Lars Eckert,Rolf-Rainer Grigat	2001	Biologically MotivatedPrecise and Simple Calibration and Reconstruction Using a Stereo Light Microscope.
ICCV	Probabilistic Framework for Segmenting People Under Occlusion.	Ahmed M. Elgammal,Larry S. Davis	2001	Probabilistic Framework for Segmenting People Under Occlusion.
ICCV	Blind Removal of Image Non-Linearities.	Hany Farid,Alin C. Popescu	2001	Blind Removal of Image Non-Linearities.
ICCV	Very High Accuracy Velocity Estimation using Orientation Tensors Parametric Motion and Simultaneous Segmentation of the Motion Field.	Gunnar Farnebäck	2001	Very High Accuracy Velocity Estimation using Orientation Tensors Parametric Motion and Simultaneous Segmentation of the Motion Field.
ICCV	Real-time Virtual Object Insertion.	Paolo Favaro,Hailin Jin,Stefano Soatto	2001	Real-time Virtual Object Insertion.
ICCV	A Curve Evolution Approach for Image Segmentation Using Adaptive Flows.	Haihua Feng,David A. Castañon,William Clement Karl	2001	A Curve Evolution Approach for Image Segmentation Using Adaptive Flows.
ICCV	4-Sensor Camera Calibration for Image Representation Invariant to ShadingShadowsLightingand Specularities.	Graham D. Finlayson,Mark S. Drew	2001	4-Sensor Camera Calibration for Image Representation Invariant to ShadingShadowsLightingand Specularities.
ICCV	Stochastic Rigidity: Image Registration for Nowhere-Static Scenes.	Andrew W. Fitzgibbon	2001	Stochastic Rigidity: Image Registration for Nowhere-Static Scenes.
ICCV	Shape from Texture and Integrability.	David A. Forsyth	2001	Shape from Texture and Integrability.
ICCV	Robust Histogram Construction from Color Invariants.	Theo Gevers	2001	Robust Histogram Construction from Color Invariants.
ICCV	A Data-Driven Model for Monocular Face Tracking.	Salih Burak Göktürk,Jean-Yves Bouguet,Radek Grzeszczuk	2001	A Data-Driven Model for Monocular Face Tracking.
ICCV	A General Imaging Model and a Method for Finding its Parameters.	Michael D. Grossberg,Shree K. Nayar	2001	A General Imaging Model and a Method for Finding its Parameters.
ICCV	Pairwise Face Recognition.	Guodong Guo,HongJiang Zhang,Stan Z. Li	2001	Pairwise Face Recognition.
ICCV	Visual Learning by Integrating Descriptive and Generative Methods.	Cheng-en Guo,Song Chun Zhu,Ying Nian Wu	2001	Visual Learning by Integrating Descriptive and Generative Methods.
ICCV	A Background Model Initialization Algorithm for Video Surveillance.	Daniel Gutchess,Miroslav Trajkovic,Eric Cohen-Solal,Damian Lyons,Anil K. Jain	2001	A Background Model Initialization Algorithm for Video Surveillance.
ICCV	Noise in Bilinear Problems.	John A. Haddon,David A. Forsyth	2001	Noise in Bilinear Problems.
ICCV	Stripe Boundary Codes for Real-Time Structured-Light Range Scanning of Moving Objects.	Olaf A. Hall-Holt,Szymon Rusinkiewicz	2001	Stripe Boundary Codes for Real-Time Structured-Light Range Scanning of Moving Objects.
ICCV	Multiple Motion Scene Reconstruction from Uncalibrated Views.	Mei Han,Takeo Kanade	2001	Multiple Motion Scene Reconstruction from Uncalibrated Views.
ICCV	Face Recognition with Support Vector Machines: Global versus Component-based Approach.	Bernd Heisele,Purdy Ho,Tomaso Poggio	2001	Face Recognition with Support Vector Machines: Global versus Component-based Approach.
ICCV	"Finding ``Anomalies'' in an Arbitrary Image."	Toshifumi Honda,Shree K. Nayar	2001	"Finding ``Anomalies'' in an Arbitrary Image."
ICCV	Multi-Agent Event Recognition.	Somboon Hongeng,Ramakant Nevatia	2001	Multi-Agent Event Recognition.
ICCV	Harmonics Extraction Based on Higher Order Statistics Spectrum Decomposition for A Unified Texture Model.	Yong Huang,Kap Luk Chan,ZhongYang Huang	2001	Harmonics Extraction Based on Higher Order Statistics Spectrum Decomposition for A Unified Texture Model.
ICCV	Geometrical Fundamentals of Polycentric Panoramas.	Fay Huang,Shou Kang Wei,Reinhard Klette	2001	Geometrical Fundamentals of Polycentric Panoramas.
ICCV	Folds and Cuts: How Shading Flows Into Edges.	Patrick S. Huggins,Steven W. Zucker	2001	Folds and Cuts: How Shading Flows Into Edges.
ICCV	Separation of Multiple Objects in Motion Images by Clustering.	Kohei Inoue,Kiichi Urahama	2001	Separation of Multiple Objects in Motion Images by Clustering.
ICCV	Human Tracking with Mixtures of Trees.	Sergey Ioffe,David A. Forsyth	2001	Human Tracking with Mixtures of Trees.
ICCV	BraMBLe: A Bayesian Multiple-Blob Tracker.	Michael Isard,John MacCormick	2001	BraMBLe: A Bayesian Multiple-Blob Tracker.
ICCV	Region Extraction from Multiple Images.	Hiroshi Ishikawa,Ian Jermyn	2001	Region Extraction from Multiple Images.
ICCV	A Pseudo-Distance Map for the Segmentation-Free Skeletonization of Gray-Scale Images.	Jeong-Hun Jang,Ki-Sang Hong	2001	A Pseudo-Distance Map for the Segmentation-Free Skeletonization of Gray-Scale Images.
ICCV	A Framework for Segmentation of Talk and Game Shows.	Omar Javed,Zeeshan Rasheed,Mubarak Shah	2001	A Framework for Segmentation of Talk and Game Shows.
ICCV	Video Objects Segmentation Using Eulerian Region-Based Active Contours.	Stéphanie Jehan-Besson,Michel Barlaud,Gilles Aubert	2001	Video Objects Segmentation Using Eulerian Region-Based Active Contours.
ICCV	Real-Time Feature Tracking and Outlier Rejection with Changes in Illumination.	Hailin Jin,Paolo Favaro,Stefano Soatto	2001	Real-Time Feature Tracking and Outlier Rejection with Changes in Illumination.
ICCV	Separating Appearance from Deformation.	Nebojsa Jojic,Patrice Simard,Brendan J. Frey,David Heckerman	2001	Separating Appearance from Deformation.
ICCV	Segmentation of the Left Ventricle in Cardiac MR Images.	Marie-Pierre Jolly,Nicolae Duta,Gareth Funka-Lea	2001	Segmentation of the Left Ventricle in Cardiac MR Images.
ICCV	A Robust Interest Points Matching Algorithm.	Il-Kyun Jung,Simon Lacroix	2001	A Robust Interest Points Matching Algorithm.
ICCV	A Simple and Efficient Template Matching Algorithm.	Frédéric Jurie,Michel Dhome	2001	A Simple and Efficient Template Matching Algorithm.
ICCV	Ambiguous Configurations for the 1D Structure and Motion Problem.	Fredrik Kahl,Kalle Åström	2001	Ambiguous Configurations for the 1D Structure and Motion Problem.
ICCV	Euclidean Reconstruction and Auto-Calibration from Continuous Motion.	Fredrik Kahl,Anders Heyden	2001	Euclidean Reconstruction and Auto-Calibration from Continuous Motion.
ICCV	Multiple View Geometry of Non-planar Algebraic Curves.	Jeremy Yermiyahou Kaminski,Michael Fryers,Amnon Shashua,Mina Teicher	2001	Multiple View Geometry of Non-planar Algebraic Curves.
ICCV	Motion Segmentation by Subspace Separation and Model Selection.	Ken-ichi Kanatani	2001	Motion Segmentation by Subspace Separation and Model Selection.
ICCV	Do We Really Have to Consider Covariance Matrices for Image Features?.	Yasushi Kanazawa,Ken-ichi Kanatani	2001	Do We Really Have to Consider Covariance Matrices for Image Features?.
ICCV	A Linear Dual-Space Approach to 3D Surface Reconstruction from Occluding Contours using Algebraic Surfaces.	Kongbin Kang,Jean-Philippe Tarel,Richard Fishman,David B. Cooper	2001	A Linear Dual-Space Approach to 3D Surface Reconstruction from Occluding Contours using Algebraic Surfaces.
ICCV	Plane-based Projective Reconstruction.	Robert Kaucic,Richard I. Hartley,Nicolas Y. Dano	2001	Plane-based Projective Reconstruction.
ICCV	Automated 3D PDM Construction Using Deformable Models.	Michael Kaus,Vladimir Pekar,Cristian Lorenz,Roel Truyen,Steven Lobregt,Jens A. Richolt,Jürgen Weese	2001	Automated 3D PDM Construction Using Deformable Models.
ICCV	Uncalibrated Motion Capture Exploiting Articulated Structure Constraints.	David Liebowitz,Stefan Carlsson	2001	We present an algorithm for 3D reconstruction of dynamic articulated structures, such as humans, from uncalibrated multiple views. The reconstruction exploits constraints associated with a dynamic articulated structure, specifically the conservation over time of length between rotational joints. These constraints admit reconstruction of metric structure from at least two different images in each of two uncalibrated parallel projection cameras. As a by product, the calibration of the cameras can also be computed. The algorithm is based on a stratified approach, starting with affine reconstruction from factorization, followed by rectification to metric structure using the articulated structure constraints. The exploitation of these specific constraints admits reconstruction and self-calibration with fewer feature points and views compared to standard self-calibration. The method is extended to pairs of cameras that are zooming, where calibration of the cameras allows compensation for the changing scale factor in a scaled orthographic camera. Results are presented in the form of stick figures and animated 3D reconstructions using pairs of sequences from broadcast television. The technique shows promise as a means of creating 3D animations of dynamic activities such as sports events.
ICCV	Human Tracking in Multiple Cameras.	Sohaib Khan,Omar Javed,Zeeshan Rasheed,Mubarak Shah	2001	Human Tracking in Multiple Cameras.
ICCV	Efficient Dense Depth Estimation from Dense Multiperspective Panoramas.	Yin Li,Chi-Keung Tang,Heung-Yeung Shum	2001	Efficient Dense Depth Estimation from Dense Multiperspective Panoramas.
ICCV	True Single View Point Cone Mirror Omni-Directional Catadioptric System.	Shih-Schön Lin,Ruzena Bajcsy	2001	True Single View Point Cone Mirror Omni-Directional Catadioptric System.
ICCV	Improving AR using Shadows Arising from Natural Illumination Distribution in Video Sequences.	Taeone Kim,Yongduek Seo,Ki-Sang Hong	2001	Improving AR using Shadows Arising from Natural Illumination Distribution in Video Sequences.
ICCV	Region Segmentation via Deformable Model-Guided Split and Merge.	Lifeng Liu,Stan Sclaroff	2001	"Abstract An improved method for deformable shape-based image segmentation is described. Image regions are merged together and/or split apart, based on their agreement with an a priori distribution on the global deformation parameters for a shape template. The quality of a candidate region merging is evaluated by a cost measure that includes: homogeneity of image properties within the combined region, degree of overlap with a deformed shape model, and a deformation likelihood term. Perceptually-motivated criteria are used to determine where/how to split regions, based on the local shape properties of the region group''s bounding contour. A globally consistent interpretation is determined in part by the minimum description length principle. Experiments show that the model-based splitting strategy yields a significant improvement in segmention over a method that uses merging alone."
ICCV	A Gabor Feature Classifier for Face Recognition.	Chengjun Liu,Harry Wechsler	2001	A Gabor Feature Classifier for Face Recognition.
ICCV	Computing Visual Correspondence with Occlusions via Graph Cuts.	Vladimir Kolmogorov,Ramin Zabih	2001	Several new algorithms for visual correspondence based on graph cuts have recently been developed. While these methods give very strong results in practice, they do not handle occlusions properly. Specifically, they treat the two input images asymmetrically, and they do not ensure that a pixel corresponds to at most one pixel in the other image. In this paper, we present two new methods which properly address occlusions, while preserving the advantages of graph cut algorithms. We give experimental results for stereo as well as motion, which demonstrate that our methods perform well both at detecting occlusions and computing disparities.
ICCV	Learning Inhomogeneous Gibbs Model of Faces by Minimax Entropy.	Ce Liu,Song Chun Zhu,Heung-Yeung Shum	2001	Learning Inhomogeneous Gibbs Model of Faces by Minimax Entropy.
ICCV	Learning Low Dimensional Invariant Signature of 3-D Object under Varying View and Illumination from 2-D Appearances.	Stan Z. Li,Jie Yan,XinWen Hou,ZeYu Li,HongJiang Zhang	2001	Learning Low Dimensional Invariant Signature of 3-D Object under Varying View and Illumination from 2-D Appearances.
ICCV	Hierarchical Pre-Segmentation without Prior Knowledge.	Arjan Kuijper,Luc Florack	2001	Hierarchical Pre-Segmentation without Prior Knowledge.
ICCV	A Co-inference Approach to Robust Visual Tracking.	Ying Wu,Thomas S. Huang	2001	A Co-inference Approach to Robust Visual Tracking.
ICCV	New Perspectives on Geometric Reflection Theory From Rough Surfaces.	Andrew Lundberg,Lawrence B. Wolff,Diego A. Socolinsky	2001	New Perspectives on Geometric Reflection Theory From Rough Surfaces.
ICCV	Dimensional Analysis of Image Motion.	Michael S. Langer,Richard Mann	2001	Dimensional Analysis of Image Motion.
ICCV	Beyond Lambert: Reconstructing Surfaces with Arbitrary BRDFs.	Sebastian Magda,David J. Kriegman,Todd Zickler,Peter N. Belhumeur	2001	Beyond Lambert: Reconstructing Surfaces with Arbitrary BRDFs.
ICCV	Visual Servoing Invariant to Changes in Camera Intrinsic Parameters.	Ezio Malis	2001	Visual Servoing Invariant to Changes in Camera Intrinsic Parameters.
ICCV	Incorporating Differential Constraints in a 3D Reconstruction Process. Application to Stereo.	Richard Lengagne,Pascal Fua	2001	Incorporating Differential Constraints in a 3D Reconstruction Process. Application to Stereo.
ICCV	Affine Calibration from Moving Objects.	Russell A. Manning,Charles R. Dyer	2001	Affine Calibration from Moving Objects.
ICCV	Model-Based Initialisation of Vehicle Tracking: Dependency on Illumination.	Holger Leuck,Hans-Hellmut Nagel	2001	Model-Based Initialisation of Vehicle Tracking: Dependency on Illumination.
ICCV	A Database of Human Segmented Natural Images and its Application to Evaluating Segmentation Algorithms and Measuring Ecological Statistics.	David R. Martin,Charless Fowlkes,Doron Tal,Jitendra Malik	2001	A Database of Human Segmented Natural Images and its Application to Evaluating Segmentation Algorithms and Measuring Ecological Statistics.
ICCV	"The Earth Mover's Distance is the Mallows Distance: Some Insights from Statistics."	Elizaveta Levina,Peter J. Bickel	2001	"The Earth Mover's Distance is the Mallows Distance: Some Insights from Statistics."
ICCV	Efficient Sequential Karhunen-Loeve Basis Extraction.	Avraham Levy,Michael Lindenbaum	2001	Efficient Sequential Karhunen-Loeve Basis Extraction.
ICCV	A Versatile Method for Trifocal Tensor Estimation.	Bogdan Matei,Bogdan Georgescu,Peter Meer	2001	A Versatile Method for Trifocal Tensor Estimation.
ICCV	Kernel Machine Based Learning for Multi-View Face Detection and Pose Estimation.	Stan Z. Li,QingDong Fu,Lie Gu,Bernhard Schölkopf,Yimin Cheng,HongJiang Zhang	2001	Kernel Machine Based Learning for Multi-View Face Detection and Pose Estimation.
ICCV	Modelling Faces Dynamically across Views and Over Time.	Yongmin Li,Shaogang Gong,Heather M. Liddell	2001	Modelling Faces Dynamically across Views and Over Time.
ICCV	View-Based Clustering of Object Appearances Based on Independent Subspace Analysis.	Stan Z. Li,XiaoGuang Lv,HongJiang Zhang	2001	View-Based Clustering of Object Appearances Based on Independent Subspace Analysis.
ICCV	FaceTracker: A Human Face Tracking and Facial Organ Localizing System.	Jun Miao,Hongming Zhang,Wen Gao,Gang Deng,Hong Liu,Xilin Chen	2001	FaceTracker: A Human Face Tracking and Facial Organ Localizing System.
ICCV	Estimation and Interpretation of Discontinuities in Optical Flow Fields.	Markus Middendorf,Hans-Hellmut Nagel	2001	Estimation and Interpretation of Discontinuities in Optical Flow Fields.
ICCV	Indexing Based on Scale Invariant Interest Points.	Krystian Mikolajczyk,Cordelia Schmid	2001	Indexing Based on Scale Invariant Interest Points.
ICCV	Color Eigenflows: Statistical Modeling of Joint Color Changes.	Erik G. Miller,Kinh Tieu	2001	Color Eigenflows: Statistical Modeling of Joint Color Changes.
ICCV	Image Detection Under Varying Illumination and Pose.	Margarita Osadchy,Daniel Keren	2001	Image Detection Under Varying Illumination and Pose.
ICCV	Gradient Vector Flow Fast Geodesic Active Contours.	Nikos Paragios,Olivier Mellina-Gottardo,Visvanathan Ramesh	2001	Abstract--In this paper, we propose an edge-driven bidirectional geometric flow for boundary extraction. To this end, we combine the geodesic active contour flow [CHECK END OF SENTENCE] and the gradient vector flow external force for snakes [CHECK END OF SENTENCE]. The resulting motion equation is considered within a level set formulation [CHECK END OF SENTENCE], can deal with topological changes and important shape deformations. An efficient numerical schema is used for the flow implementation that exhibits robust behavior and has fast convergence rate [CHECK END OF SENTENCE], [CHECK END OF SENTENCE]. Promising results on real and synthetic images demonstrate the potentials of the flow.
ICCV	A Demonstration of Handprinted Symbol Recognition.	Jim R. Parker,Juraj Pivovarov	2001	A Demonstration of Handprinted Symbol Recognition.
ICCV	3D Object Tracking Using Shape-Encoded Particle Propagation.	Hankyu Moon,Rama Chellappa,Azriel Rosenfeld	2001	3D Object Tracking Using Shape-Encoded Particle Propagation.
ICCV	Classifying and Solving Minimal Structure and Motion Problems with Missing Data.	Magnus Oskarsson,Kalle Åström,Niels Chr. Overgaard	2001	Classifying and Solving Minimal Structure and Motion Problems with Missing Data.
ICCV	JetStream: Probabilistic Contour Extraction with Particles.	Patrick Pérez,Andrew Blake,Michel Gangnet	2001	JetStream: Probabilistic Contour Extraction with Particles.
ICCV	Photometric Image-Based Rendering for Image Generation in Arbitrary Illumination.	Yasuhiro Mukaigawa,Hajime Miyaki,Sadahiko Mihashi,Takeshi Shakunaga	2001	Photometric Image-Based Rendering for Image Generation in Arbitrary Illumination.
ICCV	Performance Evaluation of Stereo for Tele-presence.	Jane Mulligan,Volkan Isler,Konstantinos Daniilidis	2001	Performance Evaluation of Stereo for Tele-presence.
ICCV	Articulated Soft Objects for Video-based Body Modeling.	Ralf Plänkers,Pascal Fua	2001	Articulated Soft Objects for Video-based Body Modeling.
ICCV	Do Ambiguous Reconstructions Always Give Ambiguous Images?.	Marc Pollefeys,Luc J. Van Gool	2001	Do Ambiguous Reconstructions Always Give Ambiguous Images?.
ICCV	Towards Real-Time Multi-Modality 3-D Medical Image Registration.	Thomas Netsch,Peter Rösch,Arianne van Muiswinkel,Jürgen Weese	2001	Towards Real-Time Multi-Modality 3-D Medical Image Registration.
ICCV	Occlusion Robust Adaptive Template Tracking.	Hieu Tat Nguyen,Marcel Worring,Rein van den Boomgaard	2001	Occlusion Robust Adaptive Template Tracking.
ICCV	Structure From Motion Using Sequential Monte Carlo Methods.	Gang Qian,Rama Chellappa	2001	In this paper, the structure from motion (SfM) problem is addressed using sequential Monte Carlo methods. A new SfM algorithm based on random sampling is derived to estimate the posterior distributions of camera motion and scene structure for the perspective projection camera model. Experimental results show that challenging issues in solving the SfM problem, due to erroneous feature tracking, feature occlusion, motion/structure ambiguity, mixed-domain sequences, mismatched features, and independently moving objects, can be well modeled and effectively addressed using the proposed method.
ICCV	Determining Reflectance Parameters and Illumination Distribution from a Sparse Set of Images for View-dependent Image Synthesis.	Ko Nishino,Zhengyou Zhang,Katsushi Ikeuchi	2001	Determining Reflectance Parameters and Illumination Distribution from a Sparse Set of Images for View-dependent Image Synthesis.
ICCV	Calibration with Robust Use of Cheirality by Quasi-Affine Reconstruction of the Set of Camera Projection Centres.	David Nistér	2001	Calibration with Robust Use of Cheirality by Quasi-Affine Reconstruction of the Set of Camera Projection Centres.
ICCV	Concentric Mosaic(s)Planar Motion and 1D Cameras.	Long Quan,Le Lu,Heung-Yeung Shum,Maxime Lhuillier	2001	Concentric Mosaic(s)Planar Motion and 1D Cameras.
ICCV	Reducing Drift in Parametric Motion Tracking.	Ali Rahimi,Louis-Philippe Morency,Trevor Darrell	2001	Reducing Drift in Parametric Motion Tracking.
ICCV	A Statistical Approach to Background Subtraction for Surveillance Systems.	Naoya Ohta	2001	A Statistical Approach to Background Subtraction for Surveillance Systems.
ICCV	Simultaneous Estimation of Super-Resolved Intensity and Depth Maps from Low Resolution Defocused Observations of a Scene.	Deepu Rajan,Subhasis Chaudhuri	2001	Simultaneous Estimation of Super-Resolved Intensity and Depth Maps from Low Resolution Defocused Observations of a Scene.
ICCV	An EM Algorithm for Video SummarizationGenerative Model Approach.	Xavier Orriols,Xavier Binefa	2001	An EM Algorithm for Video SummarizationGenerative Model Approach.
ICCV	Model-Based Bundle Adjustment with Application to Face Modeling.	Ying Shan,Zicheng Liu,Zhengyou Zhang	2001	Model-Based Bundle Adjustment with Application to Face Modeling.
ICCV	3D-Mode: A 3D Modelling and Measurement System Using a Few Photos.	Gang Xu,Takahisa Nakayama,Takeshi Kajikawa,Jun-ichi Terai	2001	3D-Mode: A 3D Modelling and Measurement System Using a Few Photos.
ICCV	A Maximum Likelihood Framework for Iterative Eigendecomposition.	Antonio Robles-Kelly,Edwin R. Hancock	2001	A Maximum Likelihood Framework for Iterative Eigendecomposition.
ICCV	Multi-Frame Infinitesimal Motion Model for the Reconstruction of (Dynamic) Scenes with Multiple Linearly Moving Objects.	Amnon Shashua,Anat Levin	2001	Multi-Frame Infinitesimal Motion Model for the Reconstruction of (Dynamic) Scenes with Multiple Linearly Moving Objects.
ICCV	Computationally Efficient Face Detection.	Sami Romdhani,Philip H. S. Torr,Bernhard Schölkopf,Andrew Blake	2001	Computationally Efficient Face Detection.
ICCV	3D Hand Pose Reconstruction Using Specialized Mappings.	Rómer Rosales,Vassilis Athitsos,Leonid Sigal,Stan Sclaroff	2001	3D Hand Pose Reconstruction Using Specialized Mappings.
ICCV	Color Constancy Using KL-Divergence.	Charles R. Rosenberg,Martial Hebert,Sebastian Thrun	2001	Color Constancy Using KL-Divergence.
ICCV	Continuous Global Evidence-Based Bayesian Modality Fusion for Simultaneous Tracking of Multiple Objects.	Jamie Sherrah,Shaogang Gong	2001	Continuous Global Evidence-Based Bayesian Modality Fusion for Simultaneous Tracking of Multiple Objects.
ICCV	Precise Sub-Pixel Estimation on Area-Based Matching.	Masao Shimizu,Masatoshi Okutomi	2001	Precise Sub-Pixel Estimation on Area-Based Matching.
ICCV	Linear Multi View Reconstruction and Camera Recovery.	Carsten Rother,Stefan Carlsson	2001	Linear Multi View Reconstruction and Camera Recovery.
ICCV	Recognizing Large 3-D Objects through Next View Planning using an Uncalibrated Camera.	Sumantra Dutta Roy,Santanu Chaudhury,Subhashis Banerjee	2001	Recognizing Large 3-D Objects through Next View Planning using an Uncalibrated Camera.
ICCV	Learning Image Statistics for Bayesian Tracking.	Hedvig Sidenbladh,Michael J. Black	2001	Learning Image Statistics for Bayesian Tracking.
ICCV	Physics-based Model Acquisition and Identification in Airborne Spectral Images.	David Slater,Glenn Healey	2001	Physics-based Model Acquisition and Identification in Airborne Spectral Images.
ICCV	Shadow Carving.	Silvio Savarese,Holly E. Rushmeier,Fausto Bernardini,Pietro Perona	2001	Shadow Carving.
ICCV	Viewpoint Invariant Texture Matching and Wide Baseline Stereo.	Frederik Schaffalitzky,Andrew Zisserman	2001	Viewpoint Invariant Texture Matching and Wide Baseline Stereo.
ICCV	Generalized Mosaicing.	Yoav Y. Schechner,Shree K. Nayar	2001	Generalized Mosaicing.
ICCV	Dynamic Textures.	Stefano Soatto,Gianfranco Doretto,Ying Nian Wu	2001	Dynamic Textures.
ICCV	Stochastic Processes in Vision: From Langevin to Beltrami.	Nir A. Sochen	2001	Stochastic Processes in Vision: From Langevin to Beltrami.
ICCV	Template Matching Approach to Content Based Image Indexing by Low Dimensional Euclidean Embedding.	Haim Schweitzer	2001	Template Matching Approach to Content Based Image Indexing by Low Dimensional Euclidean Embedding.
ICCV	Stochastic Road Shape Estimation.	Ben Southall,Camillo J. Taylor	2001	Stochastic Road Shape Estimation.
ICCV	Accurate Optical Flow in Noisy Image Sequences.	Hagen Spies,Hanno Scharr	2001	Accurate Optical Flow in Noisy Image Sequences.
ICCV	Recognition of Shapes by Editing Shock Graphs.	Thomas B. Sebastian,Philip N. Klein,Benjamin B. Kimia	2001	Recognition of Shapes by Editing Shock Graphs.
ICCV	The Space of All Stereo Images.	Steven M. Seitz	2001	The Space of All Stereo Images.
ICCV	Automatic Registration of 2-D with 3-D Imagery in Urban Environments.	Ioannis Stamos,Peter K. Allen	2001	Automatic Registration of 2-D with 3-D Imagery in Urban Environments.
ICCV	Propagation of Innovative Information in Non-Linear Least-Squares Structure from Motion.	Drew Steedly,Irfan A. Essa	2001	Propagation of Innovative Information in Non-Linear Least-Squares Structure from Motion.
ICCV	Topology Free Hidden Markov Models: Application to Background Modeling.	Bjoern Stenger,Visvanathan Ramesh,Nikos Paragios,Frans Coetzee,Joachim M. Buhmann	2001	Topology Free Hidden Markov Models: Application to Background Modeling.
ICCV	Segmentation and Range Sensing Using a Moving-Aperture Lens.	Anbumani Subramanian,Lakshmi R. Iyer,A. Lynn Abbott,Amy E. Bell	2001	The use of a novel motorized lens to perform segmentation of image sequences is presented in this paper. The lens has the effect of introducing small, repeating movements of the camera center so that objects appear to translate in the image by an amount that depends on the distance from the plane of focus. For a stationary scene, optical flow magnitudes are therefore directly related to three-dimensional object distance from the observer. We describe a segmentation procedure that exploits these controlled observer movements and present experimental results that demonstrate the successful extraction of objects at different depths. Potential applications of our approach include image compositing, teleconferencing, and range estimation.
ICCV	Invariant Mixture Recognition in Hyperspectral Images.	Pei-hsiu Suen,Glenn Healey	2001	Invariant Mixture Recognition in Hyperspectral Images.
ICCV	The Impact of Viewing Geometry on Vision Through the Atmosphere.	Pei-hsiu Suen,Glenn Healey,David Slater	2001	The Impact of Viewing Geometry on Vision Through the Atmosphere.
ICCV	Self-Calibrating Camera Projector Systems for Interactive Displays and Presentations.	Rahul Sukthankar,Tat-Jen Cham,Gita Sukthankar,James M. Rehg,David Hsu,Thomas Leung	2001	Self-Calibrating Camera Projector Systems for Interactive Displays and Presentations.
ICCV	Smarter Presentations: Exploiting Homography in Camera-Projector Systems.	Rahul Sukthankar,Robert G. Stockton,Matthew D. Mullin	2001	Smarter Presentations: Exploiting Homography in Camera-Projector Systems.
ICCV	Guiding Random Particles by Deterministic Search.	Josephine Sullivan,Jens Rittscher	2001	Guiding Random Particles by Deterministic Search.
ICCV	"Surface Matching by 3D Point's Fingerprint."	Yiyong Sun,Mongi A. Abidi	2001	"Surface Matching by 3D Point's Fingerprint."
ICCV	Caustics of Catadioptric Cameras.	Rahul Swaminathan,Michael D. Grossberg,Shree K. Nayar	2001	Caustics of Catadioptric Cameras.
ICCV	Tele-Graffiti: A Pen and Paper-Based Remote Sketching System.	Naoya Takao,Jianbo Shi,Simon Baker,Iain Matthews,Bart C. Nabbe	2001	Tele-Graffiti: A Pen and Paper-Based Remote Sketching System.
ICCV	Selecting Objects With Freehand Sketches.	Kar-Han Tan,Narendra Ahuja	2001	Selecting Objects With Freehand Sketches.
ICCV	A Global Matching Framework for Stereo Computation.	Hai Tao,Harpreet S. Sawhney,Rakesh Kumar	2001	A Global Matching Framework for Stereo Computation.
ICCV	Probabilistic Learning and Modelling of Object Dynamics for Tracking.	Terence Tay,Kah Kay Sung	2001	Probabilistic Learning and Modelling of Object Dynamics for Tracking.
ICCV	Statistical Context Priming for Object Detection.	Antonio Torralba,Pawan Sinha	2001	Statistical Context Priming for Object Detection.
ICCV	Robust Principal Component Analysis for Computer Vision.	Fernando De la Torre,Michael J. Black	2001	Robust Principal Component Analysis for Computer Vision.
ICCV	Probabilistic Tracking in a Metric Space.	Kentaro Toyama,Andrew Blake	2001	Probabilistic Tracking in a Metric Space.
ICCV	Joint Feature Distributions for Image Correspondence.	Bill Triggs	2001	Joint Feature Distributions for Image Correspondence.
ICCV	Empirical Filter Estimation for Subpixel Interpolation and Matching.	Bill Triggs	2001	Empirical Filter Estimation for Subpixel Interpolation and Matching.
ICCV	AMILab: A 2D/3D Image Processing Software.	Agustin Trujillo,Karl Krissian	2001	AMILab: A 2D/3D Image Processing Software.
ICCV	Statistical Calibration of the CCD Imaging Process.	Yanghai Tsin,Visvanathan Ramesh,Takeo Kanade	2001	Statistical Calibration of the CCD Imaging Process.
ICCV	Image Segmentation by Data Driven Markov Chain Monte Carlo.	Zhuowen Tu,Song Chun Zhu,Heung-Yeung Shum	2001	Image Segmentation by Data Driven Markov Chain Monte Carlo.
ICCV	On the Complexity of Probabilistic Image Retrieval.	Nuno Vasconcelos	2001	On the Complexity of Probabilistic Image Retrieval.
ICCV	Flux Maximizing Geometric Flows.	Alexander Vasilevskiy,Kaleem Siddiqi	2001	Several geometric active contour models have been proposed for segmentation in computer vision and image analysis. The essential idea is to evolve a curve (in 2D) or a surface (in 3D) under constraints from image forces so that it clings to features of interest in an intensity image. Recent variations on this theme take into account properties of enclosed regions and allow for multiple curves or surfaces to be simultaneously represented. However, it is still unclear how to apply these techniques to images of narrow elongated structures, such as blood vessels, where intensity contrast may be low and reliable region statistics cannot be computed. To address this problem, we derive the gradient flows which maximize the rate of increase of flux of an appropriate vector field through a curve (in 2D) or a surface (in 3D). The key idea is to exploit the direction of the vector field along with its magnitude. The calculations lead to a simple and elegant interpretation which is essentially parameter free and has the same form in both dimensions. We illustrate its advantages with several level-set-based segmentations of 2D and 3D angiography images of blood vessels.
ICCV	Stereo Matching by Compact Windows via Minimum Ratio Cycle.	Olga Veksler	2001	Stereo Matching by Compact Windows via Minimum Ratio Cycle.
ICCV	Sequential Monte Carlo Fusion of Sound and Vision for Speaker Tracking.	Jaco Vermaak,Michel Gangnet,Andrew Blake,Patrick Pérez	2001	Sequential Monte Carlo Fusion of Sound and Vision for Speaker Tracking.
ICCV	Optimal Motion Estimation from Multiview Normalized Epipolar Constraint.	René Vidal,Yi Ma,Shawn Hsu,Shankar Sastry	2001	Optimal Motion Estimation from Multiview Normalized Epipolar Constraint.
ICCV	Robust Real-Time Face Detection.	Paul A. Viola,Michael J. Jones	2001	This paper describes a face detection framework that is capable of processing images extremely rapidly while achieving high detection rates. There are three key contributions. The first is the introduction of a new image representation called the &ldquo;Integral Image&rdquo; which allows the features used by our detector to be computed very quickly. The second is a simple and efficient classifier which is built using the AdaBoost learning algorithm (Freund and Schapire, 1995) to select a small number of critical visual features from a very large set of potential features. The third contribution is a method for combining classifiers in a &ldquo;cascade&rdquo; which allows background regions of the image to be quickly discarded while spending more computation on promising face-like regions. A set of experiments in the domain of face detection is presented. The system yields face detection performance comparable to the best previous systems (Sung and Poggio, 1998; Rowley et al., 1998; Schneiderman and Kanade, 2000; Roth et al., 2000). Implemented on a conventional desktop, face detection proceeds at 15 frames per second.
ICCV	Error Analysis of Pure Rotation-Based Self-Calibration.	Leslie Wang,Sing Bing Kang,Heung-Yeung Shum,Guangyou Xu	2001	Abstract--Self-calibration using pure rotation is a well-known technique and has been shown to be a reliable means for recovering intrinsic camera parameters. However, in practice, it is virtually impossible to ensure that the camera motion for this type of self-calibration is a pure rotation. In this paper, we present an error analysis of recovered intrinsic camera parameters due to the presence of translation. We derived closed-form error expressions for a single pair of images with nondegenerate motion; for multiple rotations for which there are no closed-form solutions, analysis was done through repeated experiments. Among others, we show that translation-independent solutions do exist under certain practical conditions. Our analysis can be used to help choose the least error-prone approach (if multiple approaches exist) for a given set of conditions.
ICCV	Image Segmentation with Minimum Mean Cut.	Song Wang,Jeffrey Mark Siskind	2001	Image Segmentation with Minimum Mean Cut.
ICCV	Shape Deformation: SVM Regression and Application to Medical Image Segmentation.	Song Wang,Weiyu Zhu,Zhi-Pei Liang	2001	Shape Deformation: SVM Regression and Application to Medical Image Segmentation.
ICCV	Control of Home Appliances Using Face and Hand Sign Recognition.	Hiroki Watanabe,Hitoshi Hongo,Mamoru Yasumoto,Yoshinori Niwa,Kazuhiko Yamamoto	2001	Control of Home Appliances Using Face and Hand Sign Recognition.
ICCV	Deriving Intrinsic Images from Image Sequences.	Yair Weiss	2001	Deriving Intrinsic Images from Image Sequences.
ICCV	Cheirality in Epipolar Geometry.	Tomás Werner,Tomás Pajdla	2001	Cheirality in Epipolar Geometry.
ICCV	Camera Calibration and 3D Reconstruction from Single Images Using Parallelepipeds.	Marta Wilczkowiak,Edmond Boyer,Peter F. Sturm	2001	Camera Calibration and 3D Reconstruction from Single Images Using Parallelepipeds.
ICCV	Video Georegistration: Algorithm and Quantitative Evaluation.	Richard P. Wildes,David J. Hirvonen,Steven C. Hsu,Rakesh Kumar,W. Brian Lehman,Bogdan Matei,Wen-Yi Zhao	2001	Video Georegistration: Algorithm and Quantitative Evaluation.
ICCV	On Projection Matrices and their Applications in Computer Vision.	Lior Wolf,Amnon Shashua	2001	On Projection Matrices and their Applications in Computer Vision.
ICCV	Affine 3-D Reconstruction from Two Projective Images of Independently Translating Planes.	Lior Wolf,Amnon Shashua	2001	Affine 3-D Reconstruction from Two Projective Images of Independently Translating Planes.
ICCV	Structure and Motion from Silhouettes.	Kwan-Yee Kenneth Wong,Roberto Cipolla	2001	Structure and Motion from Silhouettes.
ICCV	Self-Supervised Learning for Object Recognition based on Kernel Discriminant-EM Algorithm.	Ying Wu,Thomas S. Huang,Kentaro Toyama	2001	Self-Supervised Learning for Object Recognition based on Kernel Discriminant-EM Algorithm.
ICCV	Capturing Natural Hand Articulation.	Ying Wu,John Y. Lin,Thomas S. Huang	2001	Capturing Natural Hand Articulation.
ICCV	3D LAMP: A New Layered Panoramic Representation.	Zhigang Zhu,Allen R. Hanson	2001	3D LAMP: A New Layered Panoramic Representation.
ICCV	Parallel-Perspective Stereo Mosaics.	Zhigang Zhu,Edward M. Riseman,Allen R. Hanson	2001	Parallel-Perspective Stereo Mosaics.
ICCV	Omni-Rig: Linear Self-Recalibration of a Rig with Varying Internal and External Parameters.	Assaf Zomet,Lior Wolf,Amnon Shashua	2001	Omni-Rig: Linear Self-Recalibration of a Rig with Varying Internal and External Parameters.
ICCV	Stereoscopic Segmentation.	Anthony J. Yezzi,Stefano Soatto	2001	We cast the problem of multiframe stereo reconstruction of a smooth surface as the global region segmentation of a collection of images of the scene. Dually, the problem of segmenting multiple calibrated images of an object becomes that of estimating the solid shape that gives rise to such images. We assume that the radiance of the scene results in piecewise homogeneous image statistics. This simplifying assumption covers Lambertian scenes with constant albedo as well as fine homogeneous textures, which are known challenges to stereo algorithms based on local correspondence. We pose the segmentation problem within a variational framework, and use fast level set methods to find the optimal solution numerically. Our algorithm does not work in the presence of strong photometric features, where traditional reconstruction algorithms do. It enjoys significant robustness to noise under the assumptiong it is designed for.
ICCV	Feature Based Object Recognition using Statistical Occlusion Models with One-to-one Correspondence.	Zhengrong Ying,David A. Castañon	2001	Feature Based Object Recognition using Statistical Occlusion Models with One-to-one Correspondence.
ICCV	Segmentation with Pairwise Attraction and Repulsion.	Stella X. Yu,Jianbo Shi	2001	Segmentation with Pairwise Attraction and Repulsion.
ICCV	The KGBR Viewpoint-Lighting Ambiguity and its Resolution by Generic Constraints.	Alan L. Yuille,James M. Coughlan,Scott Konishi	2001	The KGBR Viewpoint-Lighting Ambiguity and its Resolution by Generic Constraints.
ICCV	Real-Time Automated Concurrent Visual Tracking of Many Animals and Subsequent Behavioral Compilation.	John S. Zelek,David J. Bullock	2001	Real-Time Automated Concurrent Visual Tracking of Many Animals and Subsequent Behavioral Compilation.
ICCV	Cloning Your Own Face with a Desktop Camera.	Zhengyou Zhang,Zicheng Liu,Dennis Adler,Michael F. Cohen,Erik Hanson,Ying Shan	2001	Cloning Your Own Face with a Desktop Camera.
ICCV	Car Detection in Low Resolution Aerial Image.	Tao Zhao,Ramakant Nevatia	2001	Car Detection in Low Resolution Aerial Image.
IEEE Visualization	Point Set Surfaces.	Marc Alexa,Johannes Behr,Daniel Cohen-Or,Shachar Fleishman,David Levin,Cláudio T. Silva	2001	We advocate the use of point sets to represent shapes. We provide a definition of a smooth manifold surface from a set of points close to the original surface. The definition is based on local maps from differential geometry, which are approximated by the method of moving least squares (MLS). We present tools to increase or decrease the density of the points, thus, allowing an adjustment of the spacing among the points to control the fidelity of the representation.To display the point set surface, we introduce a novel point rendering technique. The idea is to evaluate the local maps according to the image resolution. This results in high quality shading effects and smooth silhouettes at interactive frame rates.
IEEE Visualization	Virtual Temporal Bone Dissection: A Case Study.	Jason Bryan,Don Stredney,Gregory J. Wiet,Dennis Sessanna	2001	The Temporal Bone Dissection Simulator is an ongoing research project for the construction of a synthetic environment suitable for virtual dissection of human temporal bone and related anatomy. Funded by the National Institute on Deafness and Other Communication Disorders (NIDCD), the primary goal of this project is to provide a safe, robust, and cost-effective virtual environment for learning the anatomy and surgical procedures associated with the temporal bone. Direct volume visualization has been indispensable for the necessary level of realism and interactivity that is vital to the success of this project. This work is being conducted by the Ohio Supercomputer Center in conjunction with the Department of Otolaryngology at the Ohio State University, and NIDCD.
IEEE Visualization	Transport and Anisotropic Diffusion in Time-Dependent Flow Visualization.	David Bürkle,Tobias Preußer,Martin Rumpf	2001	The visualization of time-dependent flow is an important and challenging topic in scientific visualization. Its aim is to represent transport phenomena governed by time-dependent vector fields in an intuitively understandable way, using images and animations. Here we pick up the recently presented anisotropic diffusion method, expand and generalize it to allow a multiscale visualization of longtime, complex transport problems. Instead of streamline type patterns generated by the original method now streakline patterns are generated and advected. This process obeys a nonlinear transport diffusion equation with typically dominant transport. Starting from some noisy initial image, the diffusion actually generates and enhances patterns which are then transported in the direction of the flow field. Simultaneously the image is again sharpened in the direction orthogonal to the flow field. A careful adjustment of the model&sacute; parameters is derived to balance diffusion and transport effects in a reasonable way. Properties of the method can be discussed for the continuous model, which is solved by an efficient upwind finite element discretization. As characteristic for the class of multiscale image processing methods, we can in advance select a suitable scale for representing the flow field.
IEEE Visualization	Nonlinear Virtual Colon Unfolding.	Anna Vilanova Bartrolí,Rainer Wegenkittl,Andreas König,Eduard Gröller	2001	The majority of virtual endoscopy techniques tries to simulate a real endoscopy. A real endoscopy does not always give the optimal information due to the physical limitations it is subject to. In this paper, we deal with the unfolding of the surface of the colon as a possible visualization technique for diagnosis and polyp detection. A new two-step technique is presented which deals with the problems of double appearance of polyps and nonuniform sampling that other colon unfolding techniques suffer from. In the first step, a distance map from a central path induces nonlinear rays for unambiguous parameterization of the surface. The second step compensates for locally varying distortions of the unfolded surface. A technique similar to magnification fields in information visualization is hereby applied. The technique produces a single view of a complete, virtually dissected colon.
IEEE Visualization	Simplicial Subdivisions and Sampling Artifacts.	Hamish Carr,Torsten Möller,Jack Snoeyink	2001	We review several schemes for dividing cubical cells into simplices (tetrahedra) in 3-D for interpolating from sampled data to IR3 or for computing isosurfaces by barycentric interpolation. We present test data that reveal the geometric artifacts that these subdivision schemes generate, and discuss how these artifacts relate to the filter kernels that correspond to the subdivision schemes.
IEEE Visualization	Wavelet Representation of Contour Sets.	Martin Bertram,Daniel E. Laney,Mark A. Duchaineau,Charles D. Hansen,Bernd Hamann,Kenneth I. Joy	2001	We present a new wavelet compression and multiresolution modeling approach for sets of contours (level sets). In contrast to previous wavelet schemes, our algorithm creates a parametrization of a scalar field induced by its contours and compactly stores this parametrization rather than function values sampled on a regular grid. Our representation is based on hierarchical polygon meshes with subdivision connectivity whose vertices are transformed into wavelet coefficients. From this sparse set of coefficients, every set of contours can be efficiently reconstructed at multiple levels of resolution. When applying lossy compression, introducing high quantization errors, our method preserves contour topology, in contrast to compression methods applied to the corresponding field function. We provide numerical results for scalar fields defined on planar domains. Our approach generalizes to volumetric domains, time-varying contours, and level sets of vector fields.
IEEE Visualization	POP: A Hybrid Point and Polygon Rendering System for Large Data.	Baoquan Chen,Minh Xuan Nguye	2001	We introduce a simple but effective extension to the existing pure point rendering systems. Rather than using only points, we use both points and polygons to represent and render large mesh models. We start from triangles as leaf nodes and build up a hierarchical tree structure with intermediate nodes as points. During the rendering, the system determines whether to use a point (of a certain intermediate level node) or a triangle (of a leaf node) for display depending on the screen contribution of each node. While points are used to speedup the rendering of distant objects, triangles are used to ensure the quality of close objects. Our method can accelerate the rendering of large models, compromising little in image quality.
IEEE Visualization	Hybrid Simplification: Combining Multi-Resolution Polygon and Point Rendering.	Jonathan D. Cohen,Daniel G. Aliaga,Weiqiand Zhang	2001	Multi-resolution hierarchies of polygons and more recently of points are familiar and useful tools for achieving interactive rendering rates. We present an algorithm for tightly integrating the two into a single hierarchical data structure. The trade-off between rendering portions of a model with points or with polygons is made automatically.Our approach to this problem is to apply a bottom-up simplification process involving not only polygon simplification operations, but point replacement and point simplification operations as well. Given one or more surface meshes, our algorithm produces a hybrid hierarchy comprising both polygon and point primitives. This hierarchy may be optimized according to the relative performance characteristics of these primitive types on the intended rendering platform. We also provide a range of aggressiveness for performing point replacement operations. The most conservative approach produces a hierarchy that is better than a purely polygonal hierarchy in some places, and roughly equal in others. A less conservative approach can trade reduced complexity at the far viewing ranges for some increased complexity at the near viewing ranges.We demonstrate our approach on a number of input models, achieving primitive counts that are 1.3 to 4.7 times smaller than those of triangle-only simplification.
IEEE Visualization	Integrating Occlusion Culling with View-Dependent Rendering.	Jihad El-Sana,Neta Sokolovsky,Cláudio T. Silva	2001	We present a novel approach that integrates occlusion culling within the view-dependent rendering framework. View-dependent rendering provides the ability to change level of detail over the surface seamlessly and smoothly in real-time. The exclusive use of view-parameters to perform level-of-detail selection causes even occluded regions to be rendered in high level of detail. To overcome this serious drawback we have integrated occlusion culling into the level selection mechanism. Because computing exact visibility is expensive and it is currently not possible to perform this computation in real time, we use a visibility estimation technique instead. Our approach reduces dramatically the resolution at occluded regions.
IEEE Visualization	Case Study: Visual Debugging Of Cluster Hardware.	Patricia Crossno,Rena Haynes	2001	This paper presents a novel use of visualization applied to debugging the Cplant&trade; cluster hardware at Sandia National Laboratories. As commodity cluster systems grow in popularity and grow in size, tracking component failures within the hardware will become more and more difficult. We have developed a tool that facilitates visual debugging of errors within the switches and cables connecting the processors. Combining an abstract system model with color-coding for both error and job information enables failing components to be identified.
IEEE Visualization	Variational Classification for Visualization of 3D Ultrasound Data.	Raanan Fattal,Dani Lischinski	2001	We present a new technique for visualizing surfaces from 3D ultrasound data. 3D ultrasound datasets are typically fuzzy, contain a substantial amount of noise and speckle, and suffer from several other problems that make extraction of continuous and smooth surfaces extremely difficult. We propose a novel opacity classification algorithm for 3D ultrasound datasets, based on the variational principle. More specifically, we compute a volumetric opacity function that optimally satisfies a set of simultaneous requirements. One requirement makes the function attain nonzero values only in the vicinity of a user-specified value, resulting in soft shells of finite, approximately constant thickness around isosurfaces in the volume. Other requirements are designed to make the function smoother and less sensitive to noise and speckle. The computed opacity function lends itself well to explicit geometric surface extraction, as well as to direct volume rendering at interactive rates. We also describe a new splatting algorithm that is particularly well suited for displaying soft opacity shells. Several examples and comparisons are included to illustrate our approach and demonstrate its effectiveness on real 3D ultrasound datasets.
IEEE Visualization	Undersampling and Oversampling in Sample Based Shape Modeling.	Tamal K. Dey,Joachim Giesen,Samrat Goswami,James Hudson,Rephael Wenger,Wulue Zhao	2001	Shape modeling is an integral part of many visualization problems. Recent advances in scanning technology and a number of surface reconstruction algorithms have opened up a new paradigm for modeling shapes from samples. Many of the problems currently faced in this modeling paradigm can be traced back to two anomalies in sampling, namely undersampling and oversampling. Boundaries, non-smoothness and small features create undersampling problems, whereas oversampling leads to too many triangles. We use Voronoi cell geometry as a unified guide to detect undersampling and oversampling. We apply these detections in surface reconstruction and model simplification. Guarantees of the algorithms can be proved. In this paper we show the success of the algorithms empirically on a number of interesting data sets.
IEEE Visualization	Archaeological Data Visualization in VR: Analysis of Lamp Finds at the Great Temple of Petra, a Case Study.	Daniel Acevedo Feliz,Eileen Vote,David H. Laidlaw,Martha Sharp Joukowsky	2001	We present the results of an evaluation of the ARCHAVE system, an immersive virtual reality environment for archaeological research. ARCHAVE is implemented in a Cave. The evaluation studied researchers analyzing lamp and coin finds throughout the excavation trenches at the Petra Great Temple site in Jordan. Experienced archaeologists used our system to study excavation data, confirming existing hypotheses and postulating new theories they had not been able to discover without the system. ARCHAVE provided access to the excavation database, and researchers were able to examine the data in the context of a life-size representation of the present day architectural ruins of the temple. They also had access to a miniature model for site-wide analysis. Because users quickly became comfortable with the interface, they concentrated their efforts on examining the data being retrieved and displayed. The immersive VR visualization of the recovered information gave them the opportunity to explore it in a new and dynamic way and, in several cases, enabled them to make discoveries that opened new lines of investigation about the excavation.
IEEE Visualization	Case Study: Reconstruction, Visualization, and Quantification of Neuronal Fiber Pathways.	Zhaohua Ding,John C. Gore,Adam W. Anderson	2001	It is of significant interest for neurological studies to determine and visualize neuronal fiber pathways in the human brain. By exploiting the capability of diffusion tensor magnetic resonance imaging to detect local orientations of neuronal fibers, we have developed a system of algorithms to reconstruct, visualize and quantify neuronal fiber pathways in vivo. Illustrative results show that the system is a promising tool for visual analysis of fiber connectivity and quantitative studies of neuronal fibers.
IEEE Visualization	Volume Rendering of Fine Details Within Medical Data.	Feng Dong,Gordon Clapworthy,Meleagros A. Krokos	2001	"This paper presents a method concerning the volume rendering of fine details, such as blood vessels and nerves, from medical data. The realistic and efficient visualization of such structures is often of great medical interest, and conventional rendering techniques do not always deal with them adequately. Our method uses preprocessing to reconstruct fine details that are difficult to segment and label. It detects the presence of fine geometrical structures, such as cracks or cylinders that suggest the existence of, for example, blood vessels or nerves; the subsequent volume rendering then displays fine geometrical objects that lie on a surface. The method can also show structures within the volume, using a special ""integration sampling"" scheme to portray reconstructed volume texture, such as that exhibited by muscle fibers. By combining the surface structure and volume texture in the rendering, realistic results can be produced; examples are provided."
IEEE Visualization	Fast Extraction of Adaptive Multiresolution Meshes with Guaranteed Properties from Volumetric Data.	Marcel Gavriliu,Joel Carranza,David E. Breen,Alan H. Barr	2001	"We present a new algorithm for extracting adaptive multiresolution triangle meshes from volume datasets. The algorithm guarantees that the topological genus of the generated mesh is the same as the genus of the surface embedded in the volume dataset at all levels of detail. In addition to this ""hard constraint"" on the genus of the mesh, the user can choose to specify some number of soft geometric constraints, such as triangle aspect ratio, minimum or maximum total number of vertices, minimum and/or maximum triangle edge lengths, maximum magnitude of various error metrics per triangle or vertex, including maximum curvature (area) error, maximum distance to the surface, and others. The mesh extraction process is fully automatic and does not require manual adjusting of parameters to produce the desired results as long as the user does not specify incompatible constraints. The algorithm robustly handles special topological cases, such as trimmed surfaces (intersections of the surface with the volume boundary), and manifolds with multiple disconnected components (several closed surfaces embedded in the same volume dataset). The meshes may self-intersect at coarse resolutions. However, the self-intersections are corrected automatically as the resolution of the meshes increase. We show several examples of meshes extracted from complex volume datasets."
IEEE Visualization	Case Study: An Environment for Understanding Protein Simulations Using Game Graphics.	Donna L. Gresh,Frank Suits,Yuk Yin Sham	2001	We describe a visualization system designed for interactive study of proteins in the field of computational biology. Our system incorporates multiple, custom, three-dimensional and two-dimensional linked views of the proteins. We take advantage of modern commodity graphics cards, which are typically designed for games rather than scientific visualization applications, to provide instantaneous linking between views and three-dimensional interactivity on standard personal computers. Furthermore, we anticipate the usefulness of game techniques such as bump maps and skinning for scientific applications.
IEEE Visualization	Normal Bounds for Subdivision-Surface Interference Detection.	Eitan Grinspun,Peter Schröder	2001	Subdivision surfaces are an attractive representation when modeling arbitrary topology free-form surfaces and show great promise for applications in engineering design [5, 6] and computer animation [10]. Interference detection is a critical tool in many of these applications. In this paper we derive normal bounds for subdivision surfaces and use these to develop an efficient algorithm for (self-) interference detection.
IEEE Visualization	PingTV: A Case Study in Visual Network Monitoring.	Alexander Gubin,William Yurcik,Larry Brumbaugh	2001	"PingTV generates a logical map of a network that is used as an overlay on a physical geographical image of the location from the user perspective (buildings, floors within buildings, etc.). PingTV is used at Illinois State University as a visualization tool to communicate real-time network conditions to the university community via a dedicated channel on the campus cable TV system. Colored symbols allow students and staff to discern high-congestion ""rush hours"" and understand why their specific Internet connectivity is ""broken"" from the wide range of potential causes. Lessons learned include the use of color to visually convey confidence intervals using color shading and the visualization of cyclical network traffic patterns. Our implementation is general and flexible with potential for application for other domains."
IEEE Visualization	Real-Time Decompression and Visualization of Animated Volume Data.	Stefan Guthe,Wolfgang Straßer	2001	Interactive exploration of animated volume data is required by many application, but the huge amount of computational time and storage space needed for rendering does not allow the visualization of animated volumes by now. In this paper we introduce an algorithm running at interactive frame rates using 3d wavelet transforms that allows for any wavelet, motion compensation techniques and various encoding schemes of the resulting wavelet coefficients to be used. We analyze different families and orders of wavelets for compression ratio and the introduced error. We use a quantization that has been optimized for the visual impression of the reconstructed volume independent of the viewing. This enables us to achieve very high compression ratios while still being able to reconstruct the volume with as few visual artifacts as possible. A further improvement of the compression ratio has been achieved by applying a motion compensation scheme to exploit temporal coherency. Using these scheme we are capable of decompressing each volume of our animation at interactive frame rates, while visualizing these decompressed volumes on a single PC. We also present a number of improved visualization algorithms for high quality display using OpenGL hardware running at interactive frame rates on a standard PC.
IEEE Visualization	Smooth Approximation and Rendering of Large Scattered Data Sets.	Jörg Haber,Frank Zeilfelder,Oleg Davydov,Hans-Peter Seidel	2001	We present an efficient method to automatically compute a smooth approximation of large functional scattered data sets given over arbitrarily shaped planar domains. Our approach is based on the construction of a C1-continuous bivariate cubic spline and our method offers optimal approximation order. Both local variation and nonuniform distribution of the data are taken into account by using local polynomial least squares approximations of varying degree. Since we only need to solve small linear systems and no triangulation of the scattered data points is required, the overall complexity of the algorithm is linear in the total number of points. Numerical examples dealing with several real world scattered data sets with up to millions of points demonstrate the efficiency of our method. The resulting spline surface is of high visual quality and can be efficiently evaluated for rendering and modeling. In our implementation we achieve real-time frame rates for typical fly-through sequences and interactive frame rates for recomputing and rendering a locally modified spline surface.
IEEE Visualization	Visualization and Interaction Techniques for the Exploration of Vascular Structures.	Horst K. Hahn,Bernhard Preim,Dirk Selle,Heinz-Otto Peitgen	2001	We describe a pipeline of image processing steps for deriving symbolic models of vascular structures from radiological data which reflect the branching pattern and diameter of vessels. For the visualization of these symbolic models, concatenated truncated cones are smoothly blended at branching points. We put emphasis on the quality of the visualizations which is achieved by anti-aliasing operations in different stages of the visualization. The methods presented are referred to as HQVV (High Quality Vessel Visualization). Scalable techniques are provided to explore vascular structures of different orders of magnitude. The hierarchy as well as the diameter of the branches of vascular systems are used to restrict visualizations to relevant subtrees and to emphasize parts of vascular systems.Our research is inspired by clear visualizations in textbooks and is targeted toward medical education and therapy planning. We describe the application of vessel visualization techniques for liver surgery planning. For this application it is crucial to recognize the morphology and branching pattern of vascular systems as well as the basic spatial relations between vessels and other anatomic structures.
IEEE Visualization	A Complete Distance Field Representation.	Jian Huang,Roger Crawfis,Shao-Chiung Lu,Shuh-Yuan Liou	2001	A Complete Distance Field Representation.
IEEE Visualization	Multiresolution Feature Extraction from Unstructured Meshes.	Andreas Hubeli,Markus H. Gross	2001	We present a framework to extract mesh features from unstructured two-manifold surfaces. Our method computes a collection of piecewise linear curves describing the salient features of surfaces, such as edges and ridge lines. We extend these basic techniques to a multiresolution setting which improves the quality of the results and accelerates the extraction process. The extraction process is semi-automatic, that is, the user is required to input a few control parameters and to select the operators to be applied to the input surface. Our mesh feature extraction algorithm can be used as a preprocessor for a variety of applications in geometric modeling including mesh fairing, subdivision and simplification.
IEEE Visualization	Compressing Large Polygonal Models.	Jeffrey Ho,Kuang-Chih Lee,David J. Kriegman	2001	"We present an algorithm that uses partitioning and gluing to compress large triangular meshes which are too complex to fit in main memory. The algorithm is based largely on the existing mesh compression algorithms, most of which require an 'in-core' representation of the input mesh. Our solution is to partition the mesh into smaller submeshes and compress these submeshes separately using existing mesh compression techniques. Since a direct partition of the input mesh is out of question, instead, we partition a simplified mesh and use the partition on the simplified model to obtain a partition on the original model. In order to recover the full connectivity, we present a simple scheme for encoding/decoding the resulting boundary structure from the mesh partition. When compressing large models with few singular vertices, a negligible portion of the compressed output is devoted to gluing information. On desktop computers, we have run experiments on models with millions of vertices, which could not be compressed using standard compression software packages, and have observed compression ratios as high as 17 to 1 using our technique."
IEEE Visualization	Case Study: Interacting with Cortical Flat Maps of the Human Brain.	Monica K. Hurdal,Kevin W. Kurtz,David C. Banks	2001	The complex geometry of the human brain contains many folds and fissures, making it impossible to view the entire surface at once. Since most of the cortical activity occurs on these folds, it is desirable to be able to view the entire surface of the brain in a single view. This can be achieved using quasi-conformal flat maps of the cortical surface. Computational and visualization tools are now needed to be able to interact with these flat maps of the brain to gain information about spatial and functional relationships that might not otherwise be apparent. Such information can contribute to earlier diagnostic tools for diseases and improved treatment. Our group is developing visualization and analysis tools that will help elucidate new information about the human brain through the interaction between a cortical surface and its corresponding quasi-conformal flat map.
IEEE Visualization	Connectivity Shapes.	Martin Isenburg,Stefan Gumhold,Craig Gotsman	2001	We describe a method to visualize the connectivity graph of a mesh using a natural embedding in 3D space. This uses a 3D shape representation that is based solely on mesh connectivity --- the connectivity shape. Given a connectivity, we define its natural geometry as a smooth embedding in space with uniform edge lengths and describe efficient techniques to compute it. Our main contribution is to demonstrate that a surprising amount of geometric information is implicit in the connectivity.We also show how to generate connectivity shapes that approximate given 3D shapes. Potential applications of connectivity shapes to modeling and mesh coding are described.
IEEE Visualization	Case Study: Medical Web Service for the Automatic 3D Documentation for Neuroradiological Diagnosis.	Sabine Iserhardt-Bauer,Peter Hastreiter,Thomas Ertl,K. Eberhardt,Bernd Tomandl	2001	This case study presents a medical web service for the automatic analysis of CTA (Computer Tomography Angiography) datasets. It aims at the detection and evaluation of intracranial aneurysms which are malformations of cerebral blood vessels. To obtain a standardized 3D visualization digital videos are automatically generated. The time-consuming video production caused by the manual delineation of structures, software based volume rendering, and the interactive definition of an optimized camera path is considerably improved with a fully automatic strategy. Therefore, a previously suggested approach [11] is applied which uses an optimized transfer function as a template and automatically adapts it to an individual dataset. Furthermore, we introduce hardware-accelerated morphologic filtering in order to detect the location of mid-size and giant aneurysms. The actual generation of the video is finally integrated into a hardware accelerated off-screen rendering process based on 3D texture mapping ensuring fast visualization of high quality. Overall, clinical routine can be considerably assisted by providing a web based service combining automatic detection and standardized visualization.
IEEE Visualization	Dynamic Shadow Removal from Front Projection Displays.	Christopher O. Jaynes,Stephen B. Webb,R. Matt Steele,Michael S. Brown,W. Brent Seales	2001	Front-projection display environments suffer from a fundamental problem: users and other objects in the environment can easily and inadvertently block projectors, creating shadows on the displayed image. We introduce a technique that detects and corrects transient shadows in a multi-projector display. Our approach is to minimize the difference between predicted (generated) and observed (camera) images by continuous modification of the projected image values for each display device. We are unaware of any other technique that directly addresses this problem. Furthermore, we speculate that the general predictive monitoring framework introduced here is capable of addressing more general radiometric consistency problems such as display-surface inter-reflections and the changes in display color and intensity due to projector bulb temperature variation.Using an automatically-derived relative position of cameras and projectors in the display environment and a straightforward color correction scheme, the system renders an expected image for each camera location. Cameras observe the displayed image, which is compared with the expected image to detect shadowed regions. These regions are transformed to the appropriate projector frames, where corresponding pixel values are increased. In display regions where more than one projector contributes to the image, shadow regions are eliminated. We demonstrate an implementation of the technique to remove shadows in a multi-projector front projection system.
IEEE Visualization	Lagrangian-Eulerian Advection for Unsteady Flow Visualization.	Bruno Jobard,Gordon Erlebacher,M. Yousuff Hussaini	2001	In this paper, we propose a new technique to visualize dense representations of time-dependent vector fields based on a Lagrangian-Eulerian Advection (LEA) scheme. The algorithm produces animations with high spatio-temporal correlation at interactive rates. With this technique, every still frame depicts the instantaneous structure of the flow, whereas an animated sequence of frames reveals the motion a dense collection of particles would take when released into the flow. The simplicity of both the resulting data structures and the implementation suggest that LEA could become a useful component of any scientific visualization toolkit concerned with the display of unsteady flows.
IEEE Visualization	Computed Tomography Angiography: A Case Study of Peripheral Vessel Investigation.	Armin Kanitsar,Rainer Wegenkittl,Petr Felkel,Dominik Fleischmann,Dominique Sandner,Eduard Gröller	2001	Computed Tomography Angiography: A Case Study of Peripheral Vessel Investigation.
IEEE Visualization	Visualizing 2D Probability Distributions from EOS Satellite Image-Derived Data Sets: A Case Study.	David T. Kao,Jennifer L. Dungan,Alex Pang	2001	"Maps of biophysical and geophysical variables using Earth Observing System (EOS) satellite image data are an important component of Earth science. These maps have a single value derived at every grid cell and standard techniques are used to visualize them. Current tools fall short, however, when it is necessary to describe a distribution of values at each grid cell. Distributions may represent a frequency of occurrence over time, frequency of occurrence from multiple runs of an ensemble forecast or possible values from an uncertainty model. We identify these ""distribution data sets"" and present a case study to visualize such 2D distributions. Distribution data sets are different from multivariate data sets in the sense that the values are for a single variable instead of multiple variables. Data for this case study consists of multiple realizations of percent forest cover, generated using a geostatistical technique that combines ground measurements and satellite imagery to model uncertainty about forest cover. We present two general approaches for analyzing and visualizing such data sets. The first is a pixel-wise analysis of the probability density functions for the 2D image while the second is an analysis of features identified within the image. Such pixel-wise and feature-wise views will give Earth scientists a more complete understanding of distribution data sets. See www.cse.ucsc.edu/research/avis/nasa_is for additional information."
IEEE Visualization	Quantitative Comparative Evaluation of 2D Vector Field Visualization Methods.	David H. Laidlaw,Robert M. Kirby,J. Scott Davidson,Timothy S. Miller,Marco da Silva,William H. Warren,Michael J. Tarr	2001	Quantitative Comparative Evaluation of 2D Vector Field Visualization Methods.
IEEE Visualization	Cell-Projection of Cyclic Meshes.	Martin Kraus,Thomas Ertl	2001	We present the first algorithm that employs hardware-accelerated cell-projection for direct volume rendering of cyclic meshes, i.e., meshes with visibility cycles. The visibility sorting of a cyclic mesh is performed by an extended topological sorting, which computes and isolates visibility cycles. Measured sorting times are comparable to previously published algorithms, which are, however, restricted to acyclic meshes. In practice, our algorithm is also useful for acyclic meshes as numerical instabilities can lead to false visibility cycles.Our method includes a simple, hardware-assisted algorithm based on image compositing that renders visibility cycles correctly. For tetrahedral meshes this algorithm allows us to render each tetrahedral cell (whether it is part of a cycle or not) by hardware-accelerated cell-projection. In its basic form our method applies only to convex cyclic meshes; however, we present an exact and a simpler but inexact extension of our method for nonconvex meshes.
IEEE Visualization	Chromatin Decondensation: A Case Study of Tracking Features in Confocal Dat.	Wim C. de Leeuw,Robert van Liere	2001	In this case study we discuss an interactive feature tracking system and its use for the analysis of chromatin decondensation. Features are described as points in a multidimensional attribute space. Distances between points are used as a measure for feature correspondence. Users can interactively experiment with the correspondence measure in order to gain insight in chromatin movement. In addition, by defining time as an attribute, tracking problems related to noisy confocal data can be circumvented.
IEEE Visualization	Interactive Volume Rendering Using Multi-Dimensional Transfer Functions and Direct Manipulation Widgets.	Joe Kniss,Gordon L. Kindlmann,Charles D. Hansen	2001	Most direct volume renderings produced today employ one-dimensional transfer functions, which assign color and opacity to the volume based solely on the single scalar quantity which comprises the dataset. Though they have not received widespread attention, multi-dimensional transfer functions are a very effective way to extract specific material boundaries and convey subtle surface properties. However, identifying good transfer functions is difficult enough in one dimension, let alone two or three dimensions. This paper demonstrates an important class of three-dimensional transfer functions for scalar data (based on data value, gradient magnitude, and a second directional derivative), and describes a set of direct manipulation widgets which make specifying such transfer functions intuitive and convenient. We also describe how to use modern graphics hardware to interactively render with multi-dimensional transfer functions. The transfer functions, widgets, and hardware combine to form a powerful system for interactive volume exploration.
IEEE Visualization	Case Study: Application of Feature Tracking to Analysis of Autoignition Simulation Data.	Wendy S. Koegler	2001	The focus of this paper is to evaluate the usefulness of some basic feature tracking algorithms as analysis tools for combustion datasets by application to a dataset modeling autoignition. Features defined as areas of high intermediate concentrations were examined to explore the initial phases in the autoigniton process.
IEEE Visualization	Circular Incident Edge Lists: A Data Structure for Rendering Complex Unstructured Grids.	Bruno Lévy,Guillaume Caumon,Stephane Conreaux,Xavier Cavin	2001	We present the Circular Incident Edge Lists (CIEL), a new data structure and a high-performance algorithm for generating a series of iso-surfaces in a highly unstructured grid. Slicing-based volume rendering is also considered. The CIEL data structure represents all the combinatorial information of the grid, making it possible to optimize the classical propagation from local minima paradigm. The usual geometric structures are replaced by a more efficient combinatorial structure. An active edges list is maintained, and iteratively propagated from an iso-surface to the next one in a very efficient way. The intersected cells incident to each active edge are retrieved, and the intersection polygons are generated by circulating around their facets. This latter feature enables arbitrary irregular cells to be treated, such as those encountered in certain computational fluid dynamics (CFD) simulations. Since the CIEL data structure solely depends on the connections between the cells, it is possible to take into account dynamic changes in the geometry of the mesh and in property values, which only requires the sorted extrema list to be updated. Experiments have shown that our approach is significantly faster than classical methods. The major drawback of our method is its memory consumption, higher than most classical methods. However, experimental results show that it stays within a practical range.
IEEE Visualization	Visualization of Large Terrains Made Easy.	Peter Lindstrom,Valerio Pascucci	2001	We present an elegant and simple to implement framework for performing out-of-core visualization and view-dependent refinement of large terrain surfaces. Contrary to the recent trend of increasingly elaborate algorithms for large-scale terrain visualization, our algorithms and data structures have been designed with the primary goal of simplicity and efficiency of implementation. Our approach to managing large terrain data also departs from more conventional strategies based on data tiling. Rather than emphasizing how to segment and efficiently bring data in and out of memory, we focus on the manner in which the data is laid out to achieve good memory coherency for data accesses made in a top-down (coarse-to-fine) refinement of the terrain. We present and compare the results of using several different data indexing schemes, and propose a simple to compute index that yields substantial improvements in locality and speed over more commonly used data layouts.Our second contribution is a new and simple, yet easy to generalize method for view-dependent refinement. Similar to several published methods in this area, we use longest edge bisection in a top-down traversal of the mesh hierarchy to produce a continuous surface with subdivision connectivity. In tandem with the refinement, we perform view frustum culling and triangle stripping. These three components are done together in a single pass over the mesh. We show how this framework supports virtually any error metric, while still being highly memory and compute efficient.
IEEE Visualization	A Memory Insensitive Technique for Large Model Simplification.	Peter Lindstrom,Cláudio T. Silva	2001	"In this paper we propose three simple, but significant improvements to the OoCS (Out-of-Core Simplification) algorithm of Lindstrom [20] which increase the quality of approximations and extend the applicability of the algorithm to an even larger class of compute systems.The original OoCS algorithm has memory complexity that depends on the size of the output mesh, but no dependency on the size of the input mesh. That is, it can be used to simplify meshes of arbitrarily large size, but the complexity of the output mesh is limited by the amount of memory available. Our first contribution is a version of OoCS that removes the dependency of having enough memory to hold (even) the simplified mesh. With our new algorithm, the whole process is made essentially independent of the available memory on the host computer. Our new technique uses disk instead of main memory, but it is carefully designed to avoid costly random accesses.Our two other contributions improve the quality of the approximations generated by OoCS. We propose a scheme for preserving surface boundaries which does not use connectivity information, and a scheme for constraining the position of the ""representative vertex"" of a grid cell to an optimal position inside the cell."
IEEE Visualization	Fitting Subdivision Surfaces.	Nathan Litke,Adi Levin,Peter Schröder	2001	We introduce a new algorithm for fitting a Catmull-Clark subdivision surface to a given shape within a prescribed tolerance, based on the method of quasi-interpolation. The fitting algorithm is fast, local and scales well since it does not require the solution of linear systems. Its convergence rate is optimal for regular meshes and our experiments show that it behaves very well for irregular meshes. We demonstrate the power and versatility of our method with examples from interactive modeling, surface fitting, and scientific visualization.
IEEE Visualization	Texture Hardware Assisted Rendering of Time-Varying Volume Data.	Eric B. Lum,Kwan-Liu Ma,John Clyne	2001	In this paper we present a hardware-assisted rendering technique coupled with a compression scheme for the interactive visual exploration of time-varying scalar volume data. A palette-based decoding technique and an adaptive bit allocation scheme are developed to fully utilize the texturing capability of a commodity 3-D graphics card. Using a single PC equipped with a modest amount of memory, a texture capable graphics card, and an inexpensive disk array, we are able to render hundreds of time steps of regularly gridded volume data (up to 45 millions voxels each time step) at interactive rates, permitting the visual exploration of large scientific data sets in both the temporal and spatial domain.
IEEE Visualization	RTVR - A Flexible Java Library for Interactive Volume Rendering.	Lukas Mroz,Helwig Hauser	2001	This paper presents several distinguishing design features of RTVR - a Java-based library for real-time volume rendering. We describe, how the careful design of data structures, which in our case are based on voxel enumeration, and an intelligent use of look-up tables enable interactive volume rendering even on low-end PC hardware. By assigning voxels to distinct objects within the volume and by using an individual setup and combination of look-up tables for each object, object-aware rendering is performed: different transfer functions, shading models, and also compositing modes can be mixed within a single scene to depict each object in the most appropriate way, while still providing rendering results in real-time. While providing frame rates similar to volume visualization using 3D consumer hardware, the approach utilized by RTVR offers much more flexibility and extensibility due to its pure software nature. Furthermore, due to the memory-efficiency of the data representation and the implementation in Java, RTVR can be used to provide volume viewing facilities over low-bandwidth networks, with almost full control over rendering and visualization mapping parameters (clipping, shading, compositing, transfer function) for the user. This paper also addresses specific problems which arise by the use of Java for interactive Visualization.
IEEE Visualization	Semi-Immersive Space Mission Design and Visualization: Case Study of the Terrestrial Planet Finder Mission.	Ken Museth,Alan H. Barr,Martin W. Lo	2001	"This paper addresses visualization issues of the Terrestrial Planet Finder Mission[2]. The goal of this mission is to search for chemical signatures of life in distant solar systems using five satellites flying in formation to simulate a large telescope. To design and visually verify such a delicate mission one has to analyze and interact with many different 3D spacecraft trajectories, which is often difficult in 2D. We employ a novel trajectory design approach using invariant manifold theory, which is best understood and utilized in an immersive setting. The visualization also addresses multi-scale issues related to the vast differences in distance, velocity, and time at different phases of the mission. Additionally, the parameterization and coordinate frames used for numerical simulations may not be suitable for direct visualization. Relative motion presents a more serious problem where the patterns of the trajectories can only be viewed in particular rotating frames. Some of these problems are greatly relieved by using interactive, animated stereo 3D visualization in a semi-immersive environment such as a Responsive Workbench. Others were solved using standard techniques such as a stratify approach with multiple windows to address the multi-scale issues, re-parameterizations of trajectories and associated 2D manifolds and relative motion of the camera to ""evoke"" the desired patterns."
IEEE Visualization	A Virtual Environment for Simulated Rat Dissection: A Case Study of Visualization for Astronaut Training.	Kevin Montgomery,LeRoy Heinrichs,Cynthia Bruyns,Simon Wildermuth	2001	A Virtual Environment for Simulated Rat Dissection: A Case Study of Visualization for Astronaut Training.
IEEE Visualization	Surgical Simulator for Hysteroscopy: A Case Study of Visualization in Surgical Training.	Kevin Montgomery,LeRoy Heinrichs,Cynthia Bruyns,Simon Wildermuth,Christopher J. Hasser,Stephanie Ozenne,David Bailey	2001	Computer-based surgical simulation promises to provide a broader scope of clinical training through the introduction of anatomic variation, simulation of untoward events, and collection of performance data. We present a haptically-enabled surgical simulator for the most common techniques in diagnostic and operative hysteroscopy- cervical dilation, endometrial resection and ablation, and lesion excision. Engineering tradeoffs in developing a real-time, haptic-rate simulator are discussed.
IEEE Visualization	Case Study on Real-Time Visualization of Virtual Tübingen on Commodity PC Hardware.	Michael Meißner,Jasmina Orman,Stephan J. Braun	2001	Case Study on Real-Time Visualization of Virtual Tübingen on Commodity PC Hardware.
IEEE Visualization	User-Centric Viewpoint Computations for Haptic Exploration and Manipulation.	Miguel A. Otaduy,Ming C. Lin	2001	"We present several techniques for user-centric viewing of the virtual objects or datasets under haptic exploration and manipulation. Depending on the type of tasks performed by the user, our algorithms compute automatic placement of the user viewpoint to navigate through the scene, to display the near-optimal views, and to reposition the viewpoint for haptic visualization. This is accomplished by conjecturing the user's intent based on the user's actions, the object geometry, and intra- and inter-object occlusion relationships. These algorithms have been implemented and interfaced with both a 3-DOF and a 6-DOF PHANToM arms. We demonstrate their application on haptic exploration and visualization of a complex structure, as well as multiresolution modeling and 3D painting with a haptic interface."
IEEE Visualization	Fast Detection of Meaningful Isosurfaces for Volume Data Visualization.	Vladimir Pekar,Rafael Wiemker,Daniel Hempel	2001	Automatic detection of meaningful isosurfaces is important for producing informative visualizations of volume data, especially when no information about the data origin and imaging protocol is available. We propose a computationally efficient method for the automated detection of intensity transitions in volume data. In this approach, the dominant transitions correspond to clear maxima in cumulative Laplacian-weighted gray value histograms. Only one pass through the data volume is required to compute the histogram. Several other features which may be useful for exploration of data of unknown origin can be efficiently computed in a similar manner, e.g. enclosed volume, isosurface area, mean gradient.The detected intensity transitions can be used for setting of visualization parameters for surface rendering, as well as for direct volume rendering of 3-D datasets. When using surface rendering, the detected dominant intensity transition values correspond to the optimal surface isovalues for extraction of boundaries of the objects of interest. In direct volume rendering, such transitions are important for generation of the transfer functions, which are used to assign visualization properties to data voxels and determine the appearance of the rendered image. The proposed method is illustrated by examples with synthetic data as well as real biomedical datasets.
IEEE Visualization	A Simple Algorithm for Surface Denoising.	Jianbo Peng,Vasily Strela,Denis Zorin	2001	We present a simple denoising technique for geometric data represented as a semiregular mesh, based on locally adaptive Wiener filtering. The degree of denoising is controlled by a single parameter (an estimate of the relative noise level) and the time required for denoising is independent of the magnitude of the estimate. The performance of the algorihm is sufficiently fast to allow interactive local denoising.
IEEE Visualization	Visualization of Sports using Motion Trajectories: Providing Insights into Performance, Style, and Strategy.	Gopal Sarma Pingali,Agata Opalach,Yves Jean,Ingrid Carlbom	2001	Remote experience of sporting events has thus far been limited mostly to watching video and the scores and statistics associated with the sport. However, a fast-developing trend is the use of visualization techniques to give new insights into performance, style, and strategy of the players. Automated techniques can extract accurate information from video about player performance that not even the most skilled observer is able to discern. When presented as static images or as a three-dimensional virtual replay, this information makes viewing a game an entirely new and exciting experience.This paper presents one such sports visualization system called LucentVision, which has been developed for the sport of tennis. LucentVision uses real-time video analysis to obtain motion trajectories of players and the ball, and offers a rich set of visualization options based on this trajectory data. The system has been used extensively in the broadcast of international tennis tournaments, both on television and the Internet.
IEEE Visualization	"The ""Which Blair Project"": A Quick Visual Method for Evaluating Perceptual Color Maps."	Bernice E. Rogowitz,Alan D. Kalvin	2001	We have developed a fast, perceptual method for selecting color scales for data visualization that takes advantage of our sensitivity to luminance variations in human faces. To do so, we conducted experiments in which we mapped various color scales onto the intensitiy values of a digitized photograph of a face and asked observers to rate each image. We found a very strong correlation between the perceived naturalness of the images and the degree to which the underlying color scales increased monotonically in luminance. Color scales that did not include a monotonically-increasing luminance component produced no positive rating scores. Since color scales with monotonic luminance profiles are widely recommended for visualizing continuous scalar data, a purely visual technique for identifying such color scales could be very useful, especially in situations where color calibration is not integrated into the visualization environment, such as over the Internet.
IEEE Visualization	A Tetrahedra-Based Stream Surface Algorithm.	Gerik Scheuermann,Tom Bobach,Hans Hagen,Karim Mahrous,Bernd Hamann,Kenneth I. Joy,Wolfgang Kollmann	2001	This paper presents a new algorithm for the calculation of stream surfaces for tetrahedral grids. It propagates the surface through the tetrahedra, one at a time, calculating the intersections with the tetrahedral faces. The method allows us to incorporate topological information from the cells, e.g., critical points. The calculations are based on barycentric coordinates, since this simplifies theory and algorithm. The stream surfaces are ruled surfaces inside each cell, and their construction is started with line segments on the faces. Our method supports the analysis of velocity fields resulting from computational fluid dynamics (CFD) simulations.
IEEE Visualization	The Perspective Shear-Warp Algorithm in a Virtual Environment.	Jürgen P. Schulze,Roland Niemeier,Ulrich Lang	2001	"Since the original paper of Lacroute and Levoy [9], where the shear-warp factorization was also shown for perspective projections, a lot of work has been carried out using the shear-warp factorization with parallel projections. However, none of it has proved or improved the algorithm for the perspective projection. Also in Lacroute's Volpack library, the perspective shear-warp volume rendering algorithm is missing.This paper reports on an implementation of the perspective shear-warp algorithm, which includes enhancements for its application in immersive virtual environments. Furthermore, a mathematical proof for the correctness of the permutation of projection and warp is provided, so far a basic assumption of the shear-warp perspective projection."
IEEE Visualization	Approximate Shading for the Re-Illumination of Synthetic Images.	Randy K. Scoggins,Raghu Machiraju,Robert J. Moorhead	2001	This paper presents a novel method to estimate illumination-dependent properties in image synthesis prior to rendering. A preprocessing step is described in which a linear image basis is developed and a lighting-independent formulation defined. A reflection function, similar to hemispherical reflectance, approximates normal Lambertian shading. Intensity errors resulting from this approximation are reduced by use of a polynomial gamma correction function and scaling to a normalized display range. This produces images that are similar to normal Lambertian shading without employing the maximum (max) function. For a single object view, images can then be expressed in a linear form so that lighting direction can be factored out. During normal rendering, image quantities for arbitrary light directions can be found without rendering. This method is demonstrated for estimating image intensity and level-of-detail (LOD) error prior to rendering an object.
IEEE Visualization	Wind Tunnel Data Fusion and Immersive Visualization: A Case Study.	Kurt Severance,Paul Brewster,Barry Lazos,Daniel F. Keef	2001	This case study describes the process of fusing the data from several wind tunnel experiments into a single coherent visualization. Each experiment was conducted independently and was designed to explore different flow features around airplane landing gear. In the past, it would have been very difficult to correlate results from the different experiments. However, with a single 3-D visualization representing the fusion of the three experiments, significant insight into the composite flowfield was observed that would have been extremely difficult to obtain by studying its component parts. The results are even more compelling when viewed in an immersive environment.
IEEE Visualization	Efficient Adaptive Simplification of Massive Meshes.	Eric Shaffer,Michael Garland	2001	The growing availability of massive polygonal models, and the inability of most existing visualization tools to work with such data, has created a pressing need for memory efficient methods capable of simplifying very large meshes. In this paper, we present a method for performing adaptive simplification of polygonal meshes that are too large to fit in-core.Our algorithm performs two passes over an input mesh. In the first pass, the model is quantized using a uniform grid, and surface information is accumulated in the form of quadrics and dual quadrics. This sampling is then used to construct a BSP-Tree in which the partitioning planes are determined by the dual quadrics. In the final pass, the original vertices are clustered using the BSP-Tree, yielding an adaptive approximation of the original mesh. The BSP-Tree describes a natural simplification hierarchy, making it possible to generate a progressive transmission and construct level-of-detail representations. In this way, the algorithm provides some of the features associated with more expensive edge contraction methods while maintaining greater computational efficiency. In addition to performing adaptive simplification, our algorithm exhibits output-sensitive memory requirements and allows fine control over the size of the simplified mesh.
IEEE Visualization	A Case Study on Interactive Exploration and Guidance Aids for Visualizing Historical Data.	Stanislav L. Stoev,Wolfgang Straßer	2001	"In this paper, we address the problem of historical data visualization. We describe the data acquisition, preparation, and visualization. Since the data contain four dimensions, the standard 3D exploration techniques have to be extended or appropriately adapted in order to enable interactive exploration. We discuss in detail two interaction concepts: (1) navigation with one fixed dimension, and (2) quasi 4D navigation allowing to simultaneously explore the four-dimensional space. In addition, we also present a picture-in-picture display mode, enabling the user to interactively view the data, while ""flying with"" a particular event, tracking its motion in time and space. Finally, we present a technique for guided exploration and animation generation, allowing for a vivid gain of insight into the historical data."
IEEE Visualization	Salient Iso-Surface Detection with Model-Independent Statistical Signatures.	Shivaraj Tenginakai,Jinho Lee,Raghu Machiraju	2001	Volume graphics has not been accepted for widespread use. One of the inhibiting reasons is the lack of general methods for data-analysis and simple interfaces for data exploration. An error-and-trial iterative procedure is often used to select a desirable transfer function or mine the dataset for salient iso-values. New semi-automatic methods that are also data-centric have shown much promise [1][7]. However, general and robust methods are still needed for data-exploration and analysis. In this paper, we propose general model-independent statistical methods based on central moments of data. Using these techniques we show how salient iso-surfaces at material boundaries can be determined. We provide examples from the medical and computational domain to demonstrate the effectiveness of our methods.
IEEE Visualization	Optimal Regular Volume Sampling.	Thomas Theußl,Torsten Möller,Eduard Gröller	2001	The classification of volumetric data sets as well as their rendering algorithms are typically based on the representation of the underlying grid. Grid structures based on a Cartesian lattice are the de-facto standard for regular representations of volumetric data. In this paper we introduce a more general concept of regular grids for the representation of volumetric data. We demonstrate that a specific type of regular lattice --- the so-called body-centered cubic --- is able to represent the same data set as a Cartesian grid to the same accuracy but with 29.3% fewer samples. This speeds up traditional volume rendering algorithms by the same ratio, which we demonstrate by adopting a splatting implementation for these new lattices. We investigate different filtering methods required for computing the normals on this lattice. The lattice representation results also in lossless compression ratios that are better than previously reported. Although other regular grid structures achieve the same sample efficiency, the body-centered cubic is particularly easy to use. The only assumption necessary is that the underlying volume is isotropic and band-limited - an assumption that is valid for most practical data sets.
IEEE Visualization	4D Space-Time Techniques: A Medical Imaging Case Study.	Melanie Tory,Niklas Röber,Torsten Möller,Anna Celler,M. Stella Atkins	2001	We present the problem of visualizing time-varying medical data. Two medical imaging modalities are compared --- MRI and dynamic SPECT. For each modality, we examine several derived scalar and vector quantities such as the change in intensity over time, the spatial gradient, and the change of the gradient over time. We compare several methods for presenting the data, including isosurfaces, direct volume rendering, and vector visualization using glyphs. These techniques may provide more information and context than methods currently used in practice; thus it is easier to discover temporal changes and abnormalities in a data set.
IEEE Visualization	Continuous Topology Simplification of Planar Vector Fields.	Xavier Tricoche,Gerik Scheuermann,Hans Hagen	2001	Vector fields can present complex structural behavior, especially in turbulent computational fluid dynamics. The topological analysis of these datasets reduces the information but one is usually still left with too many details for interpretation. In this paper, we present a simplification approach that removes pairs of critical points from the dataset, based on relevance measures. In contrast to earlier methods, no grid changes are necessary since the whole method uses small local changes of the vector values defining the vector field. An interpretation in terms of bifurcations underlines the continuous, natural flavor of the algorithm.
IEEE Visualization	Attribute Preserving Dataset Simplification.	Jason D. Walter,Christopher G. Healey	2001	"This paper describes a novel application of feature preserving mesh simplification to the problem of managing large, multidimensional datasets during scientific visualization. To allow this, we view a scientific dataset as a triangulated mesh of data elements, where the attributes embedded in each element form a set of properties arrayed across the surface of the mesh. Existing simplification techniques were not designed to address the high dimensionality that exists in these types of datasets. As well, vertex operations that relocate, insert, or remove data elements may need to be modified or restricted. Principal component analysis provides an algorithm-independent method for compressing a dataset's dimensionality during simplification. Vertex locking forces certain data elements maintain their spatial locations; this technique is also used to guarantee a minimum density in the simplified dataset. The result is a visualization that significantly reduces the number of data elements to display, while at the same time ensuring that high-variance regions of potential interest remain intact. We apply our techniques to a number of well-known feature preserving algorithms, and demonstrate their applicability in a real-world context by simplifying a multidimensional weather dataset. Our results show a significant improvement in execution time with only a small reduction in accuracy; even when the dataset was simplified to 10% of its original size, average per attribute error was less than 1%."
IEEE Visualization	Distance-Field-Based Skeletons for Virtual Navigation.	Ming Wan,Frank Dachille,Arie E. Kaufman	2001	We present a generic method for rapid flight planning, virtual navigation and effective camera control in a volumetric environment. Directly derived from an accurate distance from boundary (DFB) field, our automatic path planning algorithm rapidly generates centered flight paths, a skeleton, in the navigable region of the virtual environment. Based on precomputed flight paths and the DFB field, our dual-mode physically based camera control model supports a smooth, safe, and sticking-free virtual navigation with six degrees of freedom. By using these techniques, combined with accelerated volume rendering, we have successfully developed a real-time virtual colonoscopy system on low-cost PCs and confirmed the high speed, high accuracy and robustness of our techniques on more than 40 patient datasets.
IEEE Visualization	Case Study: Visualization of Particle Track Data.	Xiaoming Wei,Arie E. Kaufman,Timothy J. Hallman	2001	The Relativistic Heavy Ion Collider (RHIC) experiment at the Brookhaven National Lab is designed to study how the universe came into being. It is believed that after the Big Bang, the universe expanded and cooled, consisting of a soup of quarks, gluons, electrons and neutrinos. As the temperature lowered, electrons combined with protons and formed neutral atoms. Later, clouds of atoms contracted into stars. In this paper, we describe how techniques of volume rendering and information visualization are used to visualize the large particle track data set generated from this high energy physics experiment. The system, called TrackVis, is based on our earlier work of VolVis - Volume Visualization software. Example images of real particle collision data are shown, which are helpful to physicists in investigating the behavior of strongly interacting matter at high energy density.
IEEE Visualization	Hardware-Software-Balanced Resampling for the Interactive Visualization of Unstructured Grids.	Manfred Weiler,Thomas Ertl	2001	In this paper we address the problem of interactively resampling unstructured grids. Three algorithms are presented. They all allow adaptive resampling of an unstructured grid on a multiresolution hierarchy of arbitrarily sized cartesian grids according to a varying element size. Two of the algorithms presented take advantage of hardware accelerated polygon rendering and 2D texture mapping. In exploiting new features of modern PC graphics adapters, the first algorithm tries to significantly minimize the number of polygons to be rendered. Reducing rasterization requirements is the main goal of the second algorithm, which distributes the computational workload differently between the main processor and the graphics chip. By comparing them to a new pure software approach, an optimal software-hardware balance is studied. We end up with a hybrid approach which greatly improves the performance of hardware-assisted resampling by involving the main processor to a higher degree and thus enabling resampling at nearly interactive rates.
IEEE Visualization	Accelerated Volume Ray-Casting using Texture Mapping.	Rüdiger Westermann,Bernd Sevenich	2001	Acceleration techniques for volume ray-casting are primarily based on pre-computed data structures that allow one to efficiently traverse empty or homogeneous regions. In order to display volume data that successively undergoes color lookups, however, the data structures have to be re-built continuously. In this paper we propose a technique that circumvents this drawback using hardware accelerated texture mapping. In a first rendering pass we employ graphics hardware to interactively determine for each ray where the material is hit. In a second pass ray-casting is performed, but ray traversal starts right in front of the previously determined regions. The algorithm enables interactive classification and it considerably accelerates the view dependent display of selected materials and surfaces from volume data. In contrast to other techniques that are solely based on texture mapping our approach requires less memory and accurately performs the composition of material contributions along the ray.
IEEE Visualization	Enridged Contour Maps.	Jarke J. van Wijk,Alexandru Telea	2001	The visualization of scalar functions of two variables is a classic and ubiquitous application. We present a new method to visualize such data. The method is based on a non-linear mapping of the function to a height field, followed by visualization as a shaded mountain landscape. The method is easy to implement and efficient, and leads to intriguing and insightful images: The visualization is enriched by adding ridges. Three types of applications are discussed: visualization of iso-levels, clusters (multivariate data visualization), and dense contours (flow visualization).
IEEE Visualization	EWA Volume Splatting.	Matthias Zwicker,Hanspeter Pfister,Jeroen van Baar,Markus H. Gross	2001	"In this paper we present a novel framework for direct volume rendering using a splatting approach based on elliptical Gaussian kernels. To avoid aliasing artifacts, we introduce the concept of a resampling filter combining a reconstruction with a low-pass kernel. Because of the similarity to Heckbert's EWA (elliptical weighted average) filter for texture mapping we call our technique EWA volume splatting. It provides high image quality without aliasing artifacts or excessive blurring even with non-spherical kernels. Hence it is suitable for regular, rectilinear, and irregular volume data sets. Moreover, our framework introduces a novel approach to compute the footprint function. It facilitates efficient perspective projection of arbitrary elliptical kernels at very little additional cost. Finally, we show that EWA volume reconstruction kernels can be reduced to surface reconstruction kernels. This makes our splat primitive universal in reconstructing surface and volume data."
IEEE Visualization	Graphical Strategies to Convey Functional Relationships in the Human Brain: A Case Study.	Tomihisa Welsh,Klaus Mueller,Wei Zhu,Nora D. Volkow,Jeffrey Meade	2001	Brain imaging methods used in experimental brain research such as Positron Emission Tomography (PET) and Functional Magnetic Resonance (fMRI) require the analysis of large amounts of data. Exploratory statistical methods can be used to generate new hypotheses and to provide a reliable measure of a given effect. Typically, researchers report their findings by listing those regions which show significant statistical activity in a group of subjects under some experimental condition or task. A number of methods create statistical parametric maps (SPMs) of the brain on a voxelbasis. In our approach statistics are computed not on individual voxels but on predefined anatomical regions-of-interest (ROIs). A correlation coefficient is used to quantify similarity in response for various regions during an experimental setting. Since the functional inter-relationships can become rather complex and spatially widespread, they are best understood in the context of the underlying 3-D brain anatomy. However, despite the power of the 3-D model, the relative location of ROIs in 3-D can be obscured due the inherent problem of presenting 3-D spatial information on a 2-D screen. In order to address this problem, we have explored a number of visualization techniques to aid the brain researcher in exploring the spatial relationships of brain activity. In this paper, we present a novel 3-D interface that allows the interactive exploration of correlation datasets.
IEEE Visualization	PixelFlex: A Reconfigurable Multi-Projector Display System.	Ruigang Yang,David Gotz,Justin Hensley,Herman Towles,Michael S. Brown	2001	This paper presents PixelFlex --- a spatially reconfigurable multi-projector display system. The PixelFlex system is composed of ceiling-mounted projectors, each with computer-controlled pan, tilt, zoom and focus; and a camera for closed-loop calibration. Working collectively, these controllable projectors function as a single logical display capable of being easily modified into a variety of spatial formats of differing pixel density, size and shape. New layouts are automatically calibrated within minutes to generate the accurate warping and blending functions needed to produce seamless imagery across planar display surfaces, thus giving the user the flexibility to quickly create, save and restore multiple screen configurations.Overall, PixelFlex provides a new level of automatic reconfigurability and usage, departing from the static, one-size-fits-all design of traditional large format displays. As a front-projection system, PixelFlex can be installed in most environments with space constraints and requires little or no post-installation mechanical maintenance because of the closed-loop calibration.
IEEE Visualization	Nonmanifold Subdivision.	Lexing Ying,Denis Zorin	2001	Commonly-used subdivision schemes require manifold control meshes and produce manifold surfaces. However, it is often necessary to model nonmanifold surfaces, such as several surface patches meeting at a common boundary.In this paper, we describe a subdivision algorithm that makes it possible to model nonmanifold surfaces. Any triangle mesh, subject only to the restriction that no two vertices of any triangle coincide, can serve as an input to the algorithm. Resulting surfaces consist of collections of manifold patches joined along nonmanifold curves and vertices. If desired, constraints may be imposed on the tangent planes of manifold patches sharing a curve or a vertex.The algorithm is an extension of a well-known Loop subdivision scheme, and uses techniques developed for piecewise smooth surfaces.
IEEE Visualization	An Immersive Virtual Environment for DT-MRI Volume Visualization Applications: A Case Study.	Song Zhang,Çagatay Demiralp,Daniel F. Keefe,M. DaSilva,David H. Laidlaw,B. D. Greenberg,Peter J. Basser,Carlo Pierpaoli,E. A. Chiocca,Thomas S. Deisboeck	2001	We describe a virtual reality environment for visualizing tensor-valued volumetric datasets acquired with diffusion tensor magnetic resonance imaging (DT-MRI). We have prototyped a virtual environment that displays geometric representations of the volumetric second-order diffusion tensor data and are developing interaction and visualization techniques for two application areas: studying changes in white-matter structures after gamma-knife capsulotomy and pre-operative planning for brain tumor surgery. Our feedback shows that compared to desktop displays, our system helps the user better interpret the large and complex geometric models, and facilitates communication among a group of users.
IEEE Visualization	The MetVR Case Study: Meteorological Visualization in an Immersive Virtual Environment.	Sean Ziegeler,Robert J. Moorhead,Paul J. Croft,Duanjun Lu	2001	"Traditional methods for displaying weather products are generally two-dimensional (2D) plots or just text format. It is hard for forecasters to get the entire picture of the atmosphere using these methods. The problems apparent in 2D with comparing and correlating multiple layers are overcome simply by adding a dimension. This is important because pertinent features in the data sets may lie in multiple layers and span several time steps. However, simply using a three-dimensional (3D) approach is not enough. The capacity for analysis of small-scale, but important, features in 2D is lost when transitioning to 3D. We propose that 3D's advantages can be incorporated with 2D's small-scale analysis by using an immersive virtual environment. In this case study, we evaluate our current standing with the project: have we met our goals, and how should we proceed from this point? To evaluate our application, we invited meteorologists to use the application to explore a data set. Then we presented our goals and asked which ones had we met, from a meteorologist's perspective. The results qualitatively reflected that our application was effective and further research would be worthwhile."
IEEE Visualization	IEEE Visualization 2001, October 24-26, 2001, San Diego, CA, USA, Proceedings	Thomas Ertl,Kenneth I. Joy,Amitabh Varshney	2001	IEEE Visualization 2001, October 24-26, 2001, San Diego, CA, USA, Proceedings
IEEE Trans. Vis. Comput. Graph.	Nonlinear Multiresolution Techniques with Applications to Scientific Visualization in a Haptic Environment.	Mohammad Waqas Asghar,Kenneth E. Barner	2001	This paper develops nonlinear multiresolution techniques for scientific visualization utilizing haptic methods. The visualization of data is critical to many areas of scientific pursuit. Scientific visualization is generally accomplished through computer graphic techniques. Recent advances in haptic technologies allow visual techniques to be augmented with haptic methods. The kinesthetic feedback provided through haptic techniques provides a second modality for visualization and allows for active exploration. Moreover, haptic methods can be utilized by individuals with visual impairments. Haptic representations of large data sets, however, can be confusing to a user, especially if a visual representation is not available or cannot be used. Additionally, most haptic devices utilize point interactions, resulting in a low information bandwidth and further complicating data exploration. Multiresolution techniques can be utilized to address the issues of low information bandwidth and data complexity. Commonly used multiresolution techniques are based on the wavelet decomposition. Such linear techniques, however, tend to smooth important data features, such as discontinuities or edges. In contrast, nonlinear techniques can be utilized that preserve edge structures while removing fine data details. This paper develops a multiresolution data decomposition method based on the affine median filter. This results in a hybrid structure that can be tuned to yield a decomposition that varies from a linear wavelet decomposition to that produced by the median filter. The performance of this hybrid structure is analyzed utilizing deterministic signals and statistically in the frequency domain. This analysis and qualitative and quantitative implementation results show that the affine median decomposition has advantages over previously proposed methods. In addition to multiresolution decomposition development, analysis, and results, haptic implementation methods are presented.
IEEE Trans. Vis. Comput. Graph.	Texture Mixing and Texture Movie Synthesis Using Statistical Learning.	Ziv Bar-Joseph,Ran El-Yaniv,Dani Lischinski,Michael Werman	2001	We present an algorithm based on statistical learning for synthesizing static and time-varying textures matching the appearance of an input texture. Our algorithm is general and automatic and it works well on various types of textures, including 1D sound textures, 2D texture images, and 3D texture movies. The same method is also used to generate 2D texture mixtures that simultaneously capture the appearance of a number of different input textures. In our approach, input textures are treated as sample signals generated by a stochastic process. We first construct a tree representing a hierarchical multiscale transform of the signal using wavelets. From this tree, new random trees are generated by learning and sampling the conditional probabilities of the paths in the original tree. Transformation of these random trees back into signals results in new random textures. In the case of 2D texture synthesis, our algorithm produces results that are generally as good as or better than those produced by previously described methods in this field. For texture mixtures, our results are better and more general than those produced by earlier methods. For texture movies, we present the first algorithm that is able to automatically generate movie clips of dynamic phenomena such as waterfalls, fire flames, a school of jellyfish, a crowd of people, etc. Our results indicate that the proposed technique is effective and robust.
IEEE Trans. Vis. Comput. Graph.	Parameterization and Reconstruction from 3D Scattered Points Based on Neural Network and PDE Techniques.	Jacob Barhak,Anath Fischer	2001	Reverse engineering ordinarily uses laser scanners since they can sample 3D data quickly and accurately relative to other systems. These laser scanner systems, however, yield an enormous amount of irregular and scattered digitized point data that requires intensive reconstruction processing. Reconstruction of freeform objects consists of two main stages: 1) parameterization and 2) surface fitting. Selection of an appropriate parameterization is essential for topology reconstruction as well as surface fitness. Current parameterization methods have topological problems that lead to undesired surface fitting results, such as noisy self-intersecting surfaces. Such problems are particularly common with concave shapes whose parametric grid is self-intersecting, resulting in a fitted surface that considerably twists and changes its original shape. In such cases, other parameterization approaches should be used in order to guarantee non-self-intersecting behavior. The parameterization method described in this paper is based on two stages: 1) 2D initial parameterization and 2) 3D adaptive parameterization. Two methods were developed for the first stage: Partial Differential Equation (PDE) parameterization and neural network Self Organizing Maps (SOM) parameterization. PDE parameterization yields a parametric grid without self-intersections. Neural network SOM parameterization creates a grid where all the sampled points, not only the boundary points, affect the grid, leading to a uniform and smooth surface. In the second stage, a 3D base surface was created and then adaptively modified. To this end, the Gradient Descent Algorithm (GDA) and Random Surface Error Correction (RSEC), both of which are iterative surface fitting methods, were developed and implemented. The feasibility of the developed parameterization methods and fitting algorithms is demonstrated on several examples using sculptured free objects.
IEEE Trans. Vis. Comput. Graph.	Minimally Immersive Flow Visualization.	David S. Ebert,Christopher D. Shaw	2001	This paper describes a minimally immersive interactive system for flow visualization of multivariate volumetric data. The system, SFA, uses perceptually motivated rendering to increase the quantity and clarity of information perceived. Proprioception, stereopsis, perceptually motivated shape visualization, and three-dimensional interaction are combined in SFA to allow the three-dimensional volumetric visualization, manipulation, navigation, and analysis of multivariate, time-varying flow data.
IEEE Trans. Vis. Comput. Graph.	High-Quality Texture Reconstruction from Multiple Scans.	Fausto Bernardini,Ioana M. Martin,Holly E. Rushmeier	2001	The creation of three-dimensional digital content by scanning real objects has become common practice in graphics applications for which visual quality is paramount, such as animation, e-commerce, and virtual museums. While a lot of attention has been devoted recently to the problem of accurately capturing the geometry of scanned objects, the acquisition of high-quality textures is equally important, but not as widely studied. In this paper, we focus on methods to construct accurate digital models of scanned objects by integrating high-quality texture and normal maps with geometric data. These methods are designed for use with inexpensive, electronic camera-based systems in which low-resolution range images and high-resolution intensity images are acquired. The resulting models are well-suited for interactive rendering on the latest-generation graphics hardware with support for bump mapping. Our contributions include new techniques for processing range, reflectance, and surface normal data, for image-based registration of scans, and for reconstructing high-quality textures for the output digital object.
IEEE Trans. Vis. Comput. Graph.	Penalized-Distance Volumetric Skeleton Algorithm.	Ingmar Bitter,Arie E. Kaufman,Mie Sato	2001	This paper introduces a refined general definition of a skeleton that is based on a penalized-distance function and cannot create any of the degenerate cases of the earlier Ceasar and Teasar algorithms. Additionally, we provide an algorithm that finds the skeleton accurately and rapidly. Our solution is fully automatic, which frees the user from having to engage in manual data preprocessing. We present the accurate skeletons computed on a number of test datasets. The algorithm is very efficient as demonstrated by the running times which were all below seven minutes.
IEEE Trans. Vis. Comput. Graph.	A Level-Set Approach for the Metamorphosis of Solid Models.	David E. Breen,Ross T. Whitaker	2001	This paper presents a new approach to 3D shape metamorphosis. We express the interpolation of two shapes as a process where one shape deforms to maximize its similarity with another shape. The process incrementally optimizes an objective function while deforming an implicit surface model. We represent the deformable surface as a level set (iso-surface) of a densely sampled scalar function of three dimensions. Such level-set models have been shown to mimic conventional parametric deformable surface models by encoding surface movements as changes in the grayscale values of a volume data set. Thus, a well-founded mathematical structure leads to a set of procedures that describes how voxel values can be manipulated to create deformations that are represented as a sequence of volumes. The result is a 3D morphing method that offers several advantages over previous methods, including minimal need for user input, no model parameterization, flexible topology, and subvoxel accuracy.
IEEE Trans. Vis. Comput. Graph.	Preventing Self-Intersection under Free-Form Deformation.	James E. Gain,Neil A. Dodgson	2001	"Free-Form Deformation (FFD) is a versatile and efficient modeling technique which transforms an object by warping the surrounding space. The conventional user-interface is a lattice of movable control points but this tends to be cumbersome and counterintuitive. Directly Manipulated Free-Form Deformation (DMFFD) allows the user to drag object points directly and has proven useful in an interactive sculpting context. A serious shortcoming of both FFD and DMFFD is that some deformations cause self-intersection of the object. This is unrealistic and compromises the object's validity and suitability for later use. An in-built self-intersection test is thus required for FFD and its extensions to be truly robust. In this paper, we present the following novel results: a set of theoretical conditions for preventing self-intersection by ensuring the injectivity (one-to-one mapping) of FFD, an exact (necessary and sufficient) injectivity test which is accurate but computationally costly, an efficient but approximate injectivity test which is a sufficient condition only, and a new form of DMFFD which acts by composing many small injective deformations. The latter expands the range of possible deformations without sacrificing the speed of the approximate test."
IEEE Trans. Vis. Comput. Graph.	A Phase Field Model for Continuous Clustering on Vector Fields.	Harald Garcke,Tobias Preußer,Martin Rumpf,Alexandru Telea,Ulrich Weikard,Jarke J. van Wijk	2001	A new method for the simplification of flow fields is presented. It is based on continuous clustering. A well-known physical clustering model, the Cahn Hilliard model, which describes phase separation, is modified to reflect the properties of the data to be visualized. Clusters are defined implicitly as connected components of the positivity set of a density function. An evolution equation for this function is obtained as a suitable gradient flow of an underlying anisotropic energy functional. Here, time serves as the scale parameter. The evolution is characterized by a successive coarsening of patterns¿the actual clustering¿during which the underlying simulation data specifies preferable pattern boundaries. We introduce specific physical quantities in the simulation to control the shape, orientation and distribution of the clusters as a function of the underlying flow field. In addition, the model is expanded, involving elastic effects. In the early stages of the evolution shear layer type representation of the flow field can thereby be generated, whereas, for later stages, the distribution of clusters can be influenced. Furthermore, we incorporate upwind ideas to give the clusters an oriented drop-shaped appearance. Here, we discuss the applicability of this new type of approach mainly for flow fields, where the cluster energy penalizes cross streamline boundaries. However, the method also carries provisions for other fields as well. The clusters can be displayed directly as a flow texture. Alternatively, the clusters can be visualized by iconic representations, which are positioned by using a skeletonization algorithm.
IEEE Trans. Vis. Comput. Graph.	Terrain Decimation through Quadtree Morphing.	David Cline,Parris K. Egbert	2001	We present a new terrain decimation technique called a Quadtree Morph, or Q-morph. The new approach eliminates the usual popping artifacts associated with polygon reduction, replacing them with less objectionable smooth morphing. We show that Q-morphing is fast enough to create a view-dependent terrain model for each frame in an interactive environment. In contrast to most Geomorph algorithms, Q-morphing does not use a time step to interpolate between geometric configurations. Instead, the geometry motion in a Q-morph is based solely on the position of the viewer.
IEEE Trans. Vis. Comput. Graph.	"'Meshsweeper': Dynamic Point-to-Polygonal-Mesh Distance and Applications."	André Guéziec	2001	We introduce a new algorithm for computing the distance from a point to an arbitrary polygonal mesh. Our algorithm uses a multiresolution hierarchy of bounding volumes generated by geometric simplification. Our algorithm is dynamic, exploiting coherence between subsequent queries using a priority process and achieving constant time queries in some cases. It can be applied to meshes that transform rigidly or deform nonrigidly. We illustrate our algorithm with a simulation of particle dynamics and collisions with a deformable mesh, the computation of distance maps and offset surfaces, the computation of an approximation to the expensive Hausdorff distance between two shapes, and the detection of self-intersections. We also report comparison results between our algorithm and an alternative algorithm using an octree, upon which our method permits an order-of-magnitude speed-up.
IEEE Trans. Vis. Comput. Graph.	Cutting and Stitching: Converting Sets of Polygons to Manifold Surfaces.	André Guéziec,Gabriel Taubin,Francis Lazarus,William Horn	2001	Many real-world polygonal surfaces contain topological singularities that represent a challenge for processes such as simplification, compression, and smoothing. We present an algorithm that removes singularities from nonmanifold sets of polygons to create manifold (optionally oriented) polygonal surfaces. We identify singular vertices and edges, multiply singular vertices, and cut through singular edges. In an optional stitching operation, we maintain the surface as a manifold while joining boundary edges. We present two different edge stitching strategies, called pinching and snapping. Our algorithm manipulates the surface topology and ignores physical coordinates. Except for the optional stitching, the algorithm has a linear complexity and requires no floating point operations. In addition to introducing new algorithms, we expose the complexity (and pitfalls) associated with stitching. Finally, several real-world examples are studied.
IEEE Trans. Vis. Comput. Graph.	Efficient Conservative Visibility Culling Using the Prioritized-Layered Projection Algorithm.	James T. Klosowski,Cláudio T. Silva	2001	We propose a novel conservative visibility culling technique based on the Prioritized-Layered Projection (PLP) algorithm. PLP is a time-critical rendering technique that computes, for a given viewpoint, a partially correct image by rendering only a subset of the geometric primitives, those that PLP determines to be most likely visible. Our new algorithm builds on PLP and provides an efficient way of finding the remaining visible primitives. We do this by adding a second phase to PLP which uses image-space techniques for determining the visibility status of the remaining geometry. Another contribution of our work is to show how to efficiently implement such image-space visibility queries using currently available OpenGL hardware and extensions. We report on the implementation of our techniques on several graphics architectures, analyze their complexity, and discuss a possible hardware extension that has the potential to further increase performance.
IEEE Trans. Vis. Comput. Graph.	On Simulated Annealing and the Construction of Linear Spline Approximations for Scattered Data.	Oliver Kreylos,Bernd Hamann	2001	We describe a method to create optimal linear spline approximations to arbitrary functions of one or two variables, given as scattered data without known connectivity. We start with an initial approximation consisting of a fixed number of vertices and improve this approximation by choosing different vertices, governed by a simulated annealing algorithm. In the case of one variable, the approximation is defined by line segments; in the case of two variables, the vertices are connected to define a Delaunay triangulation of the selected subset of sites in the plane. In a second version of this algorithm, specifically designed for the bivariate case, we choose vertex sets and also change the triangulation to achieve both optimal vertex placement and optimal triangulation. We then create a hierarchy of linear spline approximations, each one being a superset of all lower-resolution ones.
IEEE Trans. Vis. Comput. Graph.	Two-Level Volume Rendering.	Helwig Hauser,Lukas Mroz,Gian Italo Bischi,Eduard Gröller	2001	In this paper, we present a two-level approach for volume rendering, i.e., two-level volume rendering, which allows for selectively using different rendering techniques for different subsets of a 3D data set. Different structures within the data set are rendered locally on an object-by-object basis by either DVR, MIP, surface rendering, value integration (x-ray-like images), or nonphotorealistic rendering. All the results of subsequent object renderings are combined globally in a merging step (usually compositing in our case). This allows us to selectively choose the most suitable technique for depicting each object within the data while keeping the amount of information contained in the image at a reasonable level. This is especially useful when inner structures should be visualized together with semitransparent outer parts, similar to the focus-plus-context approach known from information visualization. We also present an implementation of our approach which allows us to explore volumetric data using two-level rendering at interactive frame rates.
IEEE Trans. Vis. Comput. Graph.	Reliable Path for Virtual Endoscopy: Ensuring Complete Examination of Human Organs.	Taosong He,Lichan Hong,Dongqing Chen,Zhengrong Liang	2001	Virtual endoscopy is a computerized, noninvasive procedure for detecting anomalies inside human organs. Several preliminary studies have demonstrated the benefits and effectiveness of this modality. Unfortunately, previous work cannot guarantee that an existing anomaly will be detected, especially for complex organs with multiple branches. In this paper, we introduce the concept of reliable navigation, which ensures the interior organ surface is fully examined by the physician performing the virtual endoscopy procedure. To achieve this, we propose computing a reliable fly-through path that ensures no blind area during the navigation. Theoretically, we discuss the criteria of evaluating a reliable path and prove that the problem of generating an optimal reliable path for virtual endoscopy is NP-complete. In practice, we develop an efficient method for the calculation of an effective reliable path. First, a small set of center observation points are automatically located inside the hollow organ. For each observation point, there exists at least one patch of interior surface visible to it, but that cannot be seen from any of the other observation points. These chosen points are then linked with a path that stays in the center of the organ. Finally, new points inside the organ are recursively selected and connected into the path until the entire organ surface is visible from the path. We present encouraging results from experiments on several data sets. For a medium size volumetric model with several hundred thousand inner voxels, an effective reliable path can be generated in several minutes.
IEEE Trans. Vis. Comput. Graph.	Multiresolution Methods for Nonmanifold Models.	Andreas Hubeli,Markus H. Gross	2001	The concept of fairing applied to triangular meshes with irregular connectivity has become more and more important. Previous contributions proposed a variety of fairing operators for manifolds and applied them to design multiresolution representations and editing tools for meshes. In this paper, we generalize these powerful techniques to handle nonmanifold models. We propose a method to construct fairing operators for nonmanifolds which is based on standard operators for the manifold setting. Furthermore, we describe novel approaches to guarantee volume preservation. We introduce various multiresolution techniques that allow us to represent, smooth, and edit nonmanifold models efficiently. Finally, we discuss a semiautomatic feature preservation strategy to retain important model information during the fairing process.
IEEE Trans. Vis. Comput. Graph.	A Vectorial Algorithm for Tracing Discrete Straight Lines in N-Dimensional Generalized Grids.	Luis Ibáñez,Chafiaâ Hamitouche,Christian Roux	2001	This paper presents an algorithm to trace discrete straight lines in regular grids of any dimension. Most known line tracing algorithms have been developed in ${\hbox{\rlap{Z}\kern 2.0pt{\hbox{Z}}}}^{2}$ and ${\hbox{\rlap{Z}\kern 2.0pt{\hbox{Z}}}}^{3}$ orthogonal grids. The contribution of this paper is the definition of a method to trace lines in nonorthogonal grids in any dimension. This method is not restricted to being used with a specific grid connectivity as other widespread methods are. Good performance can be achieved because only additions are used during line tracing.
IEEE Trans. Vis. Comput. Graph.	Extended Specifications and Test Data Sets for Data Level Comparisons of Direct Volume Rendering Algorithms.	Kwansik Kim,Craig M. Wittenbrink,Alex Pang	2001	"Direct volume rendering (DVR) algorithms do not generate intermediate geometry to create a visualization, yet they produce countless variations in the resulting images. Therefore, comparative studies are essential for objective interpretation. Even though image and data level comparison metrics are available, it is still difficult to compare results because of the numerous rendering parameters and algorithm specifications involved. Most of the previous comparison methods use information from the final rendered images only. We overcome limitations of image level comparisons with our data level approach using intermediate rendering information. We provide a list of rendering parameters and algorithm specifications to guide comparison studies. We extend Williams and Uselton's rendering parameter list with algorithm specification items and provide guidance on how to compare algorithms. Real data are often too complex to study algorithm variations with confidence. Most of the analytic test data sets reported are often useful only for a limited feature of DVR algorithms. We provide simple and easily reproducible test data sets, a checkerboard and a ramp, that can make clear differences in a wide range of algorithm variations. With data level metrics, our test data sets make it possible to perform detailed comparison studies. A number of examples illustrate how to use these tools."
IEEE Trans. Vis. Comput. Graph.	Fast Isosurface Generation Using the Volume Thinning Algorithm.	Takayuki Itoh,Yasushi Yamaguchi,Koji Koyamada	2001	One of the most effective techniques for developing efficient isosurfacing algorithms is the reduction of visits to nonisosurface cells. Recent algorithms have drastically reduced the unnecessary cost of visiting nonisosurface cells. The experimental results show almost optimal performance in their isosurfacing processes. However, most of them have a bottleneck in that they require more than $O(n)$ computation time for their preprocessing, where $n$ denotes the total number of cells. In this paper, we propose an efficient isosurfacing technique, which can be applied to unstructured as well as structured volumes and which does not require more than $O(n)$ computation time for its preprocessing. A preprocessing step generates an extrema skeleton, which consists of cells and connects all extremum points, by the volume thinning algorithm. All disjoint parts of every isosurface intersect at least one cell in the extrema skeleton. Our implementation generates isosurfaces by searching for isosurface cells in the extrema skeleton and then recursively visiting their adjacent isosurface cells, while it skips most of the nonisosurface cells. The computation time of the preprocessing is estimated as $O(n)$. The computation time of the isosurfacing process is estimated as $O(n^{1/3} m + k)$, where $k$ denotes the number of isosurface cells and $m$ denotes the number of extremum points since the number of cells in an extrema skeleton is estimated as $O(n^{1/3} m)$.
IEEE Trans. Vis. Comput. Graph.	Visualization Exploration and Encapsulation via a Spreadsheet-Like Interface.	T. J. Jankun-Kelly,Kwan-Liu Ma	2001	Exploring complex, very large data sets requires interfaces to present and navigate through the visualization of the data. Two types of audience benefit from such coherent organization and representation: First, the user of the visualization system can examine and evaluate their data more efficiently; second, collaborators or reviewers can quickly understand and extend the visualization. The needs of these two groups are addressed by the spreadsheet-like interface described here. The interface represents a two-dimensional window into a multidimensional visualization parameter space. Data is explored by navigating this space via the interface. The visualization space is presented to the user in a manner that clearly identifies which parameters correspond to which visualized result. Operations defined on this space can be applied which generate new parameters or results. Combined with a general purpose interpreter, these functions can be utilized to quickly extract desired results. Finally, by encapsulating the visualization process, redundant exploration is eliminated and collaboration is facilitated. The efficacy of this novel interface is demonstrated through examples using a variety of data sets in different domains.
IEEE Trans. Vis. Comput. Graph.	A Graphical Representation of the State Spaces of Hierarchical Level-of-Detail Scene Descriptions.	Ashton E. W. Mason,Edwin H. Blake	2001	We present a new graphical representation of the level-of-detail state spaces generated by hierarchical geometric scene descriptions with multiple levels of detail. These level-of-detail graphs permit the analytical investigation of the hierarchical level-of-detail optimization problem that arises for such descriptions. As an example of their use, we prove the equivalence of two hierarchical level-of-detail algorithms.
IEEE Trans. Vis. Comput. Graph.	An Attempt for Coloring Multichannel MR Imaging Data.	Shigeru Muraki,Toshiharu Nakai,Yasuyo Kita,Koji Tsuda	2001	This is an elementary research for assigning color values to voxels of multichannel Magnetic Resonance Imaging (MRI) volume data. The MRI volume data sets obtained under different scanning conditions are transformed to the components by independent component analysis (ICA), which enhances physical characteristics of the tissue. The transfer functions for generating color values from independent components are obtained using the radial basis function network, a kind of neural net, by training the network with sample data chosen from visible human female data set (VHF). The resultant color volume data sets correspond well with the full-color cross-sections of the visible human data sets.
IEEE Trans. Vis. Comput. Graph.	Hierarchical Model for Real Time Simulation of Virtual Human Crowds.	Soraia Raupp Musse,Daniel Thalmann	2001	This paper describes a model for simulating crowds of humans in real time. We deal with a hierarchy composed of virtual crowds, groups, and individuals. The groups are the most complex structure that can be controlled in different degrees of autonomy. This autonomy refers to the extent to which the virtual agents are independent of user intervention and also the amount of information needed to simulate crowds. Thus, depending on the complexity of the simulation, simple behaviors can be sufficient to simulate crowds. Otherwise, more complicated behavioral rules can be necessary and, in this case, it can be included in the simulation data in order to improve the realism of the animation. We present three different ways for controlling crowd behaviors: 1) by using innate and scripted behaviors, 2) by defining behavioral rules, using events and reactions, and 3) by providing an external control to guide crowd behaviors in real time. The two main contributions of our approach are: the possibility of increasing the complexity of group/agent behaviors according to the problem to be simulated and the hierarchical structure based on groups to compose a crowd.
IEEE Trans. Vis. Comput. Graph.	Interactive Direct Rendering of Trivariate B-Spline Scalar Functions.	Alon Raviv,Gershon Elber	2001	This paper presents a direct rendering paradigm of trivariate B-spline functions that is able to incrementally update complex volumetric data sets in the order of millions of coefficients at interactive rates of several frames per second on modern workstations. This incremental rendering scheme can hence be employed in modeling sessions of volumetric trivariate functions, offering interactive volumetric sculpting capabilities. The rendering is conducted from a fixed viewpoint and in two phases. The first, preprocessing, stage accumulates the effect that the coefficients of the trivariate function have on the pixels in the image. This preprocessing stage is conducted offline and only once per trivariate and viewing direction. The second stage conducts the actual rendering of the trivariate functions. As an example, during a volumetric sculpting operation, the artist can sculpt the volume and get a displayed feedback, in interactive rates.
IEEE Trans. Vis. Comput. Graph.	Volume Illustration: Nonphotorealistic Rendering of Volume Models.	Penny Rheingans,David S. Ebert	2001	Accurately and automatically conveying the structure of a volume model is a problem not fully solved by existing volume rendering approaches. Physics-based volume rendering approaches create images which may match the appearance of translucent materials in nature, but may not embody important structural details. Transfer function approaches allow flexible design of the volume appearance, but generally require substantial hand tuning for each new data set in order to be effective. We introduce the volume illustration approach, combining the familiarity of a physics-based illumination model with the ability to enhance important features using nonphotorealistic rendering techniques. Since features to be enhanced are defined on the basis of local volume characteristics rather than volume sample value, the application of volume illustration techniques requires less manual tuning than the design of a good transfer function. Volume illustration provides a flexible unified framework for enhancing structural perception of volume models through the amplification of features and the addition of illumination effects.
IEEE Trans. Vis. Comput. Graph.	"Guest Editor's Introduction: Special Issue on Visualization 2000."	Amitabh Varshney	2001	"Guest Editor's Introduction: Special Issue on Visualization 2000."
IEEE Trans. Vis. Comput. Graph.	Topology-Preserving Smoothing of Vector Fields.	Rüdiger Westermann,Christopher Johnson,Thomas Ertl	2001	In this paper, we propose a technique for topology-preserving smoothing of sampled vector fields. The vector field data is first converted into a scalar representation in which time surfaces implicitly exist as level-sets. We then locally analyze the dynamic behavior of level-sets by placing geometric primitives in the scalar field and by subsequently distorting these primitives with respect to local variations in this field. From the distorted primitives, we calculate the curvature normal and we use the normal magnitude and its direction to separate distinct flow features. Geometrical and topological considerations are then combined to successively smooth dense flow fields, at the same time retaining their topological structure.
IEEE Trans. Vis. Comput. Graph.	Detection and Visualization of Closed Streamlines in Planar Flows.	Thomas Wischgoll,Gerik Scheuermann	2001	The analysis and visualization of flows is a central problem in visualization. Topology-based methods have gained increasing interest in recent years. This article describes a method for the detection of closed streamlines in flows. It is based on a special treatment of cases where a streamline reenters a cell to prevent infinite cycling during streamline calculation. The algorithm checks for possible exits of a loop of crossed edges and detects structurally stable closed streamlines. These global features are not detected by conventional topology and feature detection algorithms.
IEEE Trans. Vis. Comput. Graph.	Extracting Objects from Range and Radiance Images.	Yizhou Yu,Andras Ferencz,Jitendra Malik	2001	In this paper, we present a pipeline and several key techniques necessary for editing a real scene captured with both cameras and laser range scanners. We develop automatic algorithms to segment the geometry from range images into distinct surfaces, register texture from radiance images with the geometry, and synthesize compact high-quality texture maps. The result is an object-level representation of the scene which can be rendered with modifications to structure via traditional rendering methods. The segmentation algorithm for geometry operates directly on the point cloud from multiple registered 3D range images instead of a reconstructed mesh. It is a top-down algorithm which recursively partitions a point set into two subsets using a pairwise similarity measure. The result is a binary tree with individual surfaces as leaves. Our image registration technique performs a very efficient search to automatically find the camera poses for arbitrary position and orientation relative to the geometry. Thus, we can take photographs from any location without precalibration between the scanner and the camera. The algorithms have been applied to large-scale real data. We demonstrate our ability to edit a captured scene by moving, inserting, and deleting objects.
IEEE Visualization	Case Study: A Virtual Environment for Genomic Data Visualization.	R. Mark Adams,Blaze Stancampiano,Michael McKenna,David Small	2002	With the completion of the human genome sequence, and with the proliferation of genome-related annotation data, the need for scalable and more intuitive means for analysis becomes critical. At Variagenics and Small Design Firm, we have addressed this problem with a coherent three-dimensional space in which all data can be seen in a single context. This tool aids in integrating information at vastly divergent scales while maintaining accurate spatial and size relationships. Our visualization was successful in communicating to project teams with diverse backgrounds the magnitude and biological implication of genetic variation.
IEEE Visualization	Sea of Images.	Daniel G. Aliaga,Thomas A. Funkhouser,Dimah Yanovsky,Ingrid Carlbom	2002	"A long-standing research problem in computer graphics is to reproduce the visual experience of walking through a large photorealistic environment interactively. On one hand, traditional geometry-based rendering systems fall short of simulating the visual realism of a complex environment. On the other hand, image-based rendering systems have to date been unable to capture and store a sampled representation of a large environment with complex lighting and visibility effects.In this paper, we present a ""Sea of Images,"" a practical approach to dense sampling, storage, and reconstruction of the plenoptic function in large, complex indoor environments. We use a motorized cart to capture omnidirectional images every few inches on a eye-height plane throughout an environment. The captured images are compressed and stored in a multiresolution hierarchy suitable for real-time prefetching during an interactive walkthrough. Later, novel images are reconstructed for a simulated observer by resampling nearby captured images.Our system acquires 15,254 images over 1,050 square feet at an average image spacing of 1.5 inches. The average capture and processing time is 7 hours. We demonstrate realistic walkthroughs of real-world environments reproducing specular reflections and occlusion effects while rendering 15-25 frames per second."
IEEE Visualization	A Case Study in Selective Visualization of Unsteady 3D Flow.	Dirk Bauer,Ronald Peikert,Mie Sato,Mirjam Sick	2002	In this case study, we explore techniques for the purpose of visualizing isolated flow structures in time-dependent data. Our primary industrial application is the visualization of the vortex rope, a rotating helical structure which builds up in the draft tube of a water turbine. The vortex rope can be characterized by high values of normalized helicity, which is a scalar field derived from the given CFD velocity data. In two related applications, the goal is to visualize the cavitation regions near the runner blades of a Kaplan turbine and a water pump, respectively. Again, the flow structure of interest can be defined by a scalar field, namely by low pressure values. We propose a particle seeding scheme based on quasi-random numbers, which minimizes visual artifacts such as clusters or patterns. By constraining the visualization to a region of interest, occlusion problems are reduced and storage efficiency is gained.
IEEE Visualization	GeneVis: Visualization Tools for Genetic Regulatory Network Dynamics.	Charles A. H. Baker,M. Sheelagh T. Carpendale,Przemyslaw Prusinkiewicz,Michael G. Surette	2002	GeneVis provides a visual environment for exploring the dynamics of genetic regulatory networks. At present time, genetic regulation is the focus of intensive research worldwide, and computational aids are being called for to help in the research of factors that are difficult to observe directly. GeneVis provides a particle-based simulation of genetic networks and visualizes the process of this simulation as it occurs. Two dynamic visualization techniques are provided, a visualization of the movement of the regulatory proteins and a visualization of the relative concentrations of these proteins. Several interactive tools relate the dynamic visualizations to the underlying genetic network structure.
IEEE Visualization	Volume Warping for Adaptive Isosurface Extraction.	Laurent Balmelli,Christopher J. Morris,Gabriel Taubin,Fausto Bernardini	2002	Polygonal approximations of isosurfaces extracted from uniformly sampled volumes are increasing in size due to the availability of higher resolution imaging techniques. The large number of primitives represented hinders the interactive exploration of the dataset. Though many solutions have been proposed to this problem, many require the creation of isosurfaces at multiple resolutions or the use of additional data structures, often hierarchical, to represent the volume.We propose a technique for adaptive isosurface extraction that is easy to implement and allows the user to decide the degree of adaptivity as well as the choice of isosurface extraction algorithm. Our method optimizes the extraction of the isosurface by warping the volume. In a warped volume, areas of importance (e.g. containing significant details) are inflated while unimportant ones are contracted. Once the volume is warped, any extraction algorithm can be applied. The extracted mesh is subsequently unwarped such that the warped areas are rescaled to their initial proportions. The resulting isosurface is represented by a mesh that is more densely sampled in regions decided as important.
IEEE Visualization	Interactive Spectral Volume Rendering.	Steven Bergner,Torsten Möller,Mark S. Drew,Graham D. Finlayson	2002	We describe a method for volume rendering using a spectral representation of colour instead of the traditional RGB model. It is shown how to use this framework for a novel exploration of datasets through enhanced transfer function design. Furthermore, our framework is extended to allow real-time re-lighting of the scene created with any rendering method. The technique of post-illumination is introduced to generate new spectral images for arbitrary light colours in real-time. Also a tool is described to design a palette of lights and materials having certain properties such as selective metamerism or colour constancy. Applied to spectral transfer functions, different light colours can accentuate or hide specific qualities of the data. In connection with post-illumination this provides a new degree of freedom for guided exploration of volumetric data, which cannot be achieved using the RGB model.
IEEE Visualization	Scalable Alignment of Large-Format Multi-Projector Displays Using Camera Homography Trees.	Han Chen,Rahul Sukthankar,Grant Wallace,Kai Li	2002	This paper presents a vision-based geometric alignment system for aligning the projectors in an arbitrarily large display wall. Existing algorithms typically rely on a single camera view and degrade in accuracy1 as the display resolution exceeds the camera resolution by several orders of magnitude. Na&iuml;ve approaches to integrating multiple zoomed camera views fail since small errors in aligning adjacent views propagate quickly over the display surface to create glaring discontinuities. Our algorithm builds and refines a camera homography2 tree to automatically register any number of uncalibrated camera images; the resulting system is both faster and significantly more accurate than competing approaches, reliably achieving alignment errors of 0.55 pixels on a 24-projector display in under 9 minutes. Detailed experiments compare our system to two recent display wall alignment algorithms, both on our 18 Megapixel display wall and in simulation. These results indicate that our approach achieves sub-pixel accuracy even on displays with hundreds of projectors.
IEEE Visualization	TetFusion: An Algorithm For Rapid Tetrahedral Mesh Simplification.	Prashant Chopra,Joerg Meyer	2002	"This paper introduces an algorithm for rapid progressive simplification of tetrahedral meshes: TetFusion. We describe how a simple geometry decimation operation steers a rapid and controlled progressive simplification of tetrahedral meshes, while also taking care of complex mesh-inconsistency problems. The algorithm features a high decimation ratio per step, and inherently discourages any cases of self-intersection of boundary, element-boundary intersection at concave boundary-regions, and negative volume tetrahedra (flipping). We achieved rigorous reduction ratios of up to 98% for meshes consisting of 827,904 elements in less than 2 minutes, progressing through a series of level-of-details (LoDs) of the mesh in a controlled manner. We describe how the approach supports a balanced re-distribution of space between tetrahedral elements, and explain some useful control parameters that make it faster and more intuitive than 'edge collapse'-based decimation methods for volumetric meshes [3, 19, 21, 22]. Finally, we discuss how this approach can be employed for rapid LoD prototyping of large time-varying datasets as an aid to interactive visualization."
IEEE Visualization	Immersive Volume Visualization of Seismic Simulations: A Case Study of Techniques Invented and Lessons Learned.	Prashant Chopra,Joerg Meyer,Antonio Fernandez	2002	This paper is a documentation of techniques invented, results obtained and lessons learned while creating visualization algorithms to render outputs of large-scale seismic simulations. The objective is the development of techniques for a collaborative simulation and visualization shared between structural engineers, seismologists, and computer scientists. The computer graphics research community has been witnessing a large number of exemplary publications addressing the challenges faced while trying to visualize both large-scale surface and volumetric datasets lately. From a visualization perspective, issues like data preprocessing (simplification, sampling, filtering, etc.); rendering algorithms (surface and volume), and interaction paradigms (large-scale, highly interactive, highly immersive, etc.) have been areas of study. In this light, we outline and describe the milestones achieved in a large-scale simulation and visualization project, which opened the scope for combining existing techniques with new methods, especially in those cases where no existing methods were suitable. We elucidate the data simplification and reorganization schemes that we used, and discuss the problems we encountered and the solutions we found. We describe both desktop (high-end local as well as remote) interfaces and immersive visualization systems that we developed to employ interactive surface and volume rendering algorithms. Finally, we describe the results obtained, challenges that still need to be addressed, and ongoing efforts to meet the challenges of large-scale visualization.
IEEE Visualization	Optimized View-Dependent Rendering for Large Polygonal Datasets.	Jihad El-Sana,Eitan Bachmant	2002	In this paper we are presenting a novel approach for rendering large datasets in a view-dependent manner. In a typical view-dependent rendering framework, an appropriate level of detail is selected and sent to the graphics hardware for rendering at each frame. In our approach, we have successfully managed to speed up the selection of the level of detail as well as the rendering of the selected levels. We have accelerated the selection of the appropriate level of detail by not scanning active nodes that do not contribute to the incremental update of the selected level of detail. Our idea is based on imposing a spatial subdivision over the view-dependence trees data-structure, which allows spatial tree cells to refine and merge in real-time rendering to comply with the changes in the active nodes list. The rendering of the selected level of detail is accelerated by using vertex arrays. To overcome the dynamic changes in the selected levels of detail we use multiple small vertex arrays whose sizes depend on the memory on the graphics hardware. These multiple vertex arrays are attached to the active cells of the spatial tree and represent the active nodes of these cells. These vertex arrays, which are sent to the graphics hardware at each frame, merge and split with respect to the changes in the cells of the spatial tree.
IEEE Visualization	Case Study: Visual Debugging of Finite Element Codes.	Patricia Crossno,David H. Rogers,Christopher J. Garasi	2002	We present an innovative application developed at Sandia National Laboratories for visual debugging of unstructured finite element physics codes. Our tool automatically locates anomalous regions, such as inverted elements or nodes whose variable values lie outside a prescribed range, then extracts mesh subsets around these features for detailed examination. The subsets are viewed using color coding of variable values superimposed on the mesh structure. This allows the values and their relative spatial locations within the mesh to be correlated at a glance. Both topological irregularities and hot spots within the data stand out visually, allowing the user to explore the exact numeric values of the grid at surrounding points over time. We demonstrate the utility of this approach by debugging a cell inversion in a simulation of an exploding wire.
IEEE Visualization	Case Study: Hardware-Accelerated Selective LIC Volume Rendering.	Gordon Erlebacher,David A. Yuen,Fabien Dubuffet	2002	Case Study: Hardware-Accelerated Selective LIC Volume Rendering.
IEEE Visualization	XFastMesh: Fast View-dependent Meshing from External Memory.	Christopher DeCoro,Renato Pajarola	2002	We present a novel disk-based multiresolution triangle mesh data structure that supports paging and view-dependent rendering of very large meshes at interactive frame rates from external memory. Our approach, called XFastMesh, is based on a view-dependent mesh simplification framework that represents half-edge collapse operations in a binary hierarchy known as a merge-tree forest. The proposed technique partitions the merge-tree forest into so-called detail blocks, which consist of binary subtrees, that are stored on disk. We present an efficient external memory data structure and file format that stores all detail information of the multiresolution triangulation method using significantly less storage then previously reported approaches. Furthermore, we present a paging algorithm that provides efficient loading and interactive rendering of large meshes from external memory at varying and view-dependent level-of-detail. The presented approach is highly efficient both in terms of space cost and paging performance.
IEEE Visualization	Assisted Navigation of Complex Information Spaces.	Brent M. Dennis,Christopher G. Healey	2002	Assisted Navigation of Complex Information Spaces.
IEEE Visualization	Interactive Visualization of Complex Plant Ecosystems.	Oliver Deussen,Carsten Colditz,Marc Stamminger,George Drettakis	2002	We present a method for interactive rendering of large outdoor scenes. Complex polygonal plant models and whole plant populations are represented by relatively small sets of point and line primitives. This enables us to show landscapes faithfully using only a limited percentage of primitives. In addition, a hierarchical data structure allows us to smoothly reduce the geometrical representation to any desired number of primitives. The scene is hierarchically divided into local portions of geometry to achieve large reduction factors for distant regions. Additionally, the data reduction is adapted to the visual importance of geometric objects. This allows us to maintain the visual fidelity of the representation while reducing most of the geometry drastically. With our system, we are able to interactively render very complex landscapes with good visual quality.
IEEE Visualization	PMR: Point to Mesh Rendering, A Feature-Based Approach.	Tamal K. Dey,James Hudson	2002	Within the field of computer graphics and visualization, it is often necessary to visualize polygonal models with large number of polygons. Display quality is mandatory, but it is also desirable to have the ability to rapidly update the display in order to facilitate interactive use. Point based rendering methods have been shown effective for this task. Building on this paradigm we introduce the PMR system which uses a hierarchy both in points and triangles for rendering. This hierarchy is fundamentally different from the ones used in existing methods. It is based on the feature geometry in the object space rather than its projection in the screen space. This provides certain advantages over the existing methods.
IEEE Visualization	A Multiphase Approach to Efficient Surface Simplification.	Michael Garland,Eric Shaffer	2002	We present a new multiphase method for efficiently simplifying polygonal surface models of arbitrary size. It operates by combining an initial out-of-core uniform clustering phase with a subsequent in-core iterative edge contraction phase. These two phases are both driven by quadric error metrics, and quadrics are used to pass information about the original surface between phases. The result is a method that produces approximations of a quality comparable to quadric-based iterative edge contraction, but at a fraction of the cost in terms of running time and memory consumption.
IEEE Visualization	A Case Study on Multiresolution Visualization of Local Rainfall from Weather Radar Measurements.	Thomas Gerstner,Dirk Meetschen,Susanne Crewell,Michael Griebel,Clemens Simmer	2002	Weather radars can measure the backscatter from rain drops in the atmosphere. A complete radar scan provides three-dimensional precipitation information. For the understanding of the underlying atmospheric processes interactive visualization of these data sets is necessary. This is a challenging task due to the size, structure and required context of the data. In this case study, a multiresolution approach for real-time simultaneous visualization of radar measurements together with the corresponding terrain data is illustrated.
IEEE Visualization	Case Study: Visualizing Ocean Flow Vertical Motions using Lagrangian-Eulerian Time Surfaces.	"Josh Grant,Gordon Erlebacher,James F. O'Brien"	2002	Ocean model simulations commonly assume the ocean is hydrostatic, resulting in near zero vertical motion. The vertical motion found is typically associated with the variations of the thermocline depth over time, which are mainly a result of the development and movement of ocean fronts, eddies, and internal waves. A new technique, extended from Lagrangian-Eulerian Advection, is presented to help understand the variation of vertical motion associated with the change in thermocline depth over time. A time surface is correctly deformed in a single direction according to the flow. The evolution of the time surface is computed via a mixture of Eulerian and Lagrangian techniques. The dominant horizontal motion is textured onto the surface using texture advection, while both the horizontal and vertical motions are used to displace the surface. The resulting surface is shaded for enhanced contrast. Timings indicate that the overhead over standard 2D texture advection is no more than 12%.
IEEE Visualization	Interactive View-Dependent Rendering of Large IsoSurfaces.	Benjamin F. Gregorski,Mark A. Duchaineau,Peter Lindstrom,Valerio Pascucci,Kenneth I. Joy	2002	We present an algorithm for interactively extracting and rendering isosurfaces of large volume datasets in a view-dependent fashion. A recursive tetrahedral mesh refinement scheme, based on longest edge bisection, is used to hierarchically decompose the data into a multiresolution structure. This data structure allows fast extraction of arbitrary isosurfaces to within user specified view-dependent error bounds. A data layout scheme based on hierarchical space filling curves provides access to the data in a cache coherent manner that follows the data access pattern indicated by the mesh refinement.
IEEE Visualization	PRIMA: A Case Study of Using Information Visualization Techniques for Patient Record Analysis.	Donna L. Gresh,David A. Rabenhorst,Amnon Shabo,Shimon Slavin	2002	We have created an application, called PRIMA (Patient Record Intelligent Monitoring and Analysis), which can be used to visualize and understand patient record data. It was developed to better understand a large collection of patient records of bone marrow transplants at Hadassah Hospital in Jerusalem, Israel. It is based on an information visualization toolkit, Opal, which has been developed at the IBM T.J. Watson Research Center. Opal allows intelligent, interactive visualization of a wide variety of different types of data. The PRIMA application is generally applicable to a wide range of patient record data, as the underlying toolkit is flexible with regard to the form of the input data. This application is a good example of the usefulness of information visualization techniques in the bioinformatics domain, as these techniques have been developed specifically to deal with diverse sets of often unfamiliar data. We illustrate several unanticipated findings which resulted from the use of a flexible and interactive information visualization environment.
IEEE Visualization	Probabilistic Surfaces: Point Based Primitives to Show Surface Uncertainty.	Gevorg Grigoryan,Penny Rheingans	2002	Efficient and informative visualization of surfaces with uncertainties is an important topic with many applications in science and engineering. Examples include environmental pollution borderline identification, identification of the limits of an oil basin, or discrimination between contaminated and healthy tissue in medicine. This paper presents an approach for such visualization using points as display primitives. Our approach is to render each polygon as a collection of points and to displace each point from the surface in the direction of the surface normal by an amount proportional to some random number multiplied by the uncertainty level at that point. This approach can be used in combination with other techniques such as pseudo-coloring and shading to give rise to efficient and revealing visualizations. The method is used to visualize real and simulated tumor formations with uncertainty of tumor boundaries.
IEEE Visualization	Maximum Entropy Light Source Placement.	Stefan Gumhold	2002	Several methods have been proposed to help the user to position lighting sources for a given view of a 3d-scene. In the first kind of methods the user defines a desired illumination through highlight and shadow locations for which the lighting system optimizes the light positions. The more automatic approaches are based on perceptual image metrics. [Marks et al. 1997] define an image metric that measures how different two images are perceived. A collection of maximally different images is presented to the user for selection. [Shacked and Lischinski 2001] define a perceptual based image quality metric composed of six contributing terms, for which the user has to specify weights before the system searches for a locally optimal light source placement.
IEEE Visualization	Interactive Rendering of Large Volume Data Sets.	Stefan Guthe,Michael Wand,Julius Gonser,Wolfgang Straßer	2002	We present a new algorithm for rendering very large volume data sets at interactive framerates on standard PC hardware. The algorithm accepts scalar data sampled on a regular grid as input. The input data is converted into a compressed hierarchical wavelet representation in a preprocessing step. During rendering, the wavelet representation is decompressed on-the-fly and rendered using hardware texture mapping. The level of detail used for rendering is adapted to the local frequency spectrum of the data and its position relative to the viewer. Using a prototype implementation of the algorithm we were able to perform an interactive walkthrough of large data sets such as the visible human on a single of-the-shelf PC.
IEEE Visualization	Isometric Embedding by Surface Reconstruction from Distances.	Ingrid Hotz	2002	To display the intuitive meaning of an abstract metric it is helpful to look on an embedded surface with the same inner geometry as the given metric. The resulting partial differential equations have no standard solution. Only for some special cases satisfactory methods are known. I present a new algorithmic approach which is not based on differential equations. In contrast to other methods this technique also works if the embedding exists only locally. The fundamental idea is to estimate Euclidean distances, from which the surface is built up. In this paper I focus on the reconstruction of a surface from these estimated distances. Particular the influence of a perturbation of the distances on the shape of the resulting surface is investigated.
IEEE Visualization	Case Study: A Look of Performance Expression.	Rumi Hiraga	2002	For most of the time, we enjoy and appreciate music performances as they are. Once we try to understand the performance not in subjective terms but in an objective way and share it with other people, visualizing the performance parameters is indispensable. In this paper, a figure for visualizing performance expressions is described. This figure helps people understand the cause and position of the performance expression as it has expressive cues, which coincide with the cognitive meaning of musical performance, and not by using only MIDI parameter values. The differences we hear between performances are clarified by visualized figures.
IEEE Visualization	Christmas Tree Case Study: Computed Tomography as a Tool for Mastering Complex Real World Objects with Applications in Computer Graphics.	Armin Kanitsar,Thomas Theußl,Lukas Mroz,Milos Srámek,Anna Vilanova Bartrolí,Balázs Csébfalvi,Jirí Hladuvka,Dominik Fleischmann,Michael Knapp,Rainer Wegenkittl,Petr Felkel,Stefan Guthe,Werner Purgathofer,Eduard Gröller	2002	Christmas Tree Case Study: Computed Tomography as a Tool for Mastering Complex Real World Objects with Applications in Computer Graphics.
IEEE Visualization	Compressing Polygon Mesh Geometry with Parallelogram Prediction.	Martin Isenburg,Pierre Alliez	2002	In this paper we present a generalization of the geometry coder by Touma and Gotsman [34] to polygon meshes. We let the polygon information dictate where to apply the parallelogram rule that they use to predict vertex positions. Since polygons tend to be fairly planar and fairly convex, it is beneficial to make predictions within a polygon rather than across polygons. This, for example, avoids poor predictions due to a crease angle between polygons. Up to 90 percent of the vertices can be predicted this way. Our strategy improves geometry compression by 10 to 40 percent depending on (a) how polygonal the mesh is and (b) on the quality (planarity/convexity) of the polygons.
IEEE Visualization	A Model for the Visualization Exploration Process.	T. J. Jankun-Kelly,Kwan-Liu Ma,Michael Gertz	2002	The current state of the art in visualization research places a strong emphasis on different techniques to derive insight from disparate types of data. However, little work has investigated the visualization process itself. The information content of the visualization process---the results, history, and relationships between those results---is addressed by this work. A characterization of the visualization process is discussed, leading to a general model of the visualization exploration process. The model, based upon a new parameter derivation calculus, can be used for automated reporting, analysis, or visualized directly. An XML-based language for expressing visualization sessions using the model is also described. These sessions can then be shared and reused by collaborators. The model, along with the XML representation, provides an effective means to utilize the information within the visualization process to further data exploration.
IEEE Visualization	A Radial Focus+Context Visualization for Multi-Dimensional Functions.	Sanjini Jayaraman,Chris North	2002	"The analysis of multidimensional functions is important in many engineering disciplines, and poses a major problem as the number of dimensions increases. Previous visualization approaches focus on representing three or fewer dimensions at a time. This paper presents a new focus+context visualization that provides an integrated overview of an entire multidimensional function space, with uniform treatment of all dimensions. The overview is displayed with respect to a user-controlled polar focal point in the function's parameter space. Function value patterns are viewed along rays that emanate from the focal point in all directions in the parameter space, and represented radially around the focal point in the visualization. Data near the focal point receives proportionally more screen space than distant data. This approach scales smoothly from two dimensions to 10-20, with a 1000 pixel range on each dimension."
IEEE Visualization	Geometric Verification of Features in Flow Fields.	Ming Jiang,Raghu Machiraju,David Thompson	2002	Geometric Verification of Features in Flow Fields.
IEEE Visualization	Rendering The First Star In The Universe - A Case Study.	Ralf Kähler,Donna Cox,Robert Patterson,Stuart Levy,Hans-Christian Hege,Tom Abel	2002	"For quantitative examination of phenomena that simultaneously occur on very different spatial and temporal scales, adaptive hierarchical schemes are required. A special numerical multilevel technique, associated with a particular hierarchical data structure, is so-called Adaptive Mesh Refinement (AMR). It allows one to bridge a wide range of spatial and temporal resolutions and therefore gains increasing popularity.We describe the interplay of several visualization and VR software packages for rendering time dependent AMR simulations of the evolution of the first star in the universe. The work was done in the framework of a television production for Discovery Channel Television, ""The Unfolding Universe."". Parts of the data were taken from one of the most complex AMR simulation ever carried out: It contained up to 27 levels of resolution, requiring modifications to the texture based AMR volume rendering algorithm that was used to depict the density distribution of the gaseous interstellar matter. A voice and gesture controlled CAVE application was utilized to define camera paths following the interesting features deep inside the computational domains. Background images created from cosmological computational data were combined with the final renderings."
IEEE Visualization	CPR - Curved Planar Reformation.	Armin Kanitsar,Dominik Fleischmann,Rainer Wegenkittl,Petr Felkel,Eduard Gröller	2002	Visualization of tubular structures such as blood vessels is an important topic in medical imaging. One way to display tubular structures for diagnostic purposes is to generate longitudinal cross-sections in order to show their lumen, wall, and surrounding tissue in a curved plane. This process is called Curved Planar Reformation (CPR). We present three different methods to generate CPR images. A tube-phantom was scanned with Computed Tomography (CT) to illustrate the properties of the different CPR methods. Furthermore we introduce enhancements to these methods: thick-CPR, rotating-CPR and multi-path-CPR.
IEEE Visualization	Efficient Compression and Rendering of Multi-Resolution Meshes.	Zachi Karni,Alexander Bogomjakov,Craig Gotsman	2002	We present a method to code the multiresolution structure of a 3D triangle mesh in a manner that allows progressive decoding and efficient rendering at a client machine. The code is based on a special ordering of the mesh vertices which has good locality and continuity properties, inducing a natural multiresolution structure. This ordering also incorporates information allowing efficient rendering of the mesh at all resolutions using the contemporary vertex buffer mechanism. The performance of our code is shown to be competitive with existing progressive mesh compression methods, while achieving superior rendering speed.
IEEE Visualization	Evaluation of a Multimodal Interface for 3D Terrain Visualization.	David M. Krum,Olugbenga Omoteso,William Ribarsky,Thad Starner,Larry F. Hodges	2002	"Novel speech and/or gesture interfaces are candidates for use in future mobile or ubiquitous applications. This paper describes an evaluation of various interfaces for visual navigation of a whole Earth 3D terrain model. A mouse driven interface, a speech interface, a gesture interface, and a multimodal speech and gesture interface were used to navigate to targets placed at various points on the Earth. This study measured each participant's recall of target identity, order, and location as a measure of cognitive load. Timing information as well as a variety of subjective measures including discomfort and user preference were taken. While the familiar and mature mouse interface scored best by most measures, the speech interface also performed well. The gesture and multimodal interface suffered from weaknesses in the gesture modality. Weaknesses in the speech and multimodal modalities are identified and areas for improvement are discussed."
IEEE Visualization	Face-based Luminance Matching for Perceptual Colormap Generation.	Gordon L. Kindlmann,Erik Reinhard,Sarah Creem	2002	"Most systems used for creating and displaying colormap-based visualizations are not photometrically calibrated. That is, the relationship between RGB input levels and perceived luminance is usually not known, due to variations in the monitor, hardware configuration, and the viewing environment. However, the luminance component of perceptually based colormaps should be controlled, due to the central role that luminance plays in our visual processing. We address this problem with a simple and effective method for performing luminance matching on an uncalibrated monitor. The method is akin to the minimally distinct border technique (a previous method of luminance matching used for measuring luminous efficiency), but our method relies on the brain's highly developed ability to distinguish human faces. We present a user study showing that our method produces equivalent results to the minimally distinct border technique, but with significantly improved precision. We demonstrate how results from our luminance matching method can be directly applied to create new univariate colormaps."
IEEE Visualization	Interactive Translucent Volume Rendering and Procedural Modeling.	Joe Kniss,Simon Premoze,Charles D. Hansen,David S. Ebert	2002	Direct volume rendering is a commonly used technique in visualization applications. Many of these applications require sophisticated shading models to capture subtle lighting effects and characteristics of volumetric data and materials. Many common objects and natural phenomena exhibit visual quality that cannot be captured using simple lighting models or cannot be solved at interactive rates using more sophisticated methods. We present a simple yet effective interactive shading model which captures volumetric light attenuation effects to produce volumetric shadows and the subtle appearance of translucency. We also present a technique for volume displacement or perturbation that allows realistic interactive modeling of high frequency detail for real and synthetic volumetric data.
IEEE Visualization	Fast View-Dependent Level-of-Detail Rendering Using Cached Geometry.	Joshua Levenberg	2002	Level-of-detail rendering is essential for rendering very large, detailed worlds in real-time. Unfortunately, level-of-detail computations can be expensive, creating a bottleneck at the CPU.This paper presents the CABTT algorithm, an extension to existing binary-triangle-tree-based level-of-detail algorithms. Instead of manipulating triangles, the CABTT algorithm instead operates on clusters of geometry called aggregate triangles. This reduces CPU overhead, eliminating a bottleneck common to level-of-detail algorithms. Since aggregate triangles stay fixed over several frames, they may be cached on the video card. This further reduces CPU load and fully utilizes the hardware accelerated rendering pipeline on modern video cards. These improvements result in a fourfold increase in frame rate over ROAM [7] at high detail levels. Our implementation renders an approximation of an 8 million triangle heightfield at 42 frames per second with an maximum error of 1 pixel on consumer hardware.
IEEE Visualization	Horizon Occlusion Culling for Real-time Rendering of Hierarchical Terrains.	Brandon Lloyd,Parris K. Egbert	2002	We present a technique to perform occlusion culling for hierarchical terrains at run-time. The algorithm is simple to implement and requires minimal pre-processing and additional storage, yet leads to 2-4 times improvement in framerate for views with high degrees of occlusion. Our method is based on the well-known occlusion horizon algorithm. We show how to adapt the algorithm for use with hierarchical terrains. The occlusion horizon is constructed as the terrain is traversed in an approximate front to back ordering. Regions of the terrain are compared to the horizon to determine when they are completely occluded from the viewpoint. Culling these regions leads to significant savings in rendering.
IEEE Visualization	Exploring Surface Characteristics with Interactive Gaussian Images (A Case Study).	Bradley C. Lowekamp,Penny Rheingans,Terry S. Yoo	2002	The Gauss map projects surface normals to a unit sphere, providing a powerful visualization of the geometry of a graphical object. It can be used to predict visual events caused by changes in lighting, shading, and camera control. We present an interactive technique for portraying the Gauss map of polygonal models, mapping surface normals and the magnitudes of surface curvature using a spherical projection. Unlike other visualizations of surface curvature, we create our Gauss map directly from polygonal meshes without requiring any complex intermediate calculations of differential geometry. For anything other than simple shapes, surface information is densely mapped into the Gaussian normal image, inviting the use of visualization techniques to amplify and emphasize details hidden within the wealth of data. We present the use of interactive visualization tools such as brushing and linking to explore the surface properties of solid shapes. The Gauss map is shown to be simple to compute, easy to view dynamically, and effective at portraying important features of polygonal models.
IEEE Visualization	Non-Photorealistic Volume Rendering Using Stippling Techniques.	Aidong Lu,Christopher J. Morris,David S. Ebert,Penny Rheingans,Charles D. Hansen	2002	Simulating hand-drawn illustration techniques can succinctly express information in a manner that is communicative and informative. We present a framework for an interactive direct volume illustration system that simulates traditional stipple drawing. By combining the principles of artistic and scientific illustration, we explore several feature enhancement techniques to create effective, interactive visualizations of scientific and medical datasets. We also introduce a rendering mechanism that generates appropriate point lists at all resolutions during an automatic preprocess, and modifies rendering styles through different combinations of these feature enhancements. The new system is an effective way to interactively preview large, complex volume datasets in a concise, meaningful, and illustrative manner. Volume stippling is effective for many applications and provides a quick and efficient method to investigate volume models.
IEEE Visualization	Semotus Visum: A Flexible Remote Visualization Framework.	Eric Luke,Charles D. Hansen	2002	By offering more detail and precision, large data sets can provide greater insights to researchers than small data sets. However, these data sets require greater computing resources to view and manage. Remote visualization techniques allow the use of computers that cannot be operated locally. The Semotus Visum framework applies a high-performance client-server paradigm to the problem. The framework utilizes both client and server resources via multiple rendering methods. Experimental results show the framework delivers high framerates and low latency across a wide range of data sets.
IEEE Visualization	Kinetic Visualization - A Technique for Illustrating 3D Shape and Structure.	Eric B. Lum,Aleksander Stompel,Kwan-Liu Ma	2002	Motion provides strong visual cues for the perception of shape and depth, as demonstrated by cognitive scientists and visual artists. This paper presents a novel visualization technique --- kinetic visualization --- that uses particle systems to add supplemental motion cues which can aid in the perception of shape and spatial relationships of static objects. Based on a set of rules following perceptual and physical principles, particles flowing over the surface of an object not only bring out, but also attract attention to, essential information on the shape of the object that might not be readily visible with conventional rendering that uses lighting and view changes. Replacing still images with animations in this fashion, we demonstrate with both surface and volumetric models in the accompanying videos that in many cases the resulting visualizations effectively enhance the perception of three-dimensional shape and structure. The results of a preliminary user study that we have conducted also show evidence that the supplemental motion cues help.
IEEE Visualization	Computing Singularities of 3D Vector Fields with Geometric Algebra .	Stephen Mann,Alyn P. Rockwood	2002	Critical points of a vector field are key to their characterization. Not only their positions but also their indexes are crucial for understanding vector fields. Considerable work exists in 2D, but less is available for 3D or higher dimensions. Geometric Algebra is a derivative of Clifford Algebra that not only enables a succinct definition of the index of a critical point in higher dimension; it also provides insight and computational pathways for calculating the index. We describe the problems in terms of Geometric Algebra and present an octree based solution using the algebra for finding critical points and their index in a 3D vector field.
IEEE Visualization	Level-Set Segmentation From Multiple Non-Uniform Volume Datasets.	Ken Museth,David E. Breen,Leonid Zhukov,Ross T. Whitaker	2002	Typically 3-D MR and CT scans have a relatively high resolution in the scanning X - Y plane, but much lower resolution in the axial Z direction. This non-uniform sampling of an object can miss small or thin structures. One way to address this problem is to scan the same object from multiple directions. In this paper we describe a method for deforming a level set model using velocity information derived from multiple volume datasets with non-uniform resolution in order to produce a single high-resolution 3D model. The method locally approximates the values of the multiple datasets by fitting a distance-weighted polynomial using moving least-squares. The proposed method has several advantageous properties: its computational cost is proportional to the object surface area, it is stable with respect to noise, imperfect registrations and abrupt changes in the data, it provides gain-correction, and it employs a distance-based weighting to ensures that the contributions from each scan are properly merged into the final result. We have demonstrated the effectiveness of our approach on four multi-scan datasets, a griffin laser scan reconstruction, a CT scan of a teapot and MR scans of a mouse embryo and a zucchini.
IEEE Visualization	Approximating Normals for Marching Cubes applied to Locally Supported Isosurfaces.	Gregory M. Nielson,Adam Huang,Steve Sylvester	2002	We present some new methods for computing estimates of normal vectors at the vertices of a triangular mesh surface approximation to an isosurface which has been computed by the marching cube algorithm. These estimates are required for the smooth rendering of triangular mesh surfaces. The conventional method of computing estimates based upon divided difference approximations of the gradient can lead to poor estimates in some applications. This is particularly true for isosurfaces obtained from a field function, which is defined only for values near to the isosurface. We describe some efficient methods for computing the topology of the triangular mesh surface, which is used for obtaining local estimates of the normals. In addition, a new, one pass, approach for these types of applications is described and compared to existing methods.
IEEE Visualization	A New Object-Order Ray-Casting Algorithm.	Benjamin Mora,Jean-Pierre Jessel,René Caubet	2002	Many direct volume rendering algorithms have been proposed during the last decade to render 2563 voxels interactively. However a lot of limitations are inherent to all of them, like low-quality images, a small viewport size or a fixed classification. In contrast, interactive high quality algorithms are still a challenge nowadays. We introduce here an efficient and accurate technique called object-order ray-casting that can achieve up to 10 fps on current workstations. Like usual ray-casting, colors and opacities are evenly sampled along the ray, but now within a new object-order algorithm. Thus, it allows to combine the main advantages of both worlds in term of speed and quality. We also describe an efficient hidden volume removal technique to compensate for the loss of early ray termination.
IEEE Visualization	QuadTIN: Quadtree based Triangulated Irregular Networks.	Renato Pajarola,Marc Antonijuan,Roberto Lario	2002	Interactive visualization of large digital elevation models is of continuing interest in scientific visualization, GIS, and virtual reality applications. Taking advantage of the regular structure of grid digital elevation models, efficient hierarchical multiresolution triangulation and adaptive level-of-detail (LOD) rendering algorithms have been developed for interactive terrain visualization. Despite the higher triangle count, these approaches generally outperform mesh simplification methods that produce irregular triangulated network (TIN) based LOD representations. In this project we combine the advantage of a TIN based mesh simplification preprocess with high-performance quadtree based LOD triangulation and rendering at run-time. This approach, called QuadTIN, generates an efficient quadtree triangulation hierarchy over any irregular point set that may originate from irregular terrain sampling or from reducing oversampling in high-resolution grid digital elevation models.
IEEE Visualization	Case Study: Interactive Rendering of Adaptive Mesh Refinement Data.	Sanghun Park,Chandrajit L. Bajaj,Vinay Siddavanahalli	2002	Adaptive mesh refinement (AMR) is a popular computational simulation technique used in various scientific and engineering fields. Although AMR data is organized in a hierarchical multi-resolution data structure, the traditional volume visualization algorithms such as ray-casting and splatting cannot handle the form without converting it to a sophisticated data structure. In this paper, we present a hierarchical multi-resolution splatting technique using k-d trees and octrees for AMR data that is suitable for implementation on the latest consumer PC graphics hardware. We describe a graphical user interface to set transfer function and viewing / rendering parameters interactively. Experimental results obtained on a general purpose PC equipped with NVIDIA GeForce card are presented to demonstrate that the technique can interactively render AMR data (over 20 frames per second). Our scheme can easily be applied to parallel rendering of time-varying AMR data.
IEEE Visualization	Efficient Computation of the Topology of Level Set.	Valerio Pascucci,Kree Cole-McLaughlin	2002	This paper introduces two efficient algorithms that compute the Contour Tree of a 3D scalar field F and its augmented version with the Betti numbers of each isosurface. The Contour Tree is a fundamental data structure in scientific visualization that is used to preprocess the domain mesh to allow optimal computation of isosurfaces with minimal overhead storage. The Contour Tree can also be used to build user interfaces reporting the complete topological characterization of a scalar field, as shown in Figure 1.The first part of the paper presents a new scheme that augments the Contour Tree with the Betti numbers of each isocontour in linear time. We show how to extend the scheme introduced in [3] with the Betti number computation without increasing its complexity. Thus, we improve on the time complexity from our previous approach [10] from O(m log m) to O(n log n + m), where m is the number of tetrahedra and n is the number of vertices in the domain of F.The second part of the paper introduces a new divide-and-conquer algorithm that computes the Augmented Contour Tree with improved efficiency. The central part of the scheme computes the output Contour Tree by merging two intermediate Contour Trees and is independent of the interpolant. In this way we confine any knowledge regarding a specific interpolant to an oracle that computes the tree for a single cell. We have implemented this oracle for the trilinear interpolant and plan to replace it with higher order interpolants when needed. The complexity of the scheme is O(n + t log n), where t is the number of critical points of F. For the first time we can compute the Contour Tree in linear time in many practical cases when t = O(n1 - &epsilon;).Lastly, we report the running times for a parallel implementation of our algorithm, showing good scalability with the number of processors.
IEEE Visualization	Efficient Simplification of Point-Sampled Surfaces.	Mark Pauly,Markus H. Gross,Leif Kobbelt	2002	In this paper we introduce, analyze and quantitatively compare a number of surface simplification methods for point-sampled geometry. We have implemented incremental and hierarchical clustering, iterative simplification, and particle simulation algorithms to create approximations of point-based models with lower sampling density. All these methods work directly on the point cloud, requiring no intermediate tesselation. We show how local variation estimation and quadric error metrics can be employed to diminish the approximation error and concentrate more samples in regions of high curvature. To compare the quality of the simplified surfaces, we have designed a new method for computing numerical and visual error estimates for point-sampled surfaces. Our algorithms are fast, easy to implement, and create high-quality surface approximations, clearly demonstrating the effectiveness of point-based surface simplification.
IEEE Visualization	Integration of Measurement Tools in Medical 3d Visualizations.	Bernhard Preim,Christian Tietjen,Wolf Spindler,Heinz-Otto Peitgen	2002	We discuss 3d interaction techniques for the quantitative analysis of spatial relations in medical visualizations. We describe the design and implementation of measurement tools to measure distances, angles and volumes in 3d visualizations. The visualization of measurement tools as recognizable 3d objects and a 3d interaction, which is both intuitive and precise, determines the usability of such facilities. Measurements may be carried out in 2d visualizations of the original radiological data and in 3d visualizations. The result of a measurement carried out in one view is also displayed in the other view appropriately. We discuss the validation of the obtained measures. Finally, we describe how some important measurement tasks may be solved automatically.
IEEE Visualization	Fast Visualization of Plane-Like Structures in Voxel Data.	Steffen Prohaska,Hans-Christian Hege	2002	"We present a robust, noise-resistant criterion characterizing plane-like skeletons in binary voxel objects. It is based on a distance map and the geodesic distance along the object's boundary. A parameter allows to control the noise sensitivity.If needed, homotopy with the original object might be reconstructed in a second step, using an improved distance ordered thinning algorithm.The skeleton is analyzed to create a geometric representation for rendering. Plane-like parts are transformed into an triangulated surface not enclosing a volume by a suitable triangulation scheme. The resulting surfaces have lower triangle count than those created with standard methods and tend to maintain the original geometry, even after simplification with a high decimation rate. Our algorithm allows to interactively render expressive images of complex 3D structures, emphasizing independently plane-like and rod-like structures.The methods are applied for visualization of the microstructure of bone biopsies."
IEEE Visualization	Visualizing Dynamic Molecular Conformations.	Johannes Schmidt-Ehrenberg,Daniel Baum,Hans-Christian Hege	2002	"The bioactivity of a molecule strongly depends on its metastable conformational shapes and the transitions between these. Therefore, conformation analysis and visualization is a basic prerequisite for the understanding of biochemical processes.We present techniques for visual analysis of metastable molecular conformations. Core of these are flexibly applicable methods for alignment of molecular geometries, as well as methods for depicting shape and 'fuzziness' of metastable conformations. All analysis tools are provided in an integrated working environment.The described techniques are demonstrated with pharmaceutically active biomolecules."
IEEE Visualization	Seamster: Inconspicuous Low-Distortion Texture Seam Layout.	Alla Sheffer,John C. Hart	2002	Surface texturing aids the visualization of polygonal meshes by providing additional surface orientation cues and feature annotations. Such texturing is usually implemented via texture mapping, which is easier and more effective when the distortion of the mapping from the surface to the texture map is kept small.We have previously shown that distortion occurs when areas of high surface curvature are flattened into the texture map. By cutting the surface in these areas one can reduce texture map distortion at the expense of additional seam artifacts.This paper describes a faster technique for guiding a texture map seam through high distortion regions, while restricting the seam to regions of low visibility. This results in distortion reducing seams that are less visually distracting and take less time to compute. We have also observed that visibility considerations improve the speed of a recent method that adds cuts to reduce a surface genus.
IEEE Visualization	"NASA's Great Zooms: A Case Study."	Gregory W. Shirah,Horace Mitchell	2002	This paper examines a series of NASA outreach visualizations created using several layers of remote sensing satellite data ranging from 4-kilometers per pixel to 1-meter per pixel. The viewer is taken on a seamless, cloud free journey from a global view of the Earth down to ground level where buildings, streets, and cars are visible. The visualizations were produced using a procedural shader that takes advantage of accurate georegistration and color matching between images. The shader accurately and efficiently maps the data sets to geometry allowing for animations with few perceptual transitions among data sets. We developed a pipeline to facilitate the production of over twenty zoom visualizations. Millions of people have seen these visualizations through national and international media coverage.
IEEE Visualization	Tensor Field Visualisation using Adaptive Filtering of Noise Fields combined with Glyph Rendering.	Andreas Sigfridsson,Tino Ebbers,Einar Heiberg,Lars Wigström	2002	While many methods exist for visualising scalar and vector data, visualisation of tensor data is still troublesome. We present a method for visualising second order tensors in three dimensions using a hybrid between direct volume rendering and glyph rendering.An overview scalar field is created by using three-dimensional adaptive filtering of a scalar field containing noise. The filtering process is controlled by the tensor field to be visualised, creating patterns that characterise the tensor field. By combining direct volume rendering of the scalar field with standard glyph rendering methods for detailed tensor visualisation, a hybrid solution is created.A combined volume and glyph renderer was implemented and tested with both synthetic tensors and strain-rate tensors from the human heart muscle, calculated from phase contrast magnetic resonance image data. A comprehensible result could be obtained, giving both an overview of the tensor field as well as detailed information on individual tensors.
IEEE Visualization	Bounded-distortion Piecewise Mesh Parameterization.	Olga Sorkine,Daniel Cohen-Or,Rony Goldenthal,Dani Lischinski	2002	Many computer graphics operations, such as texture mapping, 3D painting, remeshing, mesh compression, and digital geometry processing, require finding a low-distortion parameterization for irregular connectivity triangulations of arbitrary genus 2-manifolds. This paper presents a simple and fast method for computing parameterizations with strictly bounded distortion. The new method operates by flattening the mesh onto a region of the 2D plane. To comply with the distortion bound, the mesh is automatically cut and partitioned on-the-fly. The method guarantees avoiding global and local self-intersections, while attempting to minimize the total length of the introduced seams.To our knowledge, this is the first method to compute the mesh partitioning and the parameterization simultaneously and entirely automatically, while providing guaranteed distortion bounds. Our results on a variety of objects demonstrate that the method is fast enough to work with large complex irregular meshes in interactive applications.
IEEE Visualization	A Case Study On The Applications Of A Generic Library For Low-Cost Polychromatic Passive Stereo.	Simon Stegmaier,Dirc Rose,Thomas Ertl	2002	Active stereo has been used by engineers and industrial designers for several years to enhance the perception of computer generated three-dimensional images. Unfortunately, active stereo requires specialized hardware. Therefore, as ubiquitous computing and teleworking gain importance, using active stereo becomes a problem. The goal of this case study is to examine the concept of a generic library for polychromatic passive stereo to make stereo vision available everywhere.
IEEE Visualization	A Case Study on Automatic Camera Placement and Motion for Visualizing Historical Data.	Stanislav L. Stoev,Wolfgang Straßer	2002	In this paper, we address the problem of automatic camera positioning and automatic camera path generation in the context of historical data visualization. After short description of the given data, we elaborate on the constraints for the positioning of a virtual camera in such a way that not only the projected area is maximized, but also the depth of the displayed scene. This is especially important when displaying terrain models, which do not provide good 3D impression when only the projected area is maximized. Based on this concept, we present a method for computing an optimal camera position for each instant of time.Since the explored data are not static, but change depending on the explored scene time, we also discuss a method for animation generation. In order to avoid sudden changes of the camera position, when the previous method is applied for each frame (point in time), we introduce pseudo-events in time, which expand the bounding box defined by the currently active events of interest. In particular, this technique allows events happening in a future point in time to be taken into account such that when this time becomes current, all events of interest are already within the current viewing frustum of the camera.
IEEE Visualization	Case Study: Hardware-Accelerated Selective LIC Volume Rendering.	Yasuko Suzuki,Issei Fujishiro,Li Chen,Hiroko Nakamura	2002	Line Integral Convolution (LIC) is a promising method for visualizing 2D dense flow fields. Direct extensions of the LIC method to 3D have not been considered very effective, because optical integration in viewing directions tends to spoil the coherent structures along 3D local streamlines. In our previous reports, we have proposed a selective approach to volume rendering of LIC solid texture using 3D significance map (S-map), derived from the characteristics of flow structures, and a specific illumination model for 3D streamlines. In this paper, we take full advantage of scalar volume rendering hardware, such as VolumePro, to realize a realtime 3D flow field visualization environment with the LIC volume rendering method.
IEEE Visualization	Geometric Surface Smoothing via Anisotropic Diffusion of Normals.	Tolga Tasdizen,Ross T. Whitaker,Paul Burchard,Stanley Osher	2002	This paper introduces a method for smoothing complex, noisy surfaces, while preserving (and enhancing) sharp, geometric features. It has two main advantages over previous approaches to feature preserving surface smoothing. First is the use of level set surface models, which allows us to process very complex shapes of arbitrary and changing topology. This generality makes it well suited for processing surfaces that are derived directly from measured data. The second advantage is that the proposed method derives from a well-founded formulation, which is a natural generalization of anisotropic diffusion, as used in image processing. This formulation is based on the proposition that the generalization of image filtering entails filtering the normals of the surface, rather than processing the positions of points on a mesh.
IEEE Visualization	BLIC: Bi-Level Isosurface Compression .	Gabriel Taubin	2002	In this paper we introduce a new and simple algorithm to compress isosurface data. This is the data extracted by isosurface algorithms from scalar functions defined on volume grids, and used to generate polygon meshes or alternative representations. In this algorithm the mesh connectivity and a substantial proportion of the geometric information are encoded to a fraction of a bit per Marching Cubes vertex with a context based arithmetic coder closely related to the JBIG binary image compression standard. The remaining optional geometric information that specifies the location of each Marching Cubes vertex more precisely along its supporting intersecting grid edge, is efficiently encoded in scan-order with the same mechanism. Vertex normals can optionally be computed as normalized gradient vectors by the encoder and included in the bitstream after quantization and entropy encoding, or computed by the decoder in a postprocessing smoothing step. These choices are determined by trade-offs associated with an in-core vs. out-of-core decoder structure. The main features of our algorithm are its extreme simplicity and high compression rates
IEEE Visualization	Case Study: Interactive Visualization for Internet Security.	Soon Tee Teoh,Kwan-Liu Ma,Shyhtsun Felix Wu,Xiaoliang Zhao	2002	Internet connectivity is defined by a set of routing protocols which let the routers that comprise the Internet backbone choose the best route for a packet to reach its destination. One way to improve the security and performance of Internet is to routinely examine the routing data. In this case study, we show how interactive visualization of Border Gateway Protocol (BGP) data helps characterize routing behavior, identify weaknesses in connectivity which could potentially cripple the Internet, as well as detect and explain actual anomalous events.
IEEE Visualization	Case Study on the Adaptation of Interactive Visualization Applications to Web-Based Production for Operational Mesoscale Weather Models.	Lloyd Treinish	2002	Visualization is required for the effective utilization of data from a weather simulation. Appropriate mapping of user goals to the design of pictorial content has been useful in the development of interactive applications with sufficient bandwidth for timely access to the model data. When remote access to the model visualizations is required the limited bandwidth becomes the primary bottleneck. To help address these problems, visualizations are presented on a web page as a meta-representation of the model output and serve as an index to simplify finding other visualizations of relevance. To provide consistency with extant interactive products and to leverage their cost of development, the aforementioned applications are adapted to automatically populate a web site with images and interactions for an operational weather forecasting system.
IEEE Visualization	"Case Study: ""The Office of Real Soon Now"" for Visualization."	Samuel P. Uselton	2002	"As part of a larger effort exploring alternative display systems, Lawrence Livermore National Laboratory has installed systems in two offices that extend and update the previously described ""Office of Real Soon Now"" project to improve the value for visualization tasks. These new systems use higher resolution projectors driven by workstations that run Unix-based applications via Linux and support hardware-accelerated 3D graphics, even across the boundary between displays."
IEEE Visualization	Out-of-Core Rendering of Massive Geometric Datasets.	Gokul Varadhan,Dinesh Manocha	2002	Out-of-Core Rendering of Massive Geometric Datasets.
IEEE Visualization	Fast and Reliable Space Leaping for Interactive Volume Rendering.	Ming Wan,Aamir Sadiq,Arie E. Kaufman	2002	We present a fast and reliable space-leaping scheme to accelerate ray casting during interactive navigation in a complex volumetric scene, where we combine innovative space-leaping techniques in a number of ways. First, we derive most of the pixel depths at the current frame by exploiting the temporal coherence during navigation, where we employ a novel fast cell-based reprojection scheme that is more reliable than the traditional intersection-point based reprojection. Next, we exploit the object space coherence to quickly detect the remaining pixel depths, by using a precomputed accurate distance field that stores the Euclidean distance from each empty (background) voxel toward its nearest object boundary. In addition, we propose an effective solution to the challenging new-incoming-objects problem during navigation. Our algorithm has been implemented on a 16-processor SGI Power Challenge and reached interactive rendering rates at more than 10 Hz during the navigation inside 5123 volume data sets acquired from both a simulation phantom and actual patients.
IEEE Visualization	Exploring Scalar Fields Using Critical Isovalues.	Gunther H. Weber,Gerik Scheuermann,Hans Hagen,Bernd Hamann	2002	"Isosurfaces are commonly used to visualize scalar fields. Critical isovalues indicate isosurface topology changes: the creation of new surface components, merging of surface components or the formation of holes in a surface component. Therefore, they highlight ""interesting"" isosurface behavior and are helpful in exploration of large trivariate data sets. We present a method that detects critical isovalues in a scalar field defined by piecewise trilinear interpolation over a rectilinear grid and describe how to use them when examining volume data. We further review varieties of the Marching Cubes (MC) algorithm, with the intention to preserve topology of the trilinear interpolant when extracting an isosurface. We combine and extend two approaches in such a way that it is possible to extract meaningful isosurfaces even when a critical value is chosen as isovalue."
IEEE Visualization	Simulating Fire with Texture Splats.	Xiaoming Wei,Wei Li,Klaus Mueller,Arie E. Kaufman	2002	We propose the use of textured splats as the basic display primitives for an open surface fire model. The high-detail textures help to achieve a smooth boundary of the fire and gain the small-scale turbulence appearance. We utilize the Lattice Boltzmann Model (LBM) to simulate physically-based equations describing the fire evolution and its interaction with the environment (e.g., obstacles, wind and temperature). The property of fuel and non-burning objects are defined on the lattice of the computation domain. A temperature field is also incorporated to model the generation of smoke from the fire due to incomplete combustion. The linear and local characteristics of the LBM enable us to accelerate the computation with graphics hardware to reach real-time simulation speed, while the texture splat primitives enable interactive rendering frame rates.
IEEE Visualization	Volume Clipping via Per-Fragment Operations in Texture-Based Volume Visualization.	Daniel Weiskopf,Klaus Engel,Thomas Ertl	2002	We propose new clipping methods that are capable of using complex geometries for volume clipping. The clipping tests exploit per-fragment operations on the graphics hardware to achieve high frame rates. In combination with texture-based volume rendering, these techniques enable the user to interactively select and explore regions of the data set. We present depth-based clipping techniques that analyze the depth structure of the boundary representation of the clip geometry to decide which parts of the volume have to be clipped. In another approach, a voxelized clip object is used to identify the clipped regions.
IEEE Visualization	Volumetric Shadows Using Splatting.	Caixia Zhang,Roger Crawfis	2002	This paper describes an efficient algorithm to model the light attenuation due to a participating media with low albedo. The light attenuation is modeled using splatting volume renderer for both the viewer and the light source. During the rendering, a 2D shadow buffer attenuates the light for each pixel. When the contribution of a footprint is added to the image buffer, as seen from the eye, we add the contribution to the shadow buffer, as seen from the light source. We have generated shadows for point lights and parallel lights using this algorithm. The shadow algorithm has been extended to deal with multiple light sources and projective textured lights.
IEEE Visualization	Direct Surface Extraction from 3D Freehand Ultrasound Images.	Youwei Zhang,Robert Rohling,Dinesh K. Pai	2002	This paper presents a new technique for the extraction of surfaces from 3D ultrasound data. Surface extraction from ultrasound data is challenging for a number of reasons including noise and artifacts in the images and non-uniform data sampling. A method is proposed to fit an approximating radial basis function to the group of data samples. An explicit surface is then obtained by iso-surfacing the function. In most previous 3D ultrasound research, a pre-processing step is taken to interpolate the data into a regular voxel array and a corresponding loss of resolution. We are the first to represent the set of semi-structured ultrasound pixel data as a single function. From this we were able to extract surfaces without first reconstructing the irregularly spaced pixels into a regular 3D voxel array.
IEEE Visualization	Visibility-Guided Simplification.	Eugene Zhang,Greg Turk	2002	For some graphics applications, object interiors and hard-to-see regions contribute little to the final images and need not be processed. In this paper, we define a view-independent visibility measure on mesh surfaces based on the visibility function between the surfaces and a surrounding sphere of cameras. We demonstrate the usefulness of this measure with a visibility-guided simplification algorithm.Mesh simplification reduces the polygon counts of 3D models and speeds up the rendering process. Many mesh simplification algorithms are based on sequences of edge collapses that minimize geometric and attribute errors. By combining the surface visibility measure with a geometric error measure, we obtain simplified models with improvement proportional to the amount of low visibility regions in the original models.
IEEE Visualization	Volume Deformation For Tensor Visualization.	Xiaoqiang Zheng,Alex Pang	2002	Visualizing second-order 3D tensor fields continue to be a challenging task. Although there are several algorithms that have been presented, no single algorithm by itself is sufficient for the analysis because of the complex nature of tensor fields. In this paper, we present two new methods, based on volume deformation, to show the effects of the tensor field upon its underlying media. We focus on providing a continuous representation of the nature of the tensor fields. Each of these visualization algorithms is good at displaying some particular properties of the tensor field.
IEEE Visualization	Comparative Evaluation of Visualization and Experimental Results Using Image Comparison Metrics.	Hualin Zhou,Min Chen,Mike F. Webster	2002	"Comparative evaluation of visualization and experiment results is a critical step in computational steering. In this paper, we present a study of image comparison metrics for quantifying the magnitude of difference between a visualization of a computer simulation and a photographic image captured from an experiment. We examined eleven metrics, including three spatial domain, four spatial-frequency domain and four HVS (human-vision system) metrics. Among these metrics, a spatial-frequency domain metric called 2nd-order Fourier comparison was proposed specifically for this work. Our study consisted of two stages: base cases and field trials. The former is a general study on a controlled comparison space using purposely selected data, and the latter involves imagery results from computational fluid dynamics and a rheological experiment. This study has introduced a methodological framework for analyzing image-level methods used in comparative visualization. For the eleven metrics considered, it has offered a set of informative indicators as to the strengths and weaknesses of each metric. In particular, we have identified three image comparison metrics that are effective in separating ""similar"" and ""different"" image groups. Our 2nd-order Fourier comparison metric has compared favorably with others in two of the three tests, and has shown its potential to be used for steering computer simulation quantitatively."
IEEE Visualization	Oriented Tensor Reconstruction: Tracing Neural Pathways from Diffusion Tensor MRI.	Leonid Zhukov,Alan H. Barr	2002	In this paper we develop a new technique for tracing anatomical fibers from 3D tensor fields. The technique extracts salient tensor features using a local regularization technique that allows the algorithm to cross noisy regions and bridge gaps in the data. We applied the method to human brain DT-MRI data and recovered identifiable anatomical structures that correspond to the white matter brain-fiber pathways. The images in this paper are derived from a dataset having 121x88x60 resolution. We were able to recover fibers with less than the voxel size resolution by applying the regularization technique, i.e., using a priori assumptions about fiber smoothness. The regularization procedure is done through a moving least squares filter directly incorporated in the tracing algorithm.
IEEE Visualization	IEEE Visualization 2002, October 27-November 01, 2002, Boston, MA, USA.		2002	IEEE Visualization 2002, October 27-November 01, 2002, Boston, MA, USA.
Information Visualization	The InfoSky visual explorer: exploiting hierarchical structure and document similarities.	Keith Andrews,Wolfgang Kienreich,Vedran Sabol,Jutta Becker,Georg Droschl,Frank Kappe,Michael Granitzer,Peter Auer,Klaus Tochtermann	2002	InfoSky is a system enabling users to explore large, hierarchically structured document collections. Similar to a real-world telescope, InfoSky employs a planar graphical representation with variable magnification. Documents of similar content are placed close to each other and are visualised as stars, forming clusters with distinct shapes. For greater performance, the hierarchical structure is exploited and force-directed placement is applied recursively at each level on much fewer objects, rather than on the whole corpus. Collections of documents at a particular level in the hierarchy are visualised with bounding polygons using a modified weighted Voronoi diagram. Their area is related to the number of documents contained. Textual labels are displayed dynamically during navigation, adjusting to the visualisation content. Navigation is animated and provides a seamless zooming transition between summary and detail view. Users can map metadata such as document size or age to attributes of the visualisation such as colour and luminance. Queries can be made and matching documents or collections are highlighted. Formative usability testing is ongoing; a small baseline experiment comparing the telescope browser to a tree browser is discussed.
Information Visualization	Inventing discovery tools: combining information visualization with data mining?	Ben Shneiderman	2002	The growing use of information visualization tools and data mining algorithms stems from two separate lines of research. Information visualization researchers believe in the importance of giving users an overview and insight into the data distributions, while data mining researchers believe that statistical algorithms and machine learning can be relied on to find the interesting patterns. This paper discusses two issues that influence design of discovery tools: statistical algorithms vs visual data presentation, and hypothesis testing vs exploratory data analysis. The paper claims that a combined approach could lead to novel discovery tools that preserve user control, enable more effective exploration, and promote responsibility.
Information Visualization	Filtering and brushing with motion.	Lyn Bartram,Colin Ware	2002	Visualizing information in user interfaces to complex, large-scale systems is difficult due to visual fragmentation caused by an enormous amount of inter-related data distributed across multiple views. New display dimensions are required to help the user visually integrate and filter such spatially distributed and heterogeneous information. Motion holds promise in this regard as a perceptually efficient display dimension. It has long been known to have a strong grouping effect, suggesting it has potential for filtering and brushing techniques. However, there is little known about which properties of motion are most effective. This paper reviews the prior literature relating to the use of motion for display and discusses the requirements for how motion can be usefully applied to these problems, especially for visualizations incorporating multiple groups of data objects. Results from previous research by the authors suggested motion type was a strong distinguishing feature. Three types of motions in pairwise combinations were compared: linear, circular and expansion/contraction. Combinations of linear directions were also compared to evaluate how great angular separation needs to be to enforce perceptual distinction. The results showed that motion can effectively group objects that are otherwise dissimilar. Type differentiation is more effective than directional differences (except for 90°). Of the three types studied, circular demands the most attention. Angular separation must be 90° to be equally effective. These results suggest that motion can be usefully applied to both filtering and brushing. They also provide the beginnings of a vocabulary of simple motions that can be applied to information visualization.
Information Visualization	Rapid, Serial and Visual: a presentation technique with potential.	Robert Spence	2002	This commentary addresses the recently exploited presentation technique called Rapid Serial Visual Presentation, or RSVP, and identifies a number of applications. It describes a number of different RSVP modes, discusses potentially attractive features and reports the results of experimental investigations. Some of the many issues requiring further research are identified.
Information Visualization	Sensitivity encoding to support information space navigation: a design guideline.	Robert Spence	2002	"This article identifies, illustrates and offers justification for a guideline intended to support interaction design for navigation in electronic information spaces. The essence of the guideline is the concept of 'sensitivity encoding', a feature of a data display intended to provide guidance regarding potentially beneficial movement in information space. That navigation's objective may be the formation of a mental model, movement towards an objective, or both."
Information Visualization	Information Visualization.	Chaomei Chen	2002	Information Visualization.
Information Visualization	Aesthetics versus Functionality.	Chaomei Chen	2002	Aesthetics versus Functionality.
Information Visualization	Information visualization is growing.	Chaomei Chen	2002	Information visualization is growing.
Information Visualization	A taxonomy of glyph placement strategies for multidimensional data visualization.	Matthew O. Ward	2002	Glyphs are graphical entities that convey one or more data values via attributes such as shape, size, color, and position. They have been widely used in the visualization of data and information, and are especially well suited for displaying complex, multivariate data sets. The placement or layout of glyphs on a display can communicate significant information regarding the data values themselves as well as relationships between data points, and a wide assortment of placement strategies have been developed to date. Methods range from simply using data dimensions as positional attributes to basing placement on implicit or explicit structure within the data set. This paper presents an overview of multivariate glyphs, a list of issues regarding the layout of glyphs, and a comprehensive taxonomy of placement strategies to assist the visualization designer in selecting the technique most suitable to his or her data and task. Examples, strengths, weaknesses, and design considerations are given for each category of technique. We conclude with some general guidelines for selecting a placement strategy, along with a brief description of some of our future research directions.
Information Visualization	Cognitive measurements of graph aesthetics.	Colin Ware,Helen C. Purchase,Linda Colpoys,Matthew McGill	2002	A large class of diagrams can be informally characterized as node-link diagrams. Typically nodes represent entities, and links represent relationships between them. The discipline of graph drawing is concerned with methods for drawing abstract versions of such diagrams. At the foundation of the discipline are a set of graph aesthetics (rules for graph layout) that, it is assumed, will produce graphs that can be clearly understood. Examples of aesthetics include minimizing edge crossings and minimizing the sum of the lengths of the edges. However, with a few notable exceptions, these aesthetics are taken as axiomatic, and have not been empirically tested. We argue that human pattern perception can tell us much that is relevant to the study of graph aesthetics including providing a more detailed understanding of aesthetics and suggesting new ones. In particular, we find the importance of good continuity (ie keeping multi-edge paths as straight as possible) has been neglected. We introduce a methodology for evaluating the cognitive cost of graph aesthetics and we apply it to the task of finding the shortest paths in spring layout graphs. The results suggest that after the length of the path the two most important factors are continuity and edge crossings, and we provide cognitive cost estimates for these parameters. Another important factor is the number of branches emanating from nodes on the path.
Information Visualization	Multivariate visualization with data fusion.	Pak Chung Wong,Harlan Foote,David L. Kao,L. Ruby Leung,Jim Thomas	2002	We discuss a fusion-based visualization method to analyze a multivariate climate dataset and its metadata. The primary difference between a conventional visualization and a fusion-based visualization is that the former draws on a single image whereas the latter draws on multiple see-through layers, which are then overlaid on each other to form the final visualization. We propose optimized colormaps to highlight subtle features that would not be shown with conventional colormaps. We present fusion techniques that integrate multiple single-purpose visualization techniques into the same viewing space. Our highly flexible fusion approach allows scientists to explore multiple parameters concurrently by mixing and matching images without frequently reconstructing new visualizations from the data for every possible combination. Although our primary visualization application is climate modeling, we show with examples that our fundamental design - fusing layers of data images for multivariate visualization - can be generalized for other information visualization applications.
Information Visualization	DEViD: a media design and software ergonomics integrating visualization for document retrieval.	Maximilian Eibl	2002	Human Computer Interaction (HCI) and Media Design are two separated schools that have few common goals in the area of designing user interfaces. Whereas HCI places the accent on the most effective operation, media design emphasizes the aesthetic aspect. Thus, the design of a user interface offers a choice between the effective ugliness of ergonomics and the hard-touse elegance of media design. One example of the successful integration of both schools is the DEViD (media design and software ergonomics integrating visualization for document retrieval systems) visualization described in this article. DEViD supports users in searching databases and integrates Boolean retrieval as well as probabilistic ranking and vague retrieval within one single interface. DEViD was created through cooperation between HCI and media design. Participants in this cooperation were the Social Science Information Center, Bonn (Germany), the University of Koblenz (Germany) and the School of Design, Offenbach (Germany).
Information Visualization	Representing high-dimensional data sets as closed surfaces.	Archana P. Sangole,George K. Knopf	2002	Scientific data visualization provides scientists and engingeers with a deeper insight and greater understanding about physical phenomena through the use of graphical tools. Individuals are able to identify patterns embedded in data sets using visual cues such as color and shape, rather than directly searching through a vast volume of numbers. The visualization algorithm described in this paper utilizes a spherical self-organizing feature map (SOFM) to automatically cluster and develop a well-defined topology of arbitrary data vectors, based on a pre-defined measure of similarity, and generate a three-dimensional color-coded surface model that reflects cluster variations. Implementation of this self-organizing surface geometry for data visualization applications is illustrated by examining the graphical forms created for a small synthetic test data set and a large environmental data-base. The proposed methodology provides the researcher with a new tool to encode information into shape and easily transfer the geometric model to an immersive virtual reality (IVR) environment for interactive information analysis.
Information Visualization	Zoomable user interfaces as a medium for slide show presentations.	Lance Good,Benjamin B. Bederson	2002	In this paper, the authors propose Zoomable User Interfaces as an alternative presentation medium to address several common presentation problems. Zoomable User Interfaces offer new techniques for managing multiple versions of a presentation, providing interactive presentation navigation, and distinguishing levels of detail. These zoomable presentations may also offer several cognitive benefits over their commercial slide show counterparts. The authors also introduce CounterPoint, a tool to simplify the creation and delivery of zoomable presentations, discuss the techniques they have used to make authoring and navigation manageable in the multidimensional space. Lastly, some of the visualization principles compiled by the authors for designing these types of presentations are presented.
Information Visualization	Pixel bar charts: a visualization technique for very large multi-attribute data sets?	Daniel A. Keim,Ming C. Hao,Umeshwar Dayal,Meichun Hsu	2002	Simple presentation graphics are intuitive and easy-to-use, but show only highly aggregated data presenting only a very small number of data values (as in the case of bar charts) and may have a high degree of overlap occluding a significant portion of the data values (as in the case of the x-y plots). In this article, the authors therefore propose a generalization of traditional bar charts and x-y plots, which allows the visualization of large amounts of data. The basic idea is to use the pixels within the bars to present detailed information of the data records. The so-called pixel bar charts retain the intuitiveness of traditional bar charts while allowing very large data sets to be visualized in an effective way. It is shown that, for an effective pixel placement, a complex optimization problem has to be solved. The authors then present an algorithm which efficiently solves the problem. The application to a number of real-world e-commerce data sets shows the wide applicability and usefulness of this new idea, and a comparison to other well-known visualization techniques (parallel coordinates and spiral techniques) shows a number of clear advantages.
Information Visualization	Information visualization for advanced vehicle displays.	Aaron Marcus	2002	Almost all vehicles soon will acquire advanced user interfaces with specialized information visualization. This revolution poses significant human factors issues, but also new challenges of appeal, cultural orientation, and branding. This article describes considerations beyond traditional human factors and shows conceptual visual prototypes that explore new approaches to vehicle information displays.
Information Visualization	Visualization schemas and a web-based architecture for custom multiple-view visualization of multiple-table databases.	Chris North,Nathan Conklin,Kiran Indukuri,Varun Saini	2002	Relational databases provide significant flexibility to organize, store, and manipulate an infinite variety of complex data collections. This flexibility is enabled by the concept of relational data schemas, which allow data owners to easily design custom databases according to their unique needs. However, user interfaces and information visualizations for accessing and utilizing databases have not kept pace with this level of flexibility. Visualizations need to integrate multiple tables and diverse visualization tools into custom solutions. This paper describes advances to Snap-Together Visualization, introduces Visualization Schemas, and presents an extensible system architecture. The Snap model for custom multiple-view visualization establishes an analogy to the relational data model, enabling coordinated data design and visualization design. Visualization Schemas are a natural extension to data schemas, and provide a user interface that enables data owners to rapidly construct and disseminate custom visualizations without programming. The web-based software architecture supports run-time extensibility, enabling end-user integration and dissemination of diverse data and visualization tools from the field.
Information Visualization	Geobrowsing: creative thinking and knowledge discovery using geographic visualization.	Donna Peuquet,Menno-Jan Kraak	2002	"In the modern computing context, the map is no longer just a final product. Maps are now being used in a fundamentally different way - as a self-directed tool for deriving the desired information from geographic data. This, along with developments in GIScience and computer graphics, have led to the new field of geographic visualization. A central issue is how to design visualization capabilities that, as a process, facilitate creative thinking for discovering previously new information from large databases. The authors propose the term 'geobrowsing' to designate this process. A number of interrelated ways that visualization can be used to spark the imagination in order to derive new insights are discussed and a brief example provided. Based upon the cognitive literature, specific properties of a visual image that promote discovery and insight are discussed. These are known as preinventive properties, and include; novelty, incongruence, abstraction, and ambiguity. All of these properties, either individually or in combination, tend to produce features that are unanticipated by the viewer, and often not explicitly created or anticipated by the person generating the visual display. While traditional (i.e. non-computer generated) images can also possess these properties, as shown in the historical examples in this discussion, it is the capability of the viewer to directly and quickly manipulate these properties that provides the real power of 'geobrowsing' for uncovering new insights."
Information Visualization	Animated visualization of multiple intersecting hierarchies.	George G. Robertson,Kim Cameron,Mary Czerwinski,Daniel C. Robbins	2002	"The authors describe a new information structure composed of multiple intersecting hierarchies, which we call a Polyarchy. Visualizing polyarchies enables use of novel views for discovery of relationships which are very difficult using existing hierarchy visualization tools. This paper will describe the visualization design and system architecture challenges as well as our current solutions. Visual Pivot is a novel web-based polyarchy visualization technique, supported by a 'polyarchy server' implemented with a Mid-Tier Cache architecture. A series of five user studies guided iterative design of Visual Pivot. Finally, the effectiveness of animation in Visual Pivot is discussed."
Information Visualization	Liver cancer detection system based on the analysis of digitized color images of tissue samples obtained using needle biopsy.	Mohamed Sammouda,Rachid Sammouda,Noboru Niki,Kiyoshi Mukai	2002	In this article, the authors propose a method for automatic diagnosis of liver cancer based on analysis of digitized color images of liver tissue obtained by needle biopsy. The approach is a combination of an unsupervised segmentation algorithm, using a modified artificial Hopfield neural network (HNN), and an analysis algorithm based on image quantization. The segmentation algorithm is superior to HNN in the sense that it converges to a nearby global minimum rather than a local one in a prespecified time. Furthermore, as the segmentation of color images does not only depend on the segmentation algorithm but also on the color space representation, and in order to choose the best segmentation result, segmentation was performed with HNN and using components of the raw image with respect to each of the RGB, HLS, and HSV color spaces. Then, the segmented image was labeled based on chromaticity features and histogram analysis of the RGB color space components of the raw image. The image regions were then classified into normal and cancerous using diagnostic rules formulated based on those used by experienced pathologists in the clinic. The proposed method provides quantitative satisfactory results in diagnosing a liver pathological image set of 17 cases.
IEEE Trans. Vis. Comput. Graph.	MGV: A System for Visualizing Massive Multidigraphs.	James Abello,Jeffrey L. Korn	2002	"We describe MGV, an integrated visualization and exploration system for massive multidigraph navigation. It adheres to the Visual Information-Seeking Mantra: overview first, zoom and filter, then details on demand. MGV's only assumption is that the vertex set of the underlying digraph corresponds to the set of leaves of a predetermined tree $T$. MGV builds an out-of-core graph hierarchy and provides mechanisms to plug in arbitrary visual representations for each graph hierarchy slice. Navigation from one level to another of the hierarchy corresponds to the implementation of a drill-down interface. In order to provide the user with navigation control and interactive response, MGV incorporates a number of visualization techniques like interactive pixel-oriented 2D and 3D maps, statistical displays, color maps, multilinked views, and a zoomable label based interface. This makes the association of geographic information and graph data very natural. To automate the creation of the vertex set hierarchy for MGV, we use the notion of graph sketches. They can be thought of as visual indices that guide the navigation of a multigraph too large to fit on the available display. MGV follows the client-server paradigm and it is implemented in C and Java-3D. We highlight the main algorithmic and visualization techniques behind the tools and, along the way, point out several possible application scenarios. Our techniques are being applied to multigraphs defined on vertex sets with sizes ranging from 100 million to 250 million vertices."
IEEE Trans. Vis. Comput. Graph.	A Generic Rendering System.	Jürgen Döllner,Klaus Hinrichs	2002	We describe the software architecture of a rendering system that follows a pragmatic approach to integrating and bundling the power of different low-level rendering systems within an object-oriented framework. The generic rendering system provides higher-level abstractions to existing rendering systems and serves as a framework for developing new rendering techniques. It wraps the functionality of several, widely used rendering systems, defines a unified, object-oriented application programming interface, and provides an extensible, customizable apparatus for evaluating and interpreting hierarchical scene information. As a fundamental property, individual features of a specific rendering system can be integrated into the generic rending system in a transparent way. The system is based on a state machine, called engine, which operates on rendering components. Four major categories of rendering components constitute the generic rendering system: shapes represent geometries, attributes specify properties assigned to geometries and scenes, handlers encapsulate rendering algorithms, and techniques represent evaluation strategies for rendering components. As a proof of concept, we have implemented the described software architecture by the Virtual Rendering System which currently wraps OpenGL, Radiance, POV Ray, and RenderMan.
IEEE Trans. Vis. Comput. Graph.	An Anatomy-Based Approach to Human Muscle Modeling and Deformation.	Feng Dong,Gordon Clapworthy,Meleagros A. Krokos,Jialiang Yao	2002	Muscle simulation is an important component of human modeling, but there have been few attempts to demonstrate, in three dimensions and in an anatomically correct way, the structures of muscles and the way in which these change during motion. This paper proposes an anatomically-based approach to muscle modeling that attempts to provide models for human musculature based on the real morphological structures. These models provide a good visual description of muscle form and action and represent a sound base from which to produce further progress toward medically accurate simulation of human bodies. Three major problems have been addressed: geometric modeling, deformation, and texture. To allow for the wide variety of deformable muscle shapes encountered in the body, while retaining as many of their common properties as possible, the geometric models are classified into several categories according to the characteristics of their structures and actions. Within each category, the model for each muscle has an efficient structural form, created using anatomical data. Deformation is also performed on the basis of the categories, with all models within each category sharing the same deformation scheme. The categories cover both general and special cases. The result is an efficient, anatomically accurate muscle representation that is specifically designed to accommodate the particular form of deformation exhibited by each individual muscle. Interactions between muscles are also taken into account to avoid penetration occurring between adjacent muscles in our model. To provide a suitable visual effect, the muscle texture is generated directly on the model surface. The textures and colors are obtained from anatomical data via image analysis. Some results are presented on the geometric modeling, the deformation, and the texture of muscles related to the lower limb.
IEEE Trans. Vis. Comput. Graph.	Three-Dimensional Interfaces for Querying by Example in Content-Based Image Retrieval.	Jürgen Assfalg,Alberto Del Bimbo,Pietro Pala	2002	"Image databases are nowadays widely exploited in a number of different contexts, ranging from history of art, through medicine, to education. Existing querying paradigms are based either on the usage of textual strings, for high-level semantic queries or on 2D visual examples for the expression of perceptual queries. Semantic queries require manual annotation of the database images. Instead, perceptual queries only require that image analysis is performed on the database images in order to extract salient perceptual features that are matched with those of the example. However, usage of 2D examples is generally inadequate as effective authoring of query images, attaining a realistic reproduction of complex scenes, needs manual editing and sketching ability. Investigation of new querying paradigms is therefore an important¿yet still marginally investigated¿factor for the success of content-based image retrieval. In this paper, a novel querying paradigm is presented which is based on usage of 3D interfaces exploiting navigation and editing of 3D virtual environments. Query images are obtained by taking a snapshot of the framed environment and by using the snapshot as an example to retrieve similar database images. A comparative analysis is carried out between the usage of 3D and 2D interfaces and their related query paradigms. This analysis develops on a user test on retrieval efficiency and effectiveness, as well as on an evaluation of users' satisfaction."
IEEE Trans. Vis. Comput. Graph.	"Guest Editor's Introduction: Special issue on IEEE Visualization."	Kenneth I. Joy	2002	"Guest Editor's Introduction: Special issue on IEEE Visualization."
IEEE Trans. Vis. Comput. Graph.	Designing Effective Transfer Functions for Volume Rendering from Photographic Volumes.	David S. Ebert,Christopher J. Morris,Penny Rheingans,Terry S. Yoo	2002	Photographic volumes present a unique, interesting challenge for volume rendering. In photographic volumes, voxel color is predetermined, making color selection through transfer functions unnecessary. However, photographic data does not contain a clear mapping from the multivalued color values to a scalar density or opacity, making projection and compositing much more difficult than with traditional volumes. Moreover, because of the nonlinear nature of color spaces, there is no meaningful norm for the multivalued voxels. Thus, the individual color channels of photographic data must be treated as incomparable data tuples rather than as vector values. Traditional differential geometric tools, such as intensity gradients, density, and Laplacians, are distorted by the nonlinear nonorthonormal color spaces that are the domain of the voxel values. We have developed different techniques for managing these issues while directly rendering volumes from photographic data. We present and justify the normalization of color values by mapping RGB values to the CIE L^\ast u^\ast v^\ast color space. We explore and compare different opacity transfer functions that map three channel color values to opacity. We apply these many-to-one mappings to the original RGB values as well as to the voxels after conversion to L^\ast u^\ast v^\ast space. Direct rendering using transfer functions allows us to explore photographic volumes without having to commit to an a priori segmentation that might mask fine variations of interest. We empirically compare the combined effects of each of the two color spaces with our opacity transfer functions using source data from the Visible Human Project.
IEEE Trans. Vis. Comput. Graph.	The Classification of Volumetric Display Systems: Characteristics and Predictability of the Image Space.	Barry G. Blundell,Adam J. Schwarz	2002	A diverse range of volumetric display systems has been proposed during the last 90 years. In order to facilitate a comparison between the various approaches, the three subsystems that comprise displays of this type are identified and are used as a basis for a classification scheme. The general characteristics of a number of volumetric display system configurations are examined, with emphasis given to issues relating to the predictability of the volume within which images are depicted. Key characteristics of this image space are identified and the complex manner in which they depend upon the display unit subsystems are illustrated for several current volumetric display techniques.
IEEE Trans. Vis. Comput. Graph.	The Floating Column Algorithm for Shaded, Parallel Display of Function Surfaces without Patches.	Dan Gordon	2002	"The floating column algorithm is a new method for the shaded rendering of function surfaces. Derived from the monochromatic floating horizon algorithm, it uses the partial derivatives of the function to compute surface normals, thus enabling intensity or normal-interpolation shading. Current rendering methods require tiling the surface with patches, so higher resolution patching is required for zoom-in views or interactive modification or time-varying surfaces. The new algorithm requires no patching and uses only constant space, so it can be implemented on graphics cards and hand-held devices. Each pixel-column is displayed independently of the others, and this ¿independent column mode¿ makes the algorithm inherently parallel in image-space, so it is suitable for multiprocessor workstations and clusters and it is scalable in the resolution size. Furthermore, the sampling frequency of the surface can be controlled locally, matching local surface features, distance, or artifact elimination requirements. Space-efficient supersampling for antialiasing is also possible. The new algorithm, which allows orthogonal and perspective projections, produces pixel-wide strips which can be displayed in software or hardware. Various extensions are described, including shadows and texture mapping. These properties, together with the algorithm's parallelism, make it potentially useful for the real-time display of functionally defined textured terrains and the animated display of time-varying surfaces."
IEEE Trans. Vis. Comput. Graph.	Vertex Data Compression through Vector Quantization.	Peter H. Chou,Teresa H. Y. Meng	2002	Rendering geometrically detailed 3D models requires the transfer and processing of large amounts of triangle and vertex geometry data. Compressing the geometry bitstream can reduce bandwidth requirements and alleviate transmission bottlenecks. In this paper, we show vector quantization to be an effective compression technique for triangle mesh vertex data. We present predictive vector quantization methods using unstructured codebooks as well as a product code pyramid vector quantizer. The technique is compatible with most existing mesh connectivity encoding schemes and does not require the use of entropy coding. In addition to compression, our vector quantization scheme can be used for complexity reduction by accelerating the computation of linear vertex transformations. Consequently, an encoded set of vertices can be both decoded and transformed in approximately 60 percent of the time required by a conventional method without compression.
IEEE Trans. Vis. Comput. Graph.	Stereoscopic View-Dependent Visualization of Terrain Height Fields.	Ugur Güdükbay,Türker Yilmaz	2002	Visualization of large geometric environments has always been an important problem of computer graphics. In this paper, we present a framework for the stereoscopic view-dependent visualization of large scale terrain models. We use a quadtree based multiresolution representation for the terrain data. This structure is queried to obtain the view-dependent approximations of the terrain model at different levels of detail. In order not to lose depth information, which is crucial for the stereoscopic visualization, we make use of a different simplification criterion, namely, distance-based angular error threshold. We also present an algorithm for the construction of stereo pairs in order to speed up the view-dependent stereoscopic visualization. The approach we use is the simultaneous generation of the triangles for two stereo images using a single draw-list so that the view frustum culling and vertex activation is done only once for each frame. The cracking problem is solved using the dependency information stored for each vertex. We eliminate the popping artifacts that can occur while switching between different resolutions of the data using morphing. We implemented the proposed algorithms on personal computers and graphics workstations. Performance experiments show that the second eye image can be produced approximately 45 percent faster than drawing the two images separately and a smooth stereoscopic visualization can be achieved at interactive frame rates using continuous multiresolution representation of height fields.
IEEE Trans. Vis. Comput. Graph.	Dynamic Particle Coating.	Arash Habibi,Annie Luciani	2002	Physically-based particle models are used by an increasing community of computer graphics researchers and users in order to produce a large variety of dynamic motions. Among all of the methods dedicated to the coating of point models, the implicit surface method has proven to be one of the most powerful. However, for the visualization of a wide variety of objects ranging from smoke to solids, the time-independent coating of traditional implicit surfaces appears to be dynamically too poor and restrictive. We propose a generalization of classic implicit surfaces able to produce a larger variety of particle coatings, from rigid solids to highly deformable objects and even wave propagation and fluid flow coatings, thus handling all these disparate categories with the same paradigm. The method consists of extracting the coating from a field function which is not predetermined but calculated as the modulation of a dynamic discrete medium by particles. For these reasons, the coating behaviors present higher-order dynamic behaviors closely correlated with the dynamics of skeleton particles.
IEEE Trans. Vis. Comput. Graph.	Multidimensional Transfer Functions for Interactive Volume Rendering.	Joe Kniss,Gordon L. Kindlmann,Charles D. Hansen	2002	Most direct volume renderings produced today employ one-dimensional transfer functions which assign color and opacity to the volume based solely on the single scalar quantity which comprises the data set. Though they have not received widespread attention, multidimensional transfer functions are a very effective way to extract materials and their boundaries for both scalar and multivariate data. However, identifying good transfer functions is difficult enough in one dimension, let alone two or three dimensions. This paper demonstrates an important class of three-dimensional transfer functions for scalar data, and describes the application of multidimensional transfer functions to multivariate data. We present a set of direct manipulation widgets that make specifying such transfer functions intuitive and convenient. We also describe how to use modern graphics hardware to both interactively render with multidimensional transfer functions and to provide interactive shadows for volumes. The transfer functions, widgets, and hardware combine to form a powerful system for interactive volume exploration.
IEEE Trans. Vis. Comput. Graph.	"Editor's Note."	Hans Hagen,David S. Ebert	2002	"Editor's Note."
IEEE Trans. Vis. Comput. Graph.	Interactive Visualization of State Transition Systems.	Frank van Ham,Huub van de Wetering,Jarke J. van Wijk	2002	A new method for the visualization of state transition systems is presented. Visual information is reduced by clustering nodes, forming a tree structure of related clusters. This structure is visualized in three dimensions with concepts from cone trees and emphasis on symmetry. A number of interactive options are provided as well, allowing the user to superimpose detail information on this tree structure. The resulting visualization enables the user to relate features in the visualization of the state transition graph to semantic concepts in the corresponding process and vice versa.
IEEE Trans. Vis. Comput. Graph.	A Flexible Approach for Visual Data Mining.	Matthias Kreuseler,Heidrun Schumann	2002	The exploration of heterogenous information spaces requires suitable mining methods as well as effective visual interfaces. Most of the existing systems concentrate either on mining algorithms or on visualization techniques. This paper describes a flexible framework for Visual Data Mining which combines analytical and visual methods to achieve a better understanding of the information space. We provide several preprocessing methods for unstructured information spaces such as a flexible hierarchy generation with user controlled refinement. Moreover, we develop new visualization techniques including an intuitive Focus+Context technique to visualize complex hierarchical graphs. A special feature of our system is a new paradigm for visualizing information structures within their frame of reference.
IEEE Trans. Vis. Comput. Graph.	ThemeRiver: Visualizing Thematic Changes in Large Document Collections.	Susan Havre,Elizabeth G. Hetzler,Paul Whitney,Lucy T. Nowell	2002	"The ThemeRiver visualization depicts thematic variations over time within a large collection of documents. The thematic changes are shown in the context of a time line and corresponding external events. The focus on temporal thematic change within a context framework allows a user to discern patterns that suggest relationships or trends. For example, the sudden change of thematic strength following an external event may indicate a causal relationship. Such patterns are not readily accessible in other visualizations of the data. We use a river metaphor to convey several key notions. The document collection's time line, selected thematic content, and thematic strength are indicated by the river's directed flow, composition, and changing width, respectively. The directed flow from left to right is interpreted as movement through time and the horizontal distance between two points on the river defines a time interval. At any point in time, the vertical distance, or width, of the river indicates the collective strength of the selected themes. Colored ¿currents¿ flowing within the river represent individual themes. A current's vertical width narrows or broadens to indicate decreases or increases in the strength of the individual theme."
IEEE Trans. Vis. Comput. Graph.	General Construction of Time-Domain Filters for Orientation Data.	Jehee Lee,Sung Yong Shin	2002	Capturing live motion has gained considerable attention in computer animation as an important motion generation technique. Canned motion data are comprised of both position and orientation components. Although a great number of signal processing methods are available for manipulating position data, the majority of these methods cannot be generalized easily to orientation data due to the inherent nonlinearity of the orientation space. In this paper, we present a new scheme that enables us to apply a filter mask (or a convolution filter) to orientation data. The key idea is to transform the orientation data into their analogues in a vector space, to apply a filter mask on them, and then to transform the results back to the orientation space. This scheme gives time-domain filters for orientation data that are computationally efficient and satisfy such important properties as coordinate-invariance, time-invariance, and symmetry. Experimental results indicate that our scheme is useful for various purposes including smoothing and sharpening.
IEEE Trans. Vis. Comput. Graph.	Terrain Simplification Simplified: A General Framework for View-Dependent Out-of-Core Visualization.	Peter Lindstrom,Valerio Pascucci	2002	This paper describes a general framework for out-of-core rendering and management of massive terrain surfaces. The two key components of this framework are: view-dependent refinement of the terrain mesh and a simple scheme for organizing the terrain data to improve coherence and reduce the number of paging events from external storage to main memory. Similar to several previously proposed methods for view-dependent refinement, we recursively subdivide a triangle mesh defined over regularly gridded data using longest-edge bisection. As part of this single, per-frame refinement pass, we perform triangle stripping, view frustum culling, and smooth blending of geometry using geomorphing. Meanwhile, our refinement framework supports a large class of error metrics, is highly competitive in terms of rendering performance, and is surprisingly simple to implement. Independent of our refinement algorithm, we also describe several data layout techniques for providing coherent access to the terrain data. By reordering the data in a manner that is more consistent with our recursive access pattern, we show that visualization of gigabyte-size data sets can be realized even on low-end, commodity PCs without the need for complicated and explicit data paging techniques. Rather, by virtue of dramatic improvements in multilevel cache coherence, we rely on the built-in paging mechanisms of the operating system to perform this task. The end result is a straightforward, simple-to-implement, pointerless indexing scheme that dramatically improves the data locality and paging performance over conventional matrix-based layouts.
IEEE Trans. Vis. Comput. Graph.	Lagrangian-Eulerian Advection of Noise and Dye Textures for Unsteady Flow Visualization.	Bruno Jobard,Gordon Erlebacher,M. Yousuff Hussaini	2002	A new hybrid scheme (LEA) that combines the advantages of Eulerian and Lagrangian frameworks is applied to the visualization of dense representations of time-dependent vector fields. The algorithm encodes the particles into a texture that is then advected. By treating every particle equally, we can handle texture advection and dye advection within a single framework. High temporal and spatial correlation is achieved through the blending of successive frames. A combination of particle and dye advection enables the simultaneous visualization of streamlines, particle paths, and streaklines. We demonstrate various experimental techniques on several physical flow fields. The simplicity of both the resulting data structures and the implementation suggest that LEA could become a useful component of any scientific visualization toolkit concerned with the display of unsteady flows.
IEEE Trans. Vis. Comput. Graph.	A Hardware-Assisted Scalable Solution for Interactive Volume Rendering of Time-Varying Data.	Eric B. Lum,Kwan-Liu Ma,John Clyne	2002	We present a scalable volume rendering technique that exploits lossy compression and low-cost commodity hardware to permit highly interactive exploration of time-varying scalar volume data. A palette-based decoding technique and an adaptive bit allocation scheme are developed to fully utilize the texturing capability of a commodity 3D graphics card. Using a single PC equipped with a modest amount of memory, a texture capable graphics card, and an inexpensive disk array, we are able to render hundreds of time steps of regularly gridded volume data (up to 42 million voxels each time step) at interactive rates. By clustering multiple PCs together, we demonstrate the data-size scalability of our method. The frame rates achieved make possible the interactive exploration of data in the temporal, spatial, and transfer function domains. A comprehensive evaluation of our method based on experimental studies using data sets (up to 134 million voxels per time step) from turbulence flow simulations is also presented.
IEEE Trans. Vis. Comput. Graph.	Information Visualization and Visual Data Mining.	Daniel A. Keim	2002	Never before in history has data been generated at such high volumes as it is today. Exploring and analyzing the vast volumes of data is becoming increasingly difficult. Information visualization and visual data mining can help to deal with the flood of information. The advantage of visual data exploration is that the user is directly involved in the data mining process. There are a large number of information visualization techniques which have been developed over the last decade to support the exploration of large data sets. In this paper, we propose a classification of information visualization and visual data mining techniques which is based on the data type to be visualized, the visualization technique, and the interaction and distortion technique. We exemplify the classification using a few examples, most of them referring to techniques and systems presented in this special section.
IEEE Trans. Vis. Comput. Graph.	Hierarchical Pixel Bar Charts.	Daniel A. Keim,Ming C. Hao,Umeshwar Dayal	2002	Simple presentation graphics are intuitive and easy-to-use, but only show highly aggregated data. Bar charts, for example, only show a rather small number of data values and x-y-plots often have a high degree of overlap. Presentation techniques are often chosen depending on the considered data type--bar charts, for example, are used for categorical data and x-y plots are used for numerical data. In this article, we propose a combination of traditional bar charts and x-y-plots, which allows the visualization of large amounts of data with categorical and numerical data. The categorical data dimensions are used for the partitioning into the bars and the numerical data dimensions are used for the ordering arrangement within the bars. The basic idea is to use the pixels within the bars to present the detailed information of the data records. Our so-called pixel bar charts retain the intuitiveness of traditional bar charts while applying the principle of x-y charts within the bars. In many applications, a natural hierarchy is defined on the categorical data dimensions such as time, region, or product type. In hierarchical pixel bar charts, the hierarchy is exploited to split the bars for selected portions of the hierarchy. Our application to a number of real-world e-business and Web services data sets shows the wide applicability and usefulness of our new idea.
IEEE Trans. Vis. Comput. Graph.	Adding Support for High-Level Skeletal Animation.	Francisco J. Serón,Rafael Rodriguez,Eva Cerezo,Alfredo Pina	2002	We hereby present a data structure specially geared toward the definition and management of synthetic actors in real-time computer graphics. The relation between our proposed data structure and the Silicon Graphics API Performer¯ makes its implementation possible on a low-cost real-time platform thanks to current accelerating cards. We demonstrate how our data structure is used to generate motion by means of two different applications. Both of them make use of direct (DK) and inverse kinematics (IK) and may use motion capture. ARTgraph is a development environment devoted to the creation of high-quality real-time 3D-graphics applications (basically, 3D games) and the ALVW system is a general platform that provides and coordinates a sensing-analysis-acting loop to provide behavior for synthetic actors in their own scenario. The aim of this paper is to contribute to the standardization process of multiplatform synthetic actor programs or libraries.
IEEE Trans. Vis. Comput. Graph.	Polaris: A System for Query, Analysis, and Visualization of Multidimensional Relational Databases.	Chris Stolte,Diane Tang,Pat Hanrahan	2002	In the last several years, large multidimensional databases have become common in a variety of applications such as data warehousing and scientific computing. Analysis and exploration tasks place significant demands on the interfaces to these databases. Because of the size of the data sets, dense graphical representations are more effective for exploration than spreadsheets and charts. Furthermore, because of the exploratory nature of the analysis, it must be possible for the analysts to change visualizations rapidly as they pursue a cycle involving first hypothesis and then experimentation. In this paper, we present Polaris, an interface for exploring large multidimensional databases that extends the well-known Pivot Table interface. The novel features of Polaris include an interface for constructing visual specifications of table-based graphical displays and the ability to generate a precise set of relational queries from the visual specifications. The visual specifications can be rapidly and incrementally developed, giving the analyst visual feedback as they construct complex queries and visualizations.
IEEE Trans. Vis. Comput. Graph.	Spatial-Temporal Antialiasing.	Kelvin Sung,Andrew Pearce,Changyaw Wang	2002	A framework for discussing the motion blur image generation process is formulated. Previous work is studied in the context of the framework. Due to the implicit assumptions on low temporal frequencies in most motion blur algorithms, issues involved in large screen space movements and fast illumination changes in time have not been adequately addressed so far. A new approach that does not make these assumptions is introduced to solve the spatial-temporal geometric and shading aliasing problems separately. Based on newly developed adaptive algorithms in the spatial-temporal domain, an implementation of the new approach is developed to efficiently deliver high quality motion blurred images in general computer graphics production environments.
IEEE Trans. Vis. Comput. Graph.	VideoPlus: A Method for Capturing the Structure and Appearance of Immersive Environments.	Camillo J. Taylor	2002	VideoPlus: A Method for Capturing the Structure and Appearance of Immersive Environments.
IEEE Trans. Vis. Comput. Graph.	A Geometric Comparison of Algorithms for Fusion Control in Stereoscopic HTDs.	Zachary Wartell,Larry F. Hodges,William Ribarsky	2002	"This paper concerns stereoscopic virtual reality displays in which the head is tracked and the display is stationary, attached to a desk, tabletop, or wall. These are called stereoscopic HTDs (Head-Tracked Display). Stereoscopic displays render two perspective views of a scene, each of which is seen by one eye of the user. Ideally, the user's natural visual system combines the stereo image pair into a single, 3D perceived image. Unfortunately, users often have difficulty fusing the stereo image pair. Researchers use a number of software techniques to reduce fusion problems. This paper geometrically examines and compares a number of these techniques and reaches the following conclusions: In interactive stereoscopic applications, the combination of view placement, scale, and either false eye separation or \alpha{\hbox{-}}{\rm false} eye separation can provide fusion control geometrically similar to image shifting and image scaling. However, in stereo HTDs, image shifting and image scaling also generate additional geometric artifacts not generated by the other methods. We anecdotally link some of these artifacts to exceeding perceptual limitations of human vision. While formal perceptual studies are still needed, geometric analysis suggests that image shifting and image scaling may be less appropriate than the other methods for interactive, stereo HTDs."
IEEE Trans. Vis. Comput. Graph.	Texture Mapping Using Surface Flattening via Multidimensional Scaling.	Gil Zigelman,Ron Kimmel,Nahum Kiryati	2002	We present a novel technique for texture mapping on arbitrary surfaces with minimal distortions by preserving the local and global structure of the texture. The recent introduction of the fast marching method on triangulated surfaces made it possible to compute a geodesic distance map from a given surface point in O( n \lg n) operations, where n is the number of triangles that represent the surface. We use this method to design a surface flattening approach based on multidimensional scaling (MDS). MDS is a family of methods that map a set of points into a finite dimensional flat (Euclidean) domain, where the only given data is the corresponding distances between every pair of points. The MDS mapping yields minimal changes of the distances between the corresponding points. We then solve an inverse problem and map a flat texture patch onto the curved surface while preserving the structure of the texture.
IEEE Trans. Vis. Comput. Graph.	Robust Creation of Implicit Surfaces from Polygonal Meshes.	Gary D. Yngve,Greg Turk	2002	"Implicit surfaces are used for a number of tasks in computer graphics, including modeling soft or organic objects, morphing, collision detection, and constructive solid geometry. Although operating on implicit surfaces is usually straightforward, creating them is not. We introduce a practical method for creating implicit surfaces from polygonal models that produces high-quality results for complex surfaces. Whereas much previous work in implicit surfaces has been done with primitives such as ""blobbies,"" we use implicit surfaces based on a variational interpolation technique (the three-dimensional generalization of thin-plate interpolation). Given a polygonal mesh, we convert the data to a volumetric representation to use as a guide for creating the implicit surface iteratively. We begin by seeding the surface with a number of constraint points through which the surface must pass. Iteratively, additional constraints are added; the resulting surfaces are evaluated, and the errors guide the placement of subsequent constraints. We have applied our method successfully to a variety of polygonal meshes and consider it to be robust."
IEEE Trans. Vis. Comput. Graph.	EWA Splatting.	Matthias Zwicker,Hanspeter Pfister,Jeroen van Baar,Markus H. Gross	2002	"In this paper, we present a framework for high quality splatting based on elliptical Gaussian kernels. To avoid aliasing artifacts, we introduce the concept of a resampling filter, combining a reconstruction kernel with a low-pass filter. Because of the similarity to Heckbert's EWA (elliptical weighted average) filter for texture mapping, we call our technique EWA splatting. Our framework allows us to derive EWA splat primitives for volume data and for point-sampled surface data. It provides high image quality without aliasing artifacts or excessive blurring for volume data and, additionally, features anisotropic texture filtering for point-sampled surfaces. It also handles nonspherical volume kernels efficiently; hence, it is suitable for regular, rectilinear, and irregular volume datasets. Moreover, our framework introduces a novel approach to compute the footprint function, facilitating efficient perspective projection of arbitrary elliptical kernels at very little additional cost. Finally, we show that EWA volume reconstruction kernels can be reduced to surface reconstruction kernels. This makes our splat primitive universal in rendering surface and volume data."
CVPR	2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2003), 16-22 June 2003, Madison, WI, USA		2003	2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2003), 16-22 June 2003, Madison, WI, USA
ICCV	Globally Convergent Autocalibration.	Arrigo Benedetti,Alessandro Busti,Michela Farenzena,Andrea Fusiello	2003	Existing autocalibration techniques use numerical optimizationalgorithms that are prone to the problem of localminima. To address this problem, we have developeda method where an interval branch-and-bound method isemployed for numerical minimization. Thanks to the propertiesof Interval Analysis this method is guaranteed to convergeto the global solution with mathematical certainty andarbitrary accuracy, and the only input information it requiresfrom the user is a set of point correspondences anda search box. The cost function is based on the Huang-Faugerasconstraint of the fundamental matrix. A recentlyproposed interval extension based on Bernstein polynomialforms has been investigated to speed up the search for thesolution. Finally, some experimental results on syntheticimages are presented.
ICCV	Towards Direct Recovery of Shape and Motion Parameters from Image Sequences.	Stephen Benoit,Frank P. Ferrie	2003	A novel procedure is presented to construct image-domainfilters (receptive fields) that directly recover localmotion and shape parameters. These receptive fields arederived from training on image deformations that best discriminatebetween different shape and motion parameters.Beginning with the construction of 1-D receptive fieldsthat detect local surface shape and motion parameterswithin cross sections, we show how the recovered shape andmotion model parameters are sufficient to produce local estimatesof time to collision.In general, filter pairs (receptive fields) can be synthesizedto perform or detect specific image deformations. Atthe heart of the method is the use of a matrix to representimage deformation correspondence between individual pixelsof two views of a surface. The image correspondencematrix can be decomposed using Singular Value Decompositionto yield a pair of corresponding receptive fields thatdetect image changes due to the deformation of interest.
ICCV	Camera calibration using spheres: A semi-definite programming approach.	Motilal Agrawal,Larry S. Davis	2003	Vision algorithms utilizing camera networks with a commonfield of view are becoming increasingly feasible andimportant. Calibration of such camera networks is a challengingand cumbersome task. The current approaches forcalibration using planes or a known 3D target may not befeasible as these objects may not be simultaneously visiblein all the cameras. In this paper, we present a new algorithmto calibrate cameras using occluding contours of spheres.In general, an occluding contour of a sphere projects to anellipse in the image. Our algorithm uses the projection ofthe occluding contours of three spheres and solves for theintrinsic parameters and the locations of the spheres. Theproblem is formulated in the dual space and the parametersare solved for optimally and efficiently using semi-definiteprogramming. The technique is flexible, accurate and easyto use. In addition, since the contour of a sphere is simultaneouslyvisible in all the cameras, our approach cangreatly simplify calibration of multiple cameras with a commonfield of view. Experimental results from computer simulateddata and real world data, both for a single cameraand multiple cameras, are presented.
ICCV	Voxel Carving for Specular Surfaces.	Thomas Bonfort,Peter F. Sturm	2003	We present an novel algorithm that reconstructs voxels of a general 3D specular surface from multiple images of a calibrated camera. A calibrated scene (i.e. points whose 3D coordinates are known) is reflected by the unknown specular surface onto the image plane of the camera. For every viewpoint, surface normals are associated to the voxels traversed by each projection ray formed by the reflection of a scene point. A decision process then discards voxels whose associated surface normals are not consistent with one another. The output of the algorithm is a collection of voxels and surface normals in 3D space, whose quality and size depend on user-set thresholds. The method has been tested on synthetic and real images. Visual and quantified experimental results are presented.
ICCV	Computing Geodesics and Minimal Surfaces via Graph Cuts.	Yuri Boykov,Vladimir Kolmogorov	2003	Geodesic active contours and graph cuts are two standard image segmentation techniques. We introduce a new segmentation method combining some of their benefits. Our main intuition is that any cut on a graph embedded in some continuous space can be interpreted as a contour (in 2D) or a surface (in 3D). We show how to build a grid graph and set its edge weights so that the cost of cuts is arbitrarily close to the length (area) of the corresponding contours(surfaces) for any anisotropic Riemannian metric. There are two interesting consequences of this technical result. First, graph cut algorithms can be used to find globally minimum geodesic contours (minimal surfaces in 3D) under arbitrary Riemannian metric for a given set of boundary conditions. Second, we show how to minimize metrication artifacts in existing graph-cut based methods in vision. Theoretically speaking, our work provides an interesting link between several branches of mathematics - differential geometry, integral geometry, and combinatorial optimization. The main technical problem is solved using Cauchy-Crofton formula from integral geometry.
ICCV	What Does Motion Reveal About Transparency?	Moshe Ben-Ezra,Shree K. Nayar	2003	The perception of transparent objects from images is knownto be a very hard problem in vision. Given a single image,it is difficult to even detect the presence of transparent objectsin the scene. In this paper, we explore what can be saidabout transparent objects by a moving observer. We showhow features that are imaged through a transparent objectbehave differently from those that are rigidly attached to thescene. We present a novel model-based approach to recoverthe shapes and the poses of transparent objects from knownmotion. The objects can be complex in that they may be composedof multiple layers with different refractive indices. Wehave conducted numerous simulations to verify the practicalfeasibility of our algorithm. We have applied it to real scenesthat include transparent objects and recovered the shapes ofthe objects with high accuracy.
ICCV	Surface Classification Using Conformal Structures.	Xianfeng Gu,Shing-Tung Yau	2003	3D surface classification is a fundamental problem incomputer vision and computational geometry. Surfaces canbe classified by different transformation groups. Traditionalclassification methods mainly use topological transformationgroups and Euclidean transformation groups. This paperintroduces a novel method to classify surfaces by conformaltransformation groups. Conformal equivalent classis refiner than topological equivalent class and coarser thanisometric equivalent class, making it suitable for practicalclassification purposes. For general surfaces, the gradientfields of conformal maps form a vector space, which hasa natural structure invariant under conformal transformations.We present an algorithm to compute this conformalstructure, which can be represented as matrices, and use itto classify surfaces. The result is intrinsic to the geometry,invariant to triangulation and insensitive to resolution. Tothe best of our knowledge, this is the first paper to classifysurfaces with arbitrary topologies by global conformal invariants.The method introduced here can also be used forsurface matching problems.
ICCV	Graph Partition by Swendsen-Wang Cuts.	Adrian Barbu,Song Chun Zhu	2003	Vision tasks, such as segmentation, grouping, recognition, can be formulated as graph partition problems. The recent literature witnessed two popular graph cut algorithms: the Ncut using spectral graph analysis and the minimum-cut using the maximum flow algorithm. This paper presents a third major approach by generalizing the Swendsen-Wang method- a well celebrated algorithm in statistical mechanics. Our algorithm simulates ergodic, reversible Markov chain jumps in the space of graph partitions to sample a posterior probability. At each step, the algorithm splits, merges, or re-groups a sizable subgraph, and achieves fast mixing at low temperature enabling a fast annealing procedure. Experiments show it converges in 2-30seconds in a PC for image segmentation. This is 400 times faster than the single-site update Gibbs sampler, and 20-40 times faster than the DDMCMC algorithm. The algorithm can optimize over the number of models and works for general forms of posterior probabilities, so it is more general than the existing graph cut approaches.
ICCV	Paracatadioptric Camera Calibration Using Lines.	João P. Barreto,Helder Araújo	2003	Paracatadioptric sensors combine a parabolic shaped mirrorand a camera inducing an orthographic projection.Such a configuration provides a wide field of view whilekeeping a single effective viewpoint. Previous work in centralcatadioptric sensors proved that a line projects into aconic curve and that three line images are enough to calibratethe system. However the estimation of the coniccurves where lines are mapped is hard to accomplish. Ingeneral only a small arc of the conic is visible in the imageand conventional conic fitting techniques are unable to correctlyestimate the curve. The present work shows that a setof conic curves corresponds to paracatadioptric line imagesif, and only if, certain properties are verified. These propertiesare used to constraint the search space and correctlyestimate the curves. The accurate estimation of a minimumof three line images allows the complete calibration of theparacatadioptric camera. If the camera is skewless and theaspect ratio is known then the conic fitting problem is solvednaturally by an eigensystem. For the general situation theconic curves are estimated using non-linear optimization.
ICCV	Recognising Panoramas.	Matthew Brown,David G. Lowe	2003	"The problem considered in this paper is the fully asutomaticconstruction of panoramas.Fundamentally, thisproblem requires recognition, as we need to know whichparts of the panorama join up.Previous approaches haveused human input or restrictions on the image sequencefor the matching step.In this work we use object recognitiontechniques based on invariant local features to selectmatchings images, and a probabilistic model for verification.Because of this our method is insensitive to the ordering, orientation, scale and illumination of the images.It is also insensitive to 'noise' images which are not partof the panorama at all, that is, it recognises panoramas.This suggests a useful application for photographers: thesystem takes as input the images on an entire flash card orfilm, recognises images that form part of a panorama, andstitches them with no user input whatsoever."
ICCV	Towards Gauge Invariant Bundle Adjustment: A Solution Based on Gauge Dependent Damping.	Adrien Bartoli	2003	Bundle ajustment is used to obtain accurate visual reconstructionsby minimizing the reprojection error. The coordinateframe ambiguity, or more generality the gauge freedoms,has been dealt with in different manners. It has oftenbeen reported that standard bundle adjustment algorithmswere not gauge invariant: two iterations within differentgauges can lead to geometrically very different results. Surprisingly,most algorithms do not exploit gauge freedoms toimprove performances. We consider this issue. We analyzetheoretically the impact of the gauge on standard algorithms.We show that a sufficiently general damping matrixin Levenberg-Marquardt iteration can be used to implicitlyreproduce a gauge transformation. We show that if thedamping matrix is chosen such that the decrease in the reprojectionerror is maximized, then the iteration is gauge invariant.Experimental results on simulated and real data showthat our gauge invariant bundle adjustment algorithm outperformsexisting ones in terms of stability.
ICCV	Multiple-View Structure and Motion From Line Correspondences.	Adrien Bartoli,Peter F. Sturm	2003	We address the problem of camera motion and structure reconstruction from line correspondences across multiple views, from initialization to final bundle adjustment. One of the main difficulties when dealing with line features is their algebraic representation. First, we consider the triangulation problem. Based on Pl&uuml;cker coordinates to represent the lines, we propose a maximum likelihood algorithm, relying on linearising the Pl&uuml;cker constraint, and on a Pl&uuml;cker correction procedure to compute the closest Pl&uuml;cker coordinates to a given 6-vector. Second, we consider the bundle adjustment problem. Previous overparameterizations of 3D lines induce gauge freedoms and/or internal consistency constraints. We propose the orthonormal representation, which allows handy non-linear optimization of 3D lines using the minimum 4 parameters, within an unconstrained non-linear optimizer. We compare our algorithms to existing ones on simulated and real data.
ICCV	A Caratheodory-Fejer Approach to Robust Multiframe Tracking.	Octavia I. Camps,Hwasup Lim,Cecilia Mazzaro,Mario Sznaier	2003	A requirement common to most dynamic vision applicationsis the ability to track objects in a sequence of frames. Thisproblem has been extensively studied in the past few years,leading to several techniques, such as Unscented ParticleFilter based trackers, that exploit a combination of the (assumed)target dynamics, empirically learned noise distributionsand past position observations. While successful inmany scenarios, these trackers remain fragile to occlusionand model uncertainty in the target dynamics. As we showin this paper, these difficulties can be addressed by modelingthe dynamics of the target as an unknown operator thatsatisfies certain interpolation conditions. Results from interpolationtheory can then be used to find this operator bysolving a convex optimization problem. As illustrated withseveral examples, combining this operator with Kalman andUPF techniques leads to both robustness improvement andcomputational complexity reduction.
ICCV	Good continuations in digital image level lines.	Frédéric Cao	2003	We propose a probabilistic algorithm able to detect the curves that are unexpectedy smooth in a set of digital curves. The only parameter is a false alarme rate, influencing the detection only by its logarithm. We experiment the good continuation criterion on image level lines. One of the conclusion is that, accordingly to Gestalt Theory, onecan detect egdes in a way that is widely independent of contrast. We also use the same kind of method to detect corners and junctions.
ICCV	A Cylindrical Surface Model to Rectify the Bound Document Image.	Huaigu Cao,Xiaoqing Ding,Changsong Liu	2003	This article proposes a novel approach on how torectify the photo image of the bound document. Thesurface of the document is modeled by a cylindricalsurface. By the geometry of camera image formation, theequations using the cue of directrixes to map the points onthe surface in the 3-D scene to the points on the imageplane are achieved. Baselines of the horizontal text lineare extracted as projections of directrixes to estimate thebending extent of the surface, and then the images arerectified. The proposed method needs no auxiliary device.Experimental results are presented to demonstrate thefeasibility and the application of the method.
ICCV	A Model-Based Approach for Automated Feature Extraction in Fundus Images.	Huiqi Li,Opas Chutatape	2003	A new approach to automatically extract the main features in color fundus images are proposed in this paper. Optic disk is localized by the principal component analysis (PCA) and its shape is detected by a modified active shape model (ASM). Exudates are extracted by the combined region growing and edge detection. A fundus coordinate system is further set up based on the fovea localization to provide a better description of the features in fundus images. The success rates achieved are 99%, 94%, and 100% for disk localization, disk boundary detection, and fovea localization respectively. The sensitivity and specificity for exudate detection are 100% and 71%. The success of the proposed algorithms can be attributed to the utilization of the model-based methods.
ICCV	Using prior shape and intensity profile in medical image segmentation.	Yunmei Chen,Feng Huang,Hemant Tagare,Murali Rao,David Clifford Wilson,Edward A. Geiser	2003	"In this note we present a coupled optimization model forboundary determination. One part of the model incorporatesa prior shape into a geometric active contour modelwith a fixed parameter. The second part determines the'best' parameter used in the first part by maximizing the mutualinformation of the image geometry between the priorand an aligned novel image over all the alignments, thatare the solutions of the first part corresponding to differentparameters. We also present an alternative method, whichgenerates an intensity model formed as the average of a setof aligned training images. Experimental results on cardiacultrasound images are presented. These results indicatethat the proposed model provides close agreement withexpert traced borders, and the parameter determined in thismodel for one image can be used for images with similarproperties. The existence of a solution to the proposed minimizationproblem is also discussed."
ICCV	Robust Regression with Projection Based M-estimators.	Haifeng Chen,Peter Meer	2003	The robust regression techniques in the RANSAC family arepopular today in computer vision, but their performance dependson a user supplied threshold. We eliminate this draw-backof RANSAC by reformulating another robust method,the M-estimator, as a projection pursuit optimization problem.The projection based pbM-estimator automatically derivesthe threshold from univariate kernel density estimates.Nevertheless, the performance of the pbM-estimator equalsor exceeds that of RANSAC techniques tuned to the optimalthreshold, a value which is never available in practice.Experiments were performed both with synthetic and realdata in the affine motion and fundamental matrix estimationtasks.
ICCV	Calibration of a Hybrid Camera Network.	Xilin Chen,Jie Yang,Alex Waibel	2003	"Visual surveillance using a camera network has imposed new challenges to camera calibration. An essential problem is that a large number of cameras may not have a common field of view or even be synchronized well. We propose to use a hybrid camera network that consists of catadioptric and perspective cameras for a visual surveillance task. The relations between multiple views of a scene captured from different cameras can be then calibrated under the catadioptric camera's coordinate system. This paper addresses the important issue of how to calibrate the hybrid camera network. We calibrate the hybrid camera network in three steps. First, we calibrate the catadioptric camera using only the vanishing points. In order to reduce computational complexity, we calibrate the camera without the mirror first and then calibrate the catadioptric camera system. Second, we determine 3D positions of some points usingas few as two spatial parallel lines and some equidistance points. Finally, we calibrate other perspective cameras based on these known spatial points."
ICCV	On-Line Selection of Discriminative Tracking Features.	Robert T. Collins,Yanxi Liu	2003	This paper presents a method for evaluating multiple feature spaces while tracking, and for adjusting the set of features used to improve tracking performance. Our hypothesisis that the features that best discriminate between object and background are also best for tracking the object. We develop an on-line feature selection mechanism based on the two-class variance ratio measure, applied to log likelihood distributions computed with respect to a given feature from samples of object and background pixels. This feature selection mechanism is embedded in a tracking system that adaptively selects the top-ranked discriminative features for tracking. Examples are presented to illustrate how the method adapts to changing appearances of both tracked object and scene background.
ICCV	Variational Space-Time Motion Segmentation.	Daniel Cremers,Stefano Soatto	2003	We propose a variational method for segmenting imagesequences into spatio-temporal domains of homogeneousmotion. To this end, we formulate the problem of motionestimation in the framework of Bayesian inference, using aprior which favors domain boundaries of minimal surfacearea. We derive a cost functional which depends on a surfacein space-time separating a set of motion regions, aswell as a set of vectors modeling the motion in each region.We propose a multiphase level set formulation of thisfunctional, in which the surface and the motion regions arerepresented implicitly by a vector-valued level set function.Joint minimization of the proposed functional results in aneigenvalue problem for the motion model of each region andin a gradient descent evolution for the separating interface.Numerical results on real-world sequences demonstratethat minimization of a single cost functional generates asegmentation of space-time into multiple motion regions.
ICCV	Gaze Manipulation for One-to-one Teleconferencing.	Antonio Criminisi,Jamie Shotton,Andrew Blake,Philip H. S. Torr	2003	A new algorithm is proposed for novel view generation in one-to-one teleconferencing applications. Given the video streams acquired by two cameras placed on either side of a computer monitor, the proposed algorithm synthesises images from a virtual camera in arbitrary position (typically located within the monitor) to facilitate eye contact. Our technique is based on an improved, dynamic-programming, stereo algorithm for efficient novel-view generation. The two main contributions of this paper are: i) a new type of three-plane graph for dense-stereo dynamic-programming, that encourages correct occlusion labeling; ii) a compact geometric derivation for novel-view synthesis by direct projection of the minimum-cost surface. Furthermore, this paper presents a novel algorithm for the temporal maintenance of a background model to enhance the rendering of occlusions and reduce temporal artefacts (flicker); and a cost aggregation algorithm that acts directly on our three-dimensional matching cost space. Examples are given that demonstrate the robustness of the new algorithm to spatial and temporal artefacts for long stereo video streams. These include demonstrations of synthesis of cyclopean views of extended conversational sequences. We further demonstrate synthesis from a freely translating virtual camera.
ICCV	Calibrating pan-tilt cameras in wide-area surveillance networks.	James Davis,Xing Chen	2003	Pan-tilt cameras are often used as components of wide-area surveillance systems. It is necessary to calibrate these cameras in relation to one another in order to obtain a consistent representation of the entire space. Existing methods for calibrating pan-tilt cameras have assumed an idealized model of camera mechanics. In addition, most methods have been calibrated using only asmall range of camera motion. This paper presents a method for calibrating pan-tilt cameras that introduces a more complete model of camera motion. Pan and tilt rotations are modeled as occurring around arbitrary axes in space. In addition, the wide area surveillance system itself is used to build a large virtual calibration object, resulting in better calibration than would be possible with a single small calibration target. Finally, the proposed enhancements are validated experimentally, with comparisons showing the improvement provided over more traditional methods.
ICCV	Recognizing Human Action Efforts: An Adaptive Three-Mode PCA Framework.	James W. Davis,Hui Gao	2003	"We present a computational framework capable of labelingthe effort of an action corresponding to the perceivedlevel of exertion by the performer (low - high). The approachinitially factorizes examples (at different efforts) ofan action into its three-mode principal components to reducethe dimensionality. Then a learning phase is introducedto compute expressive-feature weights to adjust themodel's estimation of effort to conform to given perceptuallabels for the examples. Experiments are demonstrated recognizingthe efforts of a person carrying bags of differentweight and for multiple people walking at different paces."
ICCV	Real-Time Simultaneous Localisation and Mapping with a Single Camera.	Andrew J. Davison	2003	Ego-motion estimation for an agile single camera movingthrough general, unknown scenes becomes a much morechallenging problem when real-time performance is requiredrather than under the off-line processing conditionsunder which most successful structure from motion workhas been achieved. This task of estimating camera motionfrom measurements of a continuously expanding set of self-mappedvisual features is one of a class of problems knownas Simultaneous Localisation and Mapping (SLAM) in therobotics community, and we argue that such real-time mappingresearch, despite rarely being camera-based, is morerelevant here than off-line structure from motion methodsdue to the more fundamental emphasis placed on propagationof uncertainty.We present a top-down Bayesian framework for single-cameralocalisation via mapping of a sparse set of naturalfeatures using motion modelling and an information-guidedactive measurement strategy, in particular addressingthe difficult issue of real-time feature initialisation viaa factored sampling approach. Real-time handling of uncertaintypermits robust localisation via the creating andactive measurement of a sparse map of landmarks such thatregions can be re-visited after periods of neglect and localisationcan continue through periods when few features arevisible. Results are presented of real-time localisation fora hand-waved camera with very sparse prior scene knowledgeand all processing carried out on a desktop PC.
ICCV	Constraining Human Body Tracking.	David Demirdjian,Teresa Ko,Trevor Darrell	2003	Our paper addresses the problem of enforcing constraintsin human body tracking. A projection technique is derivedto impose kinematic constraints on independent multi-bodymotion: we show that for small motions the multi-body articulatedmotion space can be approximated by a linearmanifold estimated directly from the previous body pose.We propose a learning approach to model non-linear constraints;we train a support vector classifier from motioncapture data to model the boundary of the space of validposes. Linear and non-linear body pose constraints are enforcedby first projecting unconstrained motions onto thearticulated motion space and then optimizing to find pointson this linear manifold that lie within the non-linear constraintsurface modeled by the SVM classifier.
ICCV	Information Theoretic Focal Length Selection for Real-Time Active 3-D Object Tracking.	Joachim Denzler,Matthias Zobel,Heinrich Niemann	2003	Active object tracking, for example, in surveillance tasks, becomes more and more important these days. Besides the tracking algorithms themselves methodologies have to be developed for reasonable active control of the degrees of freedom of all involved cameras. In this paper we present an information theoretic approach that allows the optimal selection of the focal lengths of two cameras during active 3-D object tracking. The selection is based on the uncertainty in the 3-D estimation. This allows us to resolve the trade-off between small and large focal length: in the former case, the chance is increased to keep the object in the field of view of the cameras. In the latter one, 3-D estimation becomes more reliable. Also, more details are provided, for example for recognizing the objects.Beyond a rigorous mathematical framework we present real-time experiments demonstrating that we gain an improvementin 3-D trajectory estimation by up to 42% in comparison with tracking using a fixed focal length.
ICCV	Markov-Based Failure Prediction for Human Motion Analysis.	Shiloh L. Dockstader,Nikita S. Imennov,A. Murat Tekalp	2003	This paper presents a new method of detecting andpredicting motion tracking failures with applications inhuman motion and gait analysis. We define a trackingfailure as an event and describe its temporal characteristicsusing a hidden Markov model (HMM). Thisstochastic model is trained using previous examples oftracking failures. We derive vector observations for theHMM using the noise covariance matrices characterizinga tracked, 3-D structural model of the human body. Weshow a causal relationship between the conditional outputprobability of the HMM, as transformed using alogarithmic mapping function, and impending trackingfailures. Results are illustrated on several multi-viewsequences of complex human motion.
ICCV	Active Concept Learning for Image Retrieval in Dynamic Databases.	Anlei Dong,Bir Bhanu	2003	Concept learning in content-based image retrieval (CBIR) systems is a challenging task. This paper presents an active concept learning approach based on mixture model to deal with the two basic aspects of a database system: changing (image insertion or removal) nature of a database and user queries. To achieve concept learning, we develop a novel model selection method based on Bayesian analysis that evaluates the consistency of hypothesized models with the available information. The analysis of exploitation vs. exploration in the search space helps to find optimal model efficiently. Experimental results on Corel database show the efficacy of our approach.
ICCV	Dynamic Texture Segmentation.	Gianfranco Doretto,Daniel Cremers,Paolo Favaro,Stefano Soatto	2003	We address the problem of segmenting a sequence of imagesof natural scenes into disjoint regions that are characterizedby constant spatio-temporal statistics. We model thespatio-temporal dynamics in each region by Gauss-Markovmodels, and infer the model parameters as well as theboundary of the regions in a variational optimization framework.Numerical results demonstrate that - in contrast topurely texture-based segmentation schemes - our method iseffective in segmenting regions that differ in their dynamicseven when spatial statistics are identical.
ICCV	Selection of Scale-Invariant Parts for Object Class Recognition.	Gyuri Dorkó,Cordelia Schmid	2003	This paper introduces a novel method for constructingand selecting scale-invariant object parts. Scale-invariantlocal descriptors are first grouped into basic parts. A classifieris then learned for each of these parts, and featureselection is used to determine the most discriminative ones.This approach allows robust part detection, and it is invariantunder scale change-that is, neither the training imagesnor the test images have to be normalized.The proposed method is evaluated in car detectiontasks with significant variations in viewing conditions, andpromising results are demonstrated. Different local regions,classifiers and feature selection methods are quantitativelycompared. Our evaluation shows that local invariantdescriptors are an appropriate representation for objectclasses such as cars, and it underlines the importance offeature selection.
ICCV	A Multi-scale Generative Model for Animate Shapes and Parts.	Aleksandr Dubinskiy,Song Chun Zhu	2003	This paper presents a multi-scale generative model for representing animate shapes and extracting meaningful parts of objects. The model assumes that animate shapes (2D simple closed curves) are formed by a linear superposition of a number of shape bases. These shapebases resemble the multi-scale Gabor bases in image pyramid representation, are well localized in both spatial and frequency domains, and form an over-complete dictionary. This model is simpler than the popular B-spline representation since it does not engage a domainpartition. Thus it eliminates the interference between adjacent B-spline bases, and becomes a true linear additive model. We pursue the bases by reconstructing the shape in a coarse-to-fine procedure through curve evolution. These shape bases are further organized ina tree-structure where the bases in each subtree sum up to an intuitive part of the object. To build probabilistic model for a class of objects, we propose a Markov random field model at each level of the tree representation to account for the spatial relationship between bases. Thus the final model integrates a Markov tree (generative) model over scales and a Markov random field over space. We adopt EM-type algorithm for learning the meaningful parts for a shape class, and show some results on shape synthesis.
ICCV	Shape Representation via Harmonic Embedding.	Alessandro Duci,Anthony J. Yezzi,Sanjoy K. Mitter,Stefano Soatto	2003	We present a novel representation of shape for closedplanar contours explicitly designed to possess a linearstructure. This greatly simplifies linear operations suchas averaging, principal component analysis or differentiation in the space of shapes. The representation reliesupon embedding the contour on a subset of the space ofharmonic functions of which the original contour is thezero level set.
ICCV	Controlling Model Complexity in Flow Estimation.	Zoran Duric,Fayin Li,Harry Wechsler,Vladimir Cherkassky	2003	"This paper describes a novel application of StatisticalLearning Theory (SLT) to control model complexity in flowestimation. SLT provides analytical generalization boundssuitable for practical model selection from small and noisydata sets of image measurements (normal flow). The methodaddresses the aperture problem by using the penalized risk(ridge regression). We demonstrate an application of thismethod on both synthetic and real image sequences and useit for motion interpolation and extrapolation. Our experimentalresults show that our approach compares favorablyagainst alternative model selection methods such as theAkaike's final prediction error, Schwartz's criterion, Generalizedcross-validation, and Shibata's model selector."
ICCV	Recognizing Action at a Distance.	Alexei A. Efros,Alexander C. Berg,Greg Mori,Jitendra Malik	2003	"Our goal is to recognize human actions at a distance,at resolutions where a whole person may be, say, 30 pixelstall. We introduce a novel motion descriptor based onoptical flow measurements in a spatio-temporal volume foreach stabilized human figure, and an associated similaritymeasure to be used in a nearest-neighbor framework. Makinguse of noisy optical flow measurements is the key challenge,which is addressed by treating optical flow not asprecise pixel displacements, but rather as a spatial patternof noisy measurements which are carefully smoothed andaggregated to form our spatio-temporal motion descriptor.To classify the action being performed by a human figurein a query sequence, we retrieve nearest neighbor(s) from adatabase of stored, annotated video sequences. We can alsouse these retrieved exemplars to transfer 2D/3D skeletonsonto the figures in the query sequence, as well as two formsof data-based action synthesis ""Do as I Do"" and ""Do as ISay"". Results are demonstrated on ballet, tennis as well asfootball datasets."
ICCV	Nonmetric Lens Distortion Calibration: Closed-form Solutions, Robust Estimation and Model Selection.	Moumen T. El-Melegy,Aly A. Farag	2003	This paper addresses the problem of calibrating camera lens distortion, which can be significant in medium to wide angle lenses. While almost all existing nonmetric distortion calibration methods need user involvement in one form or another, we present an automatic approach based on the robust the-least-median-of-squares (LMedS) estimator. Our approach is thus less sensitive to erroneous input data such as image curves that are mistakenly considered as projections of 3D linear segments. Our approach uniquely uses fast, closed-form solutions to the distortion coefficients, which serve as an initial point for a non-linear optimization algorithm to straighten imaged lines. Moreover we propose a method for distortion model selection based on geometrical inference. Successful experiments to evaluate the performance of this approach on synthetic and real data are reported.
ICCV	An automatic drowning detection surveillance system for challenging outdoor pool environments.	How-Lung Eng,Kar-Ann Toh,Alvin Harvey Kam,Junxian Wang,Wei-Yun Yau	2003	Automatically understanding events happening at a site is the ultimate goal of visual surveillance system. This paper investigates the challenges faced by automated surveillance systems operating in hostile conditions and demonstrates the developed algorithms via a system that detects water crises within highly dynamic aquatic environments. An efficient segmentation algorithm based on robust block-based background modeling and thresholding-with-hysteresis methodology enables swimmers to be reliably detected amid reflections, ripples, splashes and rapid lighting changes. Partial occlusions are resolved using a Markov Random Field framework that enhances the tracking capability of the system. Visual indicators of water crises are identified based on professional knowledge of water crises detection, based on which a set of swimmer descriptors has been defined. Through seamlessly fusing the extracted swimmer descriptors based on a novel functional link network, the system achieves promising results for water crises detection. The developed algorithms have been incorporated into a live system with robust performance for different hostile environments faced by an outdoor swimming pool.
ICCV	An affine invariant deformable shape representation for general curves.	Anders Ericsson,Kalle Åström	2003	An affine invariant deformable shape representation for general curves.
ICCV	On Exploiting Occlusions in Multiple-view Geometry.	Paolo Favaro,Alessandro Duci,Yi Ma,Stefano Soatto	2003	"Occlusions are common place in man-made and natural environments; they often result in photometric features where a line terminates at an occluding boundary, resembling a ""T"". We show that the 2-D motion of such T-junctions in multiple views carries non-trivial information on the 3-D structure of the scene and its motion relative to the camera. We show how the constraint among multiple views of T-junctions can be used to reliably detect them and differentiate them from ordinary point features. Finally, we propose an integrated algorithm to recursively and causally estimate structure and motion in the presence of T-junctions along with other point-features."
ICCV	A Bayesian Approach to Unsupervised One-Shot Learning of Object Categories.	Fei-Fei Li,Robert Fergus,Pietro Perona	2003	A Bayesian Approach to Unsupervised One-Shot Learning of Object Categories.
ICCV	On the Epipolar Geometry of the Crossed-Slits Projection.	Doron Feldman,Tomás Pajdla,Daphna Weinshall	2003	The Crossed-Slits (X-Slits) camera is defined by two non-intersectingslits, which replace the pinhole in the commonperspective camera. Each point in space is projected to theimage plane by a ray which passes through the point and thetwo slits. The X-Slits projection model includes the pushb-roomcamera as a special case. In addition, it describesa certain class of panoramic images, which are generatedfrom sequences obtained by translating pinhole cameras.In this paper we develop the epipolar geometry of the X-Slitsprojection model. We show an object which is similarto the fundamental matrix; our matrix, however, describesa quadratic relation between corresponding image points(using the Veronese mapping). Similarly the equivalent ofepipolar lines are conics in the image plane. Unlike the pin-holecase, epipolar surfaces do not usually exist in the sensethat matching epipolar lines lie on a single surface; we analyzethe cases when epipolar surfaces exist, and characterizetheir properties. Finally, we demonstrate the matchingof points in pairs of X-Slits panoramic images.
ICCV	Gamut Constrained Illuminant Estimation.	Graham D. Finlayson,Steven D. Hordley,Ingeborg Tastl	2003	This paper presents a novel solution to the illuminant estimationproblem: the problem of how, given an image of ascene taken under an unknown illuminant, we can recoveran estimate of that light. The work is founded on previousgamut mapping solutions to the problem which solvefor a scene illuminant by determining the set of diagonalmappings which take image data captured under an unknownlight to a gamut of reference colours taken undera known light. Unfortunately a diagonal model is not alwaysa valid model of illumination change and so previousapproaches sometimes return a null solution. In addition,previous methods are difficult to implement. We addressthese problems by recasting the problem as one ofilluminant classification: we define a prioria set of plausiblelights thus ensuring that a scene illuminant estimatewill always be found. A plausible light is represented bythe gamut of colours observable under it and the illuminantin an image is classified by determining the plausible lightwhose gamut is most consistent with the image data. Weshow that this step (the main computational burden of thealgorithm) can be performed simply, quickly, and efficientlyby means of a non-negative least-squares optimisation. Wereport results on a large set of real images which show thatit provides excellent illuminant estimation, outperformingprevious algorithms.
ICCV	Image-based rendering using image-based priors.	Andrew W. Fitzgibbon,Yonatan Wexler,Andrew Zisserman	2003	"Given a set of images acquired from known viewpoints, wedescribe a method for synthesizing the image which wouldbe seen from a new viewpoint.In contrast to existing techniques,which explicity reconstruct the 3D geometry of thescene, we transform the problem to the reconstruction ofcolour rather than depth.This retains the benefits of geometricconstraints, but projects out the ambiquities in depthestimation which occur in textureless regions.On the other hand, regularization is still needed in orderto generate high-quality images.The paper's secondcontribution is to constrain the generated views to lie in thespace of images whose texture statistics are those of the inputimages.This amounts to a image-based prior on thereconstruction which regularizes the solution, yielding realisticsynthetic views.Examples are given of new viewgeneration for cameras interpolated between the acquisitionviewpoints-which enables synthetic steadicam stabilizationof a sequence with a high level of realism."
ICCV	Camera Calibration with Known Rotation.	Jan-Michael Frahm,Reinhard Koch	2003	We address the problem of using external rotation informationwith uncalibrated video sequences. The main problemaddressed is, what is the benefit of the orientation infomationfor camera calibration? It is shown that in case ofa rotating camera the camera calibration problem is lineareven in the case that all intrinsic parameters vary. Forarbitrarily moving cameras the calibration problem is alsolinear but underdetermined for the general case of varyingall intrinsic parameters. However, if certain constraints areapplied to the intrinsic parameters the camera calibrationcan be computed linearily. It is analyzed which constraintsare needed for camera calibration of freely moving cameras.Furthermore we address the problem of aligning thecamera data with the rotation sensor data in time. We givean approach to align these data in case of a rotating camera.
ICCV	SVM-based Nonparametric Discriminant Analysis, An Application to Face Detection.	Rik Fransens,Jan De Prins,Luc J. Van Gool	2003	Detecting the dominant normal directions to the decisionsurface is an established technique for feature selectionin high dimensional classification problems. Several approacheshave been proposed to render this strategy moreamenable to practice, but they still show a number of importantshortcomings from a pragmatic point of view. This paperintroduces a novel such approach, which combines thenormal directions idea with Support Vector Machine classifiers.The two make a natural and powerful match, as SVsare located nearby, and fully describe the decision surfaces.The approach can be included elegantly into the training ofperformant classifiers from extensive datasets. The potentialis corroborated by experiments, both on synthetic andreal data, the latter on a face detection experiment. In thisexperiment we demonstrate how our approach can lead to asignificant reduction of CPU-time, with neglectable loss ofclassification performance.
ICCV	Texture Segmentation by Multiscale Aggregation of Filter Responses and Shape Elements.	Meirav Galun,Eitan Sharon,Ronen Basri,Achi Brandt	2003	Texture segmentation is a difficult problem, as is apparentfrom camouflage pictures. A Textured region can containtexture elements of various sizes, each of which can itselfbe textured. We approach this problem using a bottom-upaggregation framework that combines structural characteristicsof texture elements with filter responses. Our processadaptively identifies the shape of texture elements and characterizethem by their size, aspect ratio, orientation, brightness,etc., and then uses various statistics of these propertiesto distinguish between different textures. At the sametime our process uses the statistics of filter responses tocharacterize textures. In our process the shape measuresand the filter responses crosstalk extensively. In addition,a top-down cleaning process is applied to avoid mixing thestatistics of neighboring segments. We tested our algorithmon real images and demonstrate that it can accurately segmentregions that contain challenging textures.
ICCV	Mean Shift Based Clustering in High Dimensions: A Texture Classification Example.	Bogdan Georgescu,Ilan Shimshoni,Peter Meer	2003	Feature space analysis is the main module in many computer vision tasks. The most popular technique, k-means clustering, however, has two inherent limitations: the clusters are constrained to be spherically symmetric and their number has to be known a priori. In nonparametric clustering methods, like the one based on mean shift, these limitations are eliminated but the amount of computation becomes prohibitively large as the dimension of the space increases. We exploit a recently proposed approximation technique, locality-sensitive hashing (LSH), to reduce the computational complexity of adaptive mean shift. In our implementation of LSH the optimal parameters of the data structure are determined by a pilot learning procedure, and the partitions are data driven. As an application, the performance of mode and k-means based textons are compared in a texture classification study.
ICCV	Incorporating the Torrance and Sparrow Model of Reflectance in Uncalibrated Photometric Stereo.	Athinodoros S. Georghiades	2003	Under the Lambertian reflectance model, uncalibrated photometricstereo with unknown light sources is inherentlyambiguous. In this paper, we consider the use of a moregeneral reflectance model, namely the Torrance and Sparrowmodel, in uncalibrated photometric stereo. We demonstratethat this can not only resolve the ambiguity when thelight sources are unknown, but can also result in more accuratesurface reconstructions and can capture the reflectanceproperties of a large number of non-Lambertian surfaces.Our method uses single light source images with unknownlighting and no knowledge about the parameters of the reflectancemodel. It can recover the 3-D shape of surfaces(up to the binary convex/concave ambiguity) together withtheir reflectance properties. We have successfully tested ouralgorithm on a variety of non-Lambertian surfaces demonstratingthe effectiveness of our approach. In the case ofhuman faces, the estimated skin reflectance has been shownto closely resemble the measured skin reflectance reportedin the literature. We also demonstrate improved recognitionresults on 4050 images of 10 faces with variable lightingand viewpoint when the synthetic image-based representationsof the faces are generated using the surface reconstructionsand reflectance properties recovered while assumingthe extended reflectance model.
ICCV	Fragmentation in the Vision of Scenes.	Jan-Mark Geusebroek,Arnold W. M. Smeulders	2003	Natural images are highly structured in their spatial configuration. Where one would expect a different spatial distribution for every image, as each image has a different spatiallay out, we show that the spatial statistics of recorded images can be explained by a single process of sequential fragmentation. The observation by a resolution limited sensory system turns out to have a profound influence on the observed statistics of natural images. The power-law and normal distribution represent the extreme cases of sequential fragmentation. Between these two extremes, spatial detail statistics deform from power-law to normal through the Weibull type distribution as receptive field size increases relative to image detail size.
ICCV	Reflectance-based Classification of Color Edges.	Theo Gevers	2003	We aim at using color information to classify the physicalnature of edges in video. To achieve physics-based edgeclassification, we first propose a novel approach to coloredge detection by automatic noise-adaptive thresholdingderived from sensor noise analysis. Then, we present a taxonomyon color edge types. As a result, a parameter-freeedge classifier is obtained labeling color transitions intoone of the following types: (1) shadow-geometry, (2) highlightedges, (3) material edges. The proposed method isempirically verified on images showing complex real worldscenes.
ICCV	Mirrors in motion: Epipolar geometry and motion estimation.	Christopher Geyer,Kostas Daniilidis	2003	In this paper we consider the images taken from pairs ofparabolic catadioptric cameras separated by discrete motions.Despite the nonlinearity of the projection model, theepipolar geometry arising from such a system, like the perspectivecase, can be encoded in a bilinear form, the catadioptricfundamental matrix. We show that all such matriceshave equal Lorentzian singular values, and they definea nine-dimensional manifold in the space of 4 × 4 matrices.Furthermore, this manifold can be identified with a quotientof two Lie groups. We present a method to estimate a matrixin this space, so as to obtain an estimate of the motion.We show that the estimation procedures are robust to modestdeviations from the ideal assumptions.
ICCV	On the Use of Marginal Statistics of Subband Images.	Joshua Gluckman	2003	"A commonly used representation of a visual pattern is the set of marginal probability distributions of the output of a bank of filters (Gaussian, Laplacian, Gabor etc...). This representation has been used effectively for a variety of vision tasks including texture classification, texture synthesis, object detection and image retrieval. This paper examines the ability of this representation to discriminate between an arbitrary pair of visual stimuli. Examples of patterns are derived that provably possess the same marginal statistical properties, yet are ""visually distinct."" These results suggest the need for either employing a large and diverse filter bank or incorporating joint statistics in order to represent a large class of visual patterns."
ICCV	An Efficient Image Similarity Measure Based on Approximations of KL-Divergence Between Two Gaussian Mixtures.	Jacob Goldberger,Shiri Gordon,Hayit Greenspan	2003	In this work we present two new methods for approximating the Kullback-Liebler (KL) divergence between two mixtures of Gaussians. The first method is based on matching between the Gaussian elements of the two Gaussian mixture densities. The second method is based on the unscented transform. The proposed methods are utilized for image retrieval tasks. Continuous probabilistic image modeling based on mixtures of Gaussians together with KL measure for image similarity, can be used for image retrieval tasks with remarkable performance. The efficiency and the performance of the KL approximation methods proposed are demonstrated on both simulated data and real image datasets. The experimental results indicate that our proposed approximations outperform previously suggested methods.
ICCV	Recognition of Group Activities using Dynamic Probabilistic Networks.	Shaogang Gong,Tao Xiang	2003	"Dynamic Probabilistic Networks (DPNs) are exploitedfor modelling the temporal relationships among a set of differentobject temporal events in the scene for a coherentand robust scene-level behaviour interpretation. In particular,we develop a Dynamically Multi-Linked Hidden MarkovModel (DML-HMM) to interpret group activities involvingmultiple objects captured in an outdoor scene. The model isbased on the discovery of salient dynamic interlinks amongmultiple temporal events using DPNs. Object temporalevents are detected and labelled using Gaussian MixtureModels with automatic model order selection. A DML-HMMis built using Schwarz's Bayesian Information Criterionbased factorisation resulting in its topology being intrinsicallydetermined by the underlying causality and temporalorder among different object events. Our experimentsdemonstrate that its performance on modelling group activitiesin a noisy outdoor scene is superior compared to thatof a Multi-Observation Hidden Markov Model (MOHMM),a Parallel Hidden Markov Model (PaHMM) and a CoupledHidden Markov Model (CHMM)."
ICCV	Fast Stereo Matching Using Reliability-Based Dynamic Programming and Consistency Constraints.	Minglun Gong,Yee-Hong Yang	2003	A method for solving binocular and multi-view stereomatching problems is presented in this paper. A weakconsistency constraint is proposed, which expresses thevisibility constraint in the image space. It can be provedthat the weak consistency constraint holds for scenes thatcan be represented by a set of 3D points. As well, alsoproposed is a new reliability measure for dynamicprogramming techniques, which evaluates the reliability ofa given match. A novel reliability-based dynamicprogramming algorithm is derived accordingly, which canselectively assign disparity values to pixels when thereliabilities of the corresponding matches exceed a giventhreshold. Consistency constraints and the new reliability-baseddynamic programming algorithm can be combinedin an iterative approach. The experimental results showthat the iterative approach can produce dense (60~90%)and reliable (total error rate of 0.1~1.1%) matching forbinocular stereo datasets. It can also generate promisingdisparity maps for trinocular and multi-view stereodatasets.
ICCV	Applying the Information Bottleneck Principle to Unsupervised Clustering of Discrete and Continuous Image Representations.	Shiri Gordon,Hayit Greenspan,Jacob Goldberger	2003	In this paper we present a method for unsupervised clustering of image databases. The method is based on a recently introduced information-theoretic principle, the information bottleneck (IB) principle. Image archives are clustered such that the mutual information between the clusters and the image content is maximally preserved. The IB principle is applied to both discrete and continuous image representations, using discrete image histograms and probabilistic continuous image modeling based on mixture of Gaussian densities, respectively. Experimental results demonstrate the performance of the proposed method forimage clustering on a large image database. Several clustering algorithms derived from the IB principle are explored and compared.
ICCV	Recovery of Epipolar Geometry as a Manifold Fitting Problem.	Liran Goshen,Ilan Shimshoni,P. Anandan,Daniel Keren	2003	The introduction of the joint image manifold allows to treatthe problem of recovering camera motion and epipolar geometryas the problem of fitting a manifold to the data measuredin a stereo pair. The manifold has a singularity andboundary, therefore care must be taken when fitting it.This paper reviews the notion of joint image manifold,and how previous motion recovery methods can be viewedin its context, and then offers a new fitting method, whichimproves upon previous results, especially when the extentof the data and/or the motion are small.
ICCV	Inferring 3D Structure with a Statistical Image-Based Shape Model.	Kristen Grauman,Gregory Shakhnarovich,Trevor Darrell	2003	"We present an image-based approach to infer 3D structureparameters using a probabilistic ""shape+structure"" model.The 3D shape of an object class is represented by setsof contours from silhouette views simultaneously observedfrom multiple calibrated cameras, while structural featuresof interest on the object are denoted by a number of 3D locations.A prior density over the multi-view shape and correspondingstructure is constructed with a mixture of probabilisticprincipal components analyzers. Given a novelset of contours, we infer the unknown structure parametersfrom the new shape's Bayesian reconstruction. Modelmatching and parameter inference are done entirely in theimage domain and require no explicit 3D construction. Ourshape model enables accurate estimation of structure despitesegmentation errors or missing views in the input silhouettes,and it works even with only a single input view.Using a training set of thousands of pedestrian images generatedfrom a synthetic model, we can accurately infer the3D locations of 19 joints on the body based on observedsilhouette contours from real images."
ICCV	Probabilistic Bilinear Models for Appearance-Based Vision.	David B. Grimes,Aaron P. Shon,Rajesh P. N. Rao	2003	"We present a probabilistic approach to learning objectrepresentations based on the ""content and style"" bilineargenerative model of Tenenbaum and Freeman. In contrastto their earlier SVD-based approach, our approach modelsimages using particle filters. We maintain separate particlefilters to represent the content and style spaces, allowing usto define arbitrary weighting functions over the particles tohelp estimate the content/style densities. We combine thisapproach with a new EM-based method for learning basisvectors that describe content-style mixing. Using a particle-basedrepresentation permits good reconstruction despitereduced dimensionality, and increases storage capacity andcomputational efficiency. We describe how learning the distributionsusing particle filters allows us to efficiently computea probabilistic ""novelty"" term. Our example applicationconsiders a dataset of faces under different lightingconditions. The system classifies faces of people it has seenbefore, and can identify previously unseen faces as new content.Using a probabilistic definition of novelty in conjunctionwith learning content-style separability provides a crucialbuilding block for designing real-world, real-time objectrecognition systems."
ICCV	Towards a Mathematical Theory of Primal Sketch and Sketchability.	Cheng-en Guo,Song Chun Zhu,Ying Nian Wu	2003	"In this paper, we present a mathematical theory for Marr'sprimal sketch. We first conduct a theoretical study ofthe descriptive Markov random field model and the generativewavelet/sparse coding model from the perspectiveof entropy and complexity. The competition between thetwo types of models defines the concept of ""sketchability"",which divides image into texture and geometry. We then proposea primal sketch model that integrates the two modelsand, in addition, a Gestalt field model for spatial organization.We also propose a sketching pursuit process that coordinatesthe competition between two pursuit algorithms:the matching pursuit [8] and the filter pursuit [12], that seekto explain the image by bases and filters respectively. Themodel can be used to learn a dictionary of image primitives,or textons in Julesz's language, for natural images.The primal sketch model is not only parsimonious for imagerepresentation, but produces meaningful sketches overa large number of generic images."
ICCV	Determining Reflectance and Light Position from a Single Image Without Distant Illumination Assumption.	Kenji Hara,Ko Nishino,Katsushi Ikeuchi	2003	"Several techniques have been developed for recovering reflectance properties of real surfaces under unknown illumination conditions. However, in most cases, those techniques assume that the light sources are located at inifinity,which cannot be applied to, for example, photometric modeling of indoor environments. In this paper, we propose two methods to estimate the surface reflectance property of an object, as well as the position of a light source from a single image without the distant illumination assumption. Given a color image of an object with specular reflection as an input, the first method estimates the light source position by fitting to the Lambertian diffuse component, while separating the specular and diffuse components by using an iterative relaxation scheme. Moreover, we extend the above method by using a single specular image as an input, thus removing its constraints on the diffuse reflectance property and the number of light sources. This method simultaneously recovers the reflectance properties and the light source positions by optimizing the linearity of a log-transformed Torrance-Sparrow model. By estimating the object's reflectance property and the light source position, we can freely generate synthetic images of the target object under arbitrary source directions and source-surface distances."
ICCV	Photo-Consistent 3D Fire by Flame-Sheet Decomposition.	Samuel W. Hasinoff,Kiriakos N. Kutulakos	2003	This paper considers the problem of reconstructing visuallyrealistic 3D models of fire from a very small setof simultaneous views (even two). By modeling fire as asemi-transparent 3D density field, we show that fire reconstructionis equivalent to a severely under-constrained computerizedtomography problem, for which traditional methodsbreak down. Our approach is based on the observationthat every pair of photographs of a semi-transparentscene defines a unique density field, called a Flame Sheet,that (1) concentrates all its density on one connected, semi-transparentsurface, (2) reproduces the two photos exactly,and (3) is the most spatially-coherent density field that doesso. From this observation, we reduce fire reconstruction tothe convex combination of sheet-like density fields, each ofwhich is derived from the Flame Sheet of two input photos.Experimental results suggest that this method enables high-qualityview extrapolation without over-fitting artifacts.
ICCV	Statistical Background Subtraction for a Mobile Observer.	Eric Hayman,Jan-Olof Eklundh	2003	"Statistical background modelling and subtraction has proved to be a popular and effective class of algorithms for segmenting independently moving foreground objects outfrom a static background, without requiring any a priori information of the properties of foreground objects. This paper presents two contributions on this topic, aimed towards robotics where an active head is mounted on a mobile vehicle. In periods when the vehicle's wheels are not driven, camera translation is virtually zero, and background subtraction techniques are applicable. Parts of this work are also highly relevant to surveillance and video conferencing. The first part of the paper presents an efficient probabilistic framework for when the camera pans and tilts. A unified approach is developed for handling various sources of error, including motion blur, sub-pixel camera motion, mixed pixels at object boundaries, and also uncertainty in background stabilisation caused by noise, unmodelled radial distortion and small translations of the camera. The second contribution regards a Bayesian approach to specifically incorporate uncertainty concerning whether the background has yet been uncovered by moving foreground objects. This is an important requirement during initialisation of a system. We cannot assume that a background model is available in advance since that would involve storing models for each possible position, in every room, of the robot's operating environment. Instead the background model must be generated online, very possibly in the presence of moving objects."
ICCV	Learning a Locality Preserving Subspace for Visual Recognition.	Xiaofei He,Shuicheng Yan,Yuxiao Hu,HongJiang Zhang	2003	Previous works have demonstrated that the face recognition performance can be improved significantly in low dimensional linear subspaces. Conventionally, principal component analysis (PCA) and linear discriminant analysis (LDA) are considered effective in deriving such a face subspace. However, both of them effectively see only the Euclidean structure of face space. In this paper, we propose a new approach to mapping face images into a sub-space obtained by Locality Preserving Projections (LPP) for face analysis. We call this Laplacian face approach. Different from PCA and LDA, LPP finds an embedding that preserves local information, and obtains a face space that best detects the essential manifold structure. In this way, the unwanted variations resulting from changes in lighting, facial expression, and pose may be eliminated or reduced. We compare the proposed Laplacian face approach with eigenface and fisherface methods on three test datasets. Experimental results show that the proposed Laplacianface approach provides a better representation and achieves lower error rates in face recognition.
ICCV	Natural Image Statistics for Natural Image Segmentation.	Matthias Heiler,Christoph Schnörr	2003	Building on recent progress in modeling £lter responsestatistics of natural images we integrate a statistical modelinto a variational framework for image segmentation. Incorporatedin a sound probabilistic distance measure themodel drives level sets toward meaningful segmentationsof complex textures and natural scenes. Since each regioncomprises two model parameters only the approachis computationally ef£cient and enables the application ofvariational segmentation to a considerably larger class ofreal-world images. We validate the statistical basis of ourapproach on thousands of natural images and demonstratethat our model outperforms recent variational segmentationmethods based on second-order statistics.
ICCV	Real Time Pattern Matching Using Projection Kernels.	Yacov Hel-Or,Hagit Hel-Or	2003	A novel approach to pattern matching is presented, whichreduces time complexity by two orders of magnitude comparedto traditional approaches. The suggested approachuses an efficient projection scheme which bounds the distancebetween a pattern and an image window using veryfew operations. The projection framework is combined witha rejection scheme which allows rapid rejection of imagewindows that are distant from the pattern. Experimentsshow that the approach is effective even under very noisyconditions. The approach described here can also be usedin classification schemes where the projection values serveas input features that are informative and fast to extract.
ICCV	Bayesian Clustering of Optical Flow Fields.	Jesse Hoey,James J. Little	2003	"We present a method for unsupervised learning of classesof motions in video. We project optical flow fields to a complete,orthogonal, a-priori set of basis functions in a probabilisticfashion, which improves the estimation of the projectionsby incorporating uncertainties in the flows. We thencluster the projections using a mixture of feature-weightedGaussians over optical flow fields. The resulting modelextracts a concise probabilistic description of the majorclasses of optical flow present. The method is demonstratedon a video of a person's facial expressions."
ICCV	Large-Scale Event Detection Using Semi-Hidden Markov Models.	Somboon Hongeng,Ramakant Nevatia	2003	We present a new approach to recognizing events invideos.We first detect and track moving objects in thescene.Based on the shape and motion properties ofthese objects, we infer probabilities of primitive eventsframe-by-frame by using Bayesian networks.Compositeevents, consisting of multiple primitive events, overextended periods of time are analyzed by using a hidden,semi-Markov finite state model.This results in morereliable event segmentation compared to the use of standardHMMs in noisy video sequences at the cost of someincrease in computational complexity.We describe ourapproach to reducing this complexity.We demonstratethe effectiveness of our algorithm using both real-worldand pertubed data.
ICCV	Outlier Correction in Image Sequences for the Affine Camera.	Du Q. Huynh,R. Hartley,Anders Heyden	2003	It is widely known that, for the affine camera model, both shape and motion can be factorized directly from the so-called image measurement matrix constructed from image point coordinates. The ability to extract both shape and motion from this matrix by a single SVD operation makes this shape-from-motion approach attractive; however, it cannot deal with missing feature points and, in the presence of outliers, a direct SVD to the matrix would yield highly unreliable shape and motion components. In this paper, we present an outlier correction scheme that iteratively updates the elements of the image measurement matrix. The magnitude and sign of the update to each element is dependent upon the residual robustly estimated in each iteration. The result is that outliers are corrected and retained, giving improved reconstruction and smaller reprojection errors. Our iterative outlier correction scheme has been applied to both synthesized and real video sequences. The results obtained are remarkably good.
ICCV	Stochastic Refinement of the Visual Hull to Satisfy Photometric and Silhouette Consistency Constraints.	John Isidoro,Stan Sclaroff	2003	"An iterative method for reconstructing a 3D polygonalmesh and color texture map from multiple views of an objectis presented. In each iteration, the method first estimates atexture map given the current shape estimate. The texturemap and its associated residual error image are obtainedvia maximum a posteriori estimation and reprojection of themultiple views into texture space. Next, the surface shape isadjusted to minimize residual error in texture space. Thesurface is deformed towards a photometrically-consistentsolution via a series of 1D epipolar searches at randomlyselected surface points. The texture space formulation hasimproved computational complexity over standard image-basederror aproaches, and allows computation of the reprojectionerror and uncertainty for any point on the surface.Moreover, shape adjustments can be constrained suchthat the recovered model's silhouette matches those of theinput images. Experiments with real world imagery demonstratethe validity of the approach."
ICCV	Tracking Across Multiple Cameras With Disjoint Views.	Omar Javed,Zeeshan Rasheed,Khurram Shafique,Mubarak Shah	2003	Conventional tracking approaches assume proximity inspace, time and appearance of objects in successive observations.However, observations of objects are often widelyseparated in time and space when viewed from multiplenon-overlapping cameras. To address this problem, wepresent a novel approach for establishing object correspondenceacross non-overlapping cameras. Our multi-cameratracking algorithm exploits the redundance in paths thatpeople and cars tend to follow, e.g. roads, walk-ways orcorridors, by using motion trends and appearance of objects,to establish correspondence. Our system does notrequire any inter-camera calibration, instead the systemlearns the camera topology and path probabilities of objectsusing Parzen windows, during a training phase. Oncethe training is complete, correspondences are assigned usingthe maximum a posteriori (MAP) estimation framework.The learned parameters are updated with changing trajectorypatterns. Experiments with real world videos are reported,which validate the proposed approach.
ICCV	Images as Bags of Pixels.	Tony Jebara	2003	"We propose modeling images and related visual objects as bags of pixels or sets of vectors. For instance, gray scale images are modeled as a collection or bag of (X, Y, I) pixel vectors. This representation implies a permutational invariance over the bag of pixels which is naturally handled by endowing each image with a permutation matrix. Each matrix permits the image to span a manifold of multiple configurations, capturing the vector set's invariance to orderings or permutation transformations. Permutation configurations are optimized while jointly modeling many images via maximum likelihood. The solution is a uniquely solvable convex program which computes correspondence simultaneously for all images (as opposed to traditional pairwise correspondence solutions). Maximum likelihood performs a nonlinear dimensionality reduction, choosing permutations that compact the permuted image vectors into a volumetrically minimal subspace. This is highly suitable for principal components analysis which, when applied to the permutationally invariant bag of pixels representation, outperforms PCA on appearance-based vectorization by orders of magnitude.Furthermore, the bag of pixels subspace benefits from automatic correspondence estimation, giving rise to meaningful linear variations such as morphings, translations, and jointly spatio-textural image transformations. Results are shown for several datasets."
ICCV	Shape Gradients for Histogram Segmentation using Active Contours.	Stéphanie Jehan-Besson,Michel Barlaud,Gilles Aubert,Olivier D. Faugeras	2003	We consider the problem of image segmentation using active contours through the minimization of an energy criterion involving both region and boundary functionals. These functionals are derived through a shape derivative approach instead of classical calculus of variation. The equations can be elegantly derived without converting the region integrals into boundary integrals. From the derivative, we deduce the evolution equation of an active contour that makes it evolve towards a minimum of the criterion. We focus more particularly on statistical features globally attached to the region and especially to the probability density functions of image features such as the color histogram of a region. A theoretical framework is set for the minimization of the distance between two histograms for matching or tracking purposes. An application of this framework to the segmentation of color histograms in video sequences is then proposed. We briefly describe our numerical scheme and show some experimental results.
ICCV	Image Registration with Global and Local Luminance Alignment.	Jiaya Jia,Chi-Keung Tang	2003	Inspired by tensor voting, we present luminance voting, a novel approach for image registration with global and local luminance alignment. The key to our modeless approach is the direct estimation of replacement function, by reducing the complex estimation problem to the robust 2D tensor voting in the corresponding voting spaces. No model for replacement function is assumed. Luminance data are first encoded into 2D ball tensors. Subject to the monotonic constraint only, we vote for an optimal replacement function by propagating the smoothness constraint using a dense tensor field. Our method effectively infers missing curve segments and rejects image outliers without assuming any simplifying or complex curve model. The voted replacement functions are used in our iterative registration algorithm for computing the best warping matrix. Unlike previous approaches, our robust method corrects exposure disparity even if the two overlapping images are initially misaligned. Luminance voting is effective in correcting exposure difference, eliminating vignettes, and thus improving image registration. We present results on a variety of images.
ICCV	Circular Motion Geometry by Minimal 2 Points in 4 Images.	Guang Jiang,Long Quan,Hung-Tat Tsui	2003	This paper describes a new and simple method of recovering the geometry of uncalibrated circular motion or single axis motion using a minimal data set of 2 points in 4 images. This problem has been solved using non-minimal data either by computing the fundamental matrix and trifocal tensor in 3 images, or by fitting conics to tracked points in 5 images. Our new method first computes a planar homography from a minimum of 2 points in 4 images. It is shown that two eigenvectors of this homography are the images of the circular points. Then, other fixed image entities and rotation angles can be straightforwardly computed. The crux of the method lies in relating this planar homography from two different points to a homology naturally induced by corresponding points on different conic loci from a circular motion. The experiments on real image sequences demonstrate the simplicity, accuracy and robustness of the new method.
ICCV	Epitomic analysis of appearance and shape.	Nebojsa Jojic,Brendan J. Frey,Anitha Kannan	2003	We present novel simple appearance and shape models that we call epitomes. The epitome of an image is its miniature, condensed version containing the essence of the textural and shape properties of the image. As opposed to previously used simple image models, such as templates or basis functions, the size of the epitome is considerably smaller than the size of the image or object it represents, but the epitome still contains most constitute elements needed to reconstruct the image (Fig. 1). A collection of images often shares an epitome, e.g., when images are a few consecutive frames from a video sequence, or when they are photographs of similar objects.A particular image in a collection is defined by its epitome and a smooth mapping from the epitome to the image pixels. When the epitomic representation is used within a hierarchical generative model, appropriate inference algorithms can be derived to extract the epitome from a single image or a collection of images and at the same time perform various inference tasks, such as image segmentation, motion estimation, object removal and super-resolution.
ICCV	High resolution terrain mapping using low altitude aerial stereo imagery.	Il-Kyun Jung,Simon Lacroix	2003	This paper presents an approach to build high resolutiondigital elevation maps from a sequence of unregistered lowaltitude stereovision image pairs. The approach first uses avisual motion estimation algorithm that determines the 3Dmotions of the cameras between consecutive acquisitions,on the basis of visually detected and matched environmentfeatures. An extended Kalman filter then estimates both the6 position parameters and the 3D positions of the memorizedfeatures as images are acquired. Details are given onthe filter implementation and on the estimation of the uncertaintieson the feature observations and motion estimations.Experimental results show that the precision of the methodenables to build spatially consistent very large maps.
ICCV	Unsupervised Non-parametric Region Segmentation Using Level Sets.	Timor Kadir,Michael Brady	2003	We present a novel non-parametric unsupervised segmentationalgorithm based on Region Competition [21];but implemented within a Level Sets framework [11]. Thekey novelty of the algorithm is that it can solve N ¿ 2 classsegmentation problems using just one embedded surface;this is achieved by controlling the merging and splitting behaviourof the level sets according to a Minimum DescriptionLength (MDL) [6, 14] cost function. This is in contrastto N class region-based Level Set segmentation methods todate which operate by evolving multiple coupled embeddedsurfaces in parallel [3, 13, 20]. Furthermore, it operates inan unsupervised manner; it is necessary neither to specifythe value of N nor the class models a-priori.We argue that the Level Sets methodology provides amore convenient framework for the implementation of theRegion Competition algorithm, which is conventionally implementedusing region membership arrays due to the lackof a intrinsic curve representation. Finally, we generalisethe Gaussian region model used in standard Region Competitionto the non-parametric case. The region boundary motionand merge equations become simple expressions containingcross-entropy and entropy terms.
ICCV	Multiview Reconstruction of Space Curves.	Fredrik Kahl,Jonas August	2003	"Is the real problem in resolving correspondence using currentstereo algorithms the lack of the ""right"" matching criterion?In studying the related task of reconstructing three-dimensionalspace curves from their projections in multipleviews, we suggest that the problem is more basic: matchingand reconstruction are coupled, and so reconstruction algorithmsshould exploit this rather than assuming that matchingcan be successfully performed before reconstruction. Torealize this coupling, a generative model of curves is introducedwhich has two key components: (i) a prior distributionof general space curves and (ii) an image formation modelwhich describes how 3D curves are projected onto the imageplane. A novel aspect of the image formation model is that ituses an exact description of the gradient field of a piecewiseconstant image. Based on this forward model, a fully automaticalgorithm for solving the inverse problem is developedfor an arbitrary number of views. The resulting algorithmis robust to partial occlusion, deficiencies in image curveextraction and it does not rely on photometric information.The relative motion of the cameras is assumed to be given.Several experiments are carried out on various realistic scenarios.In particular, we focus on scenes where traditionalcorrelation-based methods would fail."
ICCV	Reliable Recovery of Piled Box-like Objects via Parabolically Deformable Superquadrics.	Dimitrios Katsoulas	2003	"Automatic unloading of piled box-like objects is undoubtedlyof great importance to the industry. In this contributiona system addressing this problem is described: Weemploy a laser range finder for data acquisition, and globallydeformable Superquadrics [2], [22] for object modeling.Our technique is based on a hypothesis generation andrefinement scheme. The vertices of the piled objects are extractedand Superquadric seeds are aligned at these vertices.The model parameter recovery task is decomposedinto two subproblems, each dealing with a subset of themodel's parameter set. Both region and boundary based informationsources are used for parameter estimation. Comparedto a widespread strategy for superquadric recovery[11], our method shows advantages in terms of robustnessand computational efficiency. In addition, our system exhibitsversatility with regard to existing industrial systems,since it can effectively deal with both neatly placed and jumbledconfigurations of objects"
ICCV	Visual Correspondence Using Energy Minimization and Mutual Information.	Junhwan Kim,Vladimir Kolmogorov,Ramin Zabih	2003	We address visual correspondence problems without assuming that scene points have similar intensities in different views. This situation is common, usually due tonon-lambertian scenes or to differences between cameras. We use maximization of mutual information, apowerful technique for registering images that requiresno apriori model of the relationship between scene intensities in different views. However, it has provendifficult to use mutual information to compute densevisual correspondence. Comparing fixed-size windowsvia mutual information suffers from the well-knownproblems of fixed windows, namely poor performanceat discontinuities and in low-texture regions. In thispaper, we show how to compute visual correspondenceusing mutual information without suffering from theseproblems. Using a simple approximation, mutual information can be incorporated into the standard energyminimization framework used in early vision. The energy can then be efficiently minimized using graph cuts,which preserve discontinuities and handle low-textureregions. The resulting algorithm combines the accuratedisparity maps that come from graph cuts with the tolerance for intensity changes that comes from mutual information.
ICCV	Fast Vehicle Detection with Probabilistic Feature Grouping and its Application to Vehicle Tracking.	Zu Whan Kim,Jitendra Malik	2003	Generating vehicle trajectories from video data is an important application of ITS (Intelligent Transportation Systems). We introduce a new tracking approach which uses model-based 3-D vehicle detection and description algorithm. Our vehicle detection and description algorithm is based on a probabilistic line feature grouping, and it is faster (by up to an order of magnitude) and more flexible than previous image-based algorithms. We present the system implementation and the vehicle detection and tracking results.
ICCV	Dynamic Stroke Information Analysis for Video-Based Handwritten Chinese Character Recognition.	Feng Lin,Xiaoou Tang	2003	Video-based handwritten Character Recognition(VCR) system is a new type of character recognitionsystem with many unique advantages over on-linecharacter recognition system. Its main problem is toeffectively extract stroke dynamic information fromvideo data for character recognition. In this paper, wepropose a new stroke extraction algorithm throughdynamic stroke information analysis for a VCR system.The experimental results on over 3000 video charactersequences show that our system can extract the Chinesecharacter stroke dynamic information similar to an on-line system.
ICCV	A Segmentation Algorithm for Contrast-Enhanced Images.	Junhwan Kim,Ramin Zabih	2003	"Medical imaging often involves the injection of contrast agents and the subsequent analysis of tissue enhancement patterns. Many important types of tissue have characteristic enhancement patterns; for example, in magnetic resonance (MR) mammography, malignancies exhibit a characteristic ""wash out"" temporal pattern, while in MR angiography, arteries, veins and parenchyma each have their own distinctive temporal signature. In such image sequences, there are substantial changes in intensities; however, this change is due primarily to the contrast agent rather than the motion of scene elements. As a result, the task of segmenting contrast-enhanced images poses interesting new challenges for computer vision. In this paper, we propose a new image segmentation algorithm for image sequences with contrast enhancement, using a model-based time series analysis of individual pixels. We use energy minimization via graph cuts to efficiently ensure spatial coherence. The energy is minimized in an expectation-maximization fashion that alternates between segmenting the image into a number of non-overlapping regions and finding the temporal profile parameters which best describe the behavior of each region. Preliminary experiments on MR mammography and MR angiography studies show the algorithm's ability to find an accurate segmentation."
ICCV	Meshfree Particle Method.	Huafeng Liu,Pengcheng Shi	2003	Many of the computer vision algorithms have been posed invarious forms of differential equations, derived from minimization of specific energy functionals, and the finite element representation and computation have become the de facto numerical strategies for solving these problems. However, for cases where domain mappings between numerical iterations or image frames involve large geometrical shape changes, such as deformable models for object segmentation and non rigid motion tracking, these strategies may exhibit considerable loss of accuracy when themesh elements become extremely skewed or compressed. We present a new computational paradigm, the meshfree particle method, where the object representation and the numerical calculation are purely based on the nodal points and do not require the meshing of the analysis domain. This meshfree strategy can naturally handle large deformation and domain discontinuity issues and achieve desired numerical accuracy through adaptive node and polynomial shape function refinement. We discuss in detail the element-free Galerkin method, including the shape function construction using the moving least square approximation and the Galerkin weak form formulation, and we demonstrate its applications to deformable model based segmentation and mechanically motivated left ventricular motion analysis.
ICCV	Integrated Edge and Junction Detection with the Boundary Tensor.	Ullrich Köthe	2003	The boundaries of image regions necessarily consist of edges (in particular, step and roof edges), corners, and junctions. Currently, different algorithms are used to detect each boundary type separately, but the integration of the results into a single boundary representation is difficult. Therefore, a method for the simultaneous detection of all boundary types is needed. We propose to combine responses of suitable polar separable filters into what we will call the boundary tensor. The trace of this tensor is a measure of boundary strength, while the small eigenvalue and its difference to the large one represent corner / junction and edge strengths respectively. We prove that the edge strength measure behaves like a rotationally invariant quadrature filter. A number of examples demonstrate the properties of the new method and illustrate its application to image segmentation.
ICCV	Combining Gradient and Albedo Data for Rotation Invariant Classification of 3D Surface Texture.	Jiahua Wu,Mike J. Chantler	2003	We present a new texture classification scheme whichis invariant to surface-rotation. Many textureclassification approaches have been presented in the pastthat are image-rotation invariant, However, imagerotation is not necessarily the same as surface rotation.We have therefore developed a classifier that usesinvariants that are derived from surface properties ratherthan image properties. Previously we developed ascheme that used surface gradient (normal) fieldsestimated using photometric stereo. In this paper weaugment these data with albedo information and an alsoemploy an additional feature set: the radial spectrum.We used 30 real textures to test the new classifier. Aclassification accuracy of 91% was achieved whenalbedo and gradient 1D polar and radial features werecombined. The best performance was also achieved byusing 2D albedo and gradient spectra. The classificationaccuracy is 99%.
ICCV	Discriminative Random Fields: A Discriminative Framework for Contextual Interaction in Classification.	Sanjiv Kumar,Martial Hebert	2003	In this work we present Discriminative Random Fields(DRFs), a discriminative framework for the classification ofimage regions by incorporating neighborhood interactionsin the labels as well as the observed data. The discriminativerandom fields offer several advantages over the conventionalMarkov Random Field (MRF) framework. First,the DRFs allow to relax the strong assumption of conditionalindependence of the observed data generally used inthe MRF framework for tractability. This assumption is toorestrictive for a large number of applications in vision. Second,the DRFs derive their classification power by exploitingthe probabilistic discriminative models instead of thegenerative models used in the MRF framework. Finally, allthe parameters in the DRF model are estimated simultaneouslyfrom the training data unlike the MRF frameworkwhere likelihood parameters are usually learned separatelyfrom the field parameters. We illustrate the advantages ofthe DRFs over the MRF framework in an application ofman-made structure detection in natural images taken fromthe Corel database.
ICCV	Surface Reflectance Modeling of Real Objects with Interreflections.	Takashi Machida,Naokazu Yokoya,Haruo Takemura	2003	In mixed reality, especially in augmented virtuality which virtualizes real objects, it is important to estimate object surface reflectance properties to render the objects under arbitrary illumination conditions. Though several methods have been explored to estimate the surface reflectance properties, it is still difficult to estimate surface reflectance parameters faithfully for complex objects which have non-uniform surface reflectance properties and exhibitinter reflections. This paper describes a new method for densely estimating non-uniform surface reflectance properties of real objects constructed of convex and concave surfaces with interreflections. We use registered range and surface color texture images obtained by a laser range finder. Experiments show the usefulness of the proposed method.
ICCV	Minimum Risk Distance Measure for Object Recognition.	Shyjan Mahamud,Martial Hebert	2003	The optimal distance measure for a given discrimination task under the nearest neighbor framework has been shown to be the likelihood that a pair of measurements have different class labels [5]. For implementation and efficiency considerations, the optimal distance measure was approximated by combining more elementary distance measures defined on simple feature spaces. In this paper, we address two important issues that arise in practice for such an approach:(a) What form should the elementary distance measure in each feature space take? We motivate the need to use the optimal distance measure in simple feature spaces as the elementary distance measures; such distance measures have the desirable property that they are invariant to distance-respecting transformations. (b) How do we combine the elementary distance measures? We present the precise statistical assumptions under which a linear logistic model holds exactly. We benchmark our model with three other methods on a challenging face discrimination task and show that our approach is competitive with the state of the art.
ICCV	Space-time Interest Points.	Ivan Laptev,Tony Lindeberg	2003	Local image features or interest points provide compact and abstract representations of patterns in an image. In this paper, we propose to extend the notion of spatial interest points into the spatio-temporal domain and show how the resulting features often reflect interesting events that can be used for a compact representation of video data as well as for its interpretation. To detect spatio-temporal events, we build on the idea of the Harris and F&ouml;rstner interest point operators and detect local structures in space-time where the image values have significant local variations in both space and time. We then estimate the spatio-temporal extents of the detected events and compute their scale-invariant spatio-temporal descriptors. Using such descriptors, we classify events and construct video representation in terms of labeled space-time points. For the problem of human motion analysis, we illustrate how the proposed method allows for detection of walking people in scenes with occlusions and dynamic back-grounds.
ICCV	The Local Projective Shape of Smooth Surfaces and their Outlines.	Svetlana Lazebnik,Jean Ponce	2003	"This paper examines projectively invariant local properties of smooth curves and surfaces. Oriented projective differential geometry is proposed as a theoretical framework for establishing such invariants and describing the local shape of surfaces and their outlines. This framework is applied to two problems: a projective proof of Koenderink's famous characterization of convexities, concavities, and inflections of apparent contours; and the determination of the relative orientation of rim tangents at frontier points."
ICCV	Affine-Invariant Local Descriptors and Neighborhood Statistics for Texture Recognition.	Svetlana Lazebnik,Cordelia Schmid,Jean Ponce	2003	This paper presents a framework for texture recognitionbased on local affine-invariant descriptors and their spatiallayout. At modeling time, a generative model of localdescriptors is learned from sample images using the EM algorithm.The EM framework allows the incorporation ofunsegmented multi-texture images into the training set. Thesecond modeling step consists of gathering co-occurrencestatistics of neighboring descriptors. At recognition time,initial probabilities computed from the generative modelare refined using a relaxation step that incorporates co-occurrencestatistics. Performance is evaluated on imagesof an indoor scene and pictures of wild animals.
ICCV	Learning Pedestrian Models for Silhouette Refinement.	L. Lee,Gerald Dalley,Kinh Tieu	2003	We present a model-based method for accurate extraction ofpedestrian silhouettes from video sequences. Our approachis based on two assumptions, 1) there is a common appearanceto all pedestrians, and 2) each individual looks likehim/herself over a short amount of time. These assumptionsallow us to learn pedestrian models that encompassboth a pedestrian population appearance and the individualappearance variations. Using our models, we are ableto produce pedestrian silhouettes that have fewer noise pixelsand missing parts. We apply our silhouette extractionapproach to the NIST gait data set and show that under thegait recognition task, our model-based sulhouettes resultin much higher recognition rates than silhouettes directlyextracted from background subtraction, or any non-model-basedsmoothing schemes.
ICCV	Unsupervised Improvement of Visual Detectors using Co-Training.	Anat Levin,Paul A. Viola,Yoav Freund	2003	One significant challenge in the construction of visualdetection systems is the acquisition of sufficient labeleddata. This paper describes a new technique for trainingvisual detectors which requires only a small quantity of labeleddata, and then uses unlabeled data to improve performanceover time. Unsupervised improvement is based onthe co-training framework of Blum and Mitchell, in whichtwo disparate classifiers are trained simultaneously. Unlabeledexamples which are confidently labeled by one classifierare added, with labels, to the training set of the otherclassifier. Experiments are presented on the realistic task ofautomobile detection in roadway surveillance video. In thisapplication, co-training reduces the false positive rate by afactor of 2 to 11 from the classifier trained with labeled dataalone.
ICCV	Learning How to Inpaint from Global Image Statistics.	Anat Levin,Assaf Zomet,Yair Weiss	2003	Inpainting is the problem of filling-in holes in images. Considerable progress has been made by techniques that use the immediate boundary of the hole and some prior information on images to solve this problem. These algorithms successfully solve the local inpainting problem but they must, by definition, give the same completion to any two holes that have the same boundary, even when the rest of the image is vastly different. In this paper we address a different, more global inpainting problem. How can we use the rest of the image in order to learn how to inpaint? We approach this problem from the context of statistical learning. Given a training image we build an exponential family distribution over images that is based on the histograms of local features. We then use this image specific distribution to in paint the hole by finding the most probable image given the boundary and the distribution. The optimization is done using loopy belief propagation. We show that our method can successfully complete holes while taking into account the specific image statistics. In particular it can give vastly different completions even when the local neighborhoods are identical.
ICCV	Surface Reconstruction by Integrating 3D and 2D Data of Multiple Views.	Maxime Lhuillier,Long Quan	2003	Surface representation is needed for almost all modelingand visualization applications, but unfortunately, 3D datafrom a passive vision system are often insufficient for a traditionalsurface reconstruction technique that is designedfor densely scanned 3D point data. In this paper, we developa new method for surface reconstruction by combiningboth 3D data and 2D image information. The silhouetteinformation extracted from 2D images can also be integratedas an option if it is available. The new methodis a variational approach with a new functional integrating3D stereo data with 2D image information. This givesa more robust approach than existing methods using onlypure 2D information or 3D stereo data. We also propose abounded regularization method to implement efficiently thesurface evolution by level-set methods. The properties ofthe algorithms are discussed, proved for some cases, andempirically demonstrated through intensive experiments onreal sequences.
ICCV	Multiple-cue Illumination Estimation in Textured Scenes.	Yuanzhen Li,Stephen Lin,Hanqing Lu,Heung-Yeung Shum	2003	In this paper, we present a method that integrates cuesfrom shading, shadow and specular reflections for estimatingdirectional illumination in a textured scene. Textureposes a problem for lighting estimation, since texture edgescan be mistaken for changes in illumination condition, andunknown variations in albedo make reflectance model fittinginpractical. Unlike previous works which all assumeknown or uniform reflectance, our method can deal with theeffects of textures by capitalizing on physical consistenciesthat exist among the lighting cues. Since scene textures donot exhibit such coherence, we use this property to minimizethe influence of texture on illumination direction estimation.For the recovered light source directions, a technique forestimating their intensities in the presence of texture is alsoproposed.
ICCV	Using Specularities for Recognition.	Margarita Osadchy,David W. Jacobs,Ravi Ramamoorthi	2003	"Recognition systems have generally treated specular highlightsas noise. We show how to use these highlights asa positive source of information that improves recognitionof shiny objects. This also enables us to recognize verychallenging shiny transparent objects, such as wine glasses.Specifically, we show how to find highlights that are consistentwith an hypothesized pose of an object of known 3Dshape. We do this using only a qualitative description ofhighlight formation that is consistent with most models ofspecular reflection, so no specific knowledge of an object'sreflectance properties is needed. We first present a methodthat finds highlights produced by a dominant compact lightsource, whose position is roughly known. We then show howto estimate the lighting automatically for objects whose reflectionis part specular and part Lambertian. We demonstratethis method for two classes of objects. First, we showthat specular information alone can suffice to identify objectswith no Lambertian reflectance, such as transparentwine glasses. Second, we use our complete system to recognizeshiny objects, such as pottery."
ICCV	Polarization-based Transparent Surface Modeling from Two Views.	Daisuke Miyazaki,Masataka Kagesawa,Katsushi Ikeuchi	2003	"In this paper, we propose a novel method to recover thesurface shape of transparent objects. The degree of polarizationof the light reflected from the object surface dependson the reflection angle which, in turn, depends on the object'ssurface normal; thus, by measuring the degree of polarization,we are able to calculate the surface normal of theobject. However, degree of polarization and surface normaldoes not correspond one-to-one, making us to analyze twopolarization images taken from two different view in orderto solve the ambiguity. A parabolic curve will be a strongclue to correspond a point in one image to a point in theother image, where both points represent the same point onobject surface. By comparing the degree of polarization atsuch corresponding points, the true surface normal can bedetermined."
ICCV	Polarization-based Inverse Rendering from a Single View.	Daisuke Miyazaki,Robby T. Tan,Kenji Hara,Katsushi Ikeuchi	2003	This paper presents a method to estimate geometrical,photometrical, and environmental information of a single-viewedobject in one integrated framework under fixed viewingposition and fixed illumination direction. These threetypes of information are important to render a photorealisticimage of a real object. Photometrical information representsthe texture and the surface roughness of an object,while geometrical and environmental information representthe 3D shape of an object and the illumination distribution,respectively. The proposed method estimates the 3D shapeby computing the surface normal from polarization data,calculates the texture of the object from the diffuse only reflectioncomponent, determines the illumination directionsfrom the position of the brightest intensity in the specularreflection component, and finally computes the surfaceroughness of the object by using the estimated illuminationdistribution.
ICCV	Background Modeling and Subtraction of Dynamic Scenes.	Antoine Monnet,Anurag Mittal,Nikos Paragios,Visvanathan Ramesh	2003	Background modeling and subtraction is a core componentin motion analysis. The central idea behind such moduleis to create a probabilistic representation of the staticscene that is compared with the current input to performsubtraction. Such approach is efficient when the scene to bemodeled refers to a static structure with limited perturbation.In this paper, we address the problem of modeling dynamicscenes where the assumption of a static backgroundis not valid. Waving trees, beaches, escalators, naturalscenes with rain or snow are examples. Inspired by the workproposed in [4], we propose an on-line auto-regressivemodel to capture and predict the behavior of such scenes.Towards detection of events we introduce a new metric thatis based on a state-driven comparison between the predictionand the actual frame. Promising results demonstratethe potentials of the proposed framework.
ICCV	Dominant Sets and Hierarchical Clustering.	Massimiliano Pavan,Marcello Pelillo	2003	Dominant sets are a new graph-theoretic concept that has proven to be relevant in partitional (flat) clustering as well as image segmentation problems. However, in many computer vision applications, such as the organization of an image database, it is important to provide the data to be clustered with a hierarchical organization, and it is not clear how to do this within the dominant set framework. In this paper we address precisely this problem, and present a simple and elegant solution to it. To this end, we consider a family of (continuous) quadratic programs which contain a parameterized regularization term that controls the global shape of the energy landscape. When the regularization parameter is zero the local solutions are known to be in one-to-one correspondence with dominant sets, but when it is positive an interesting picture emerges. We determine bounds for the regularization parameter that allow us to exclude from the set of local solutions those inducing clusters of size smaller than a prescribed threshold. This suggests a new (divisive) hierarchical approach to clustering, which is based on the idea of properly varying the regularization parameter during the clustering process. Straight forward dynamics from evolutionary game theory are used to locate the solutions of the quadratic programs at each level of the hierarchy. We apply the proposed framework to the problem of organizing a shape database. Experiments with three different similarity matrices (and databases) reported in the literature have been conducted, and the results confirm the effectiveness of our approach.
ICCV	Image Spaces and Video Trajectories: Using Isomap to Explore Video Sequences.	Robert Pless	2003	Dimensionality reduction techniques seek to representa set of images as a set of points in a low dimensionalspace.Here we explore a video representation thatconsiders a video as two parts - a space of possibleimages and a trajectory through that space.The non-lineardimensionality reduction technique of Isomap,gives, for many interesting scenes, a very low dimensionalrepresentation of the space of possible images.Analysis of the shape of the video trajectory throughthese image spaces gives new tools for video analysis.Experiments with natural video sequences illustratemethods for the very different tasts of classifyingvideo clips and temporal super-resolution.
ICCV	A Class of Photometric Invariants: Separating Material from Shape and Illumination.	Srinivasa G. Narasimhan,Visvanathan Ramesh,Shree K. Nayar	2003	"We derive a new class of photometric invariants that can beused for a variety of vision tasks including lighting invariantmaterial segmentation, change detection and tracking, aswell as material invariant shape recognition. The key ideais the formulation of a scene radiance model for the class of""separable"" BRDFs, that can be decomposed into materialrelated terms and object shape and lighting related terms.All the proposed invariants are simple rational functions ofthe appearance parameters (say, material or shape and lighting).The invariants in this class differ from one another in thenumber and type of image measurements they require. Mostof the invariants in this class need changes in illumination orobject position between image acquisitions. The invariantscan handle large changes in lighting which pose problems formost existing vision algorithms. We demonstrate the power ofthese invariants using scenes with complex shapes, materials,textures, shadows and specularities."
ICCV	How To Deal with Point Correspondences and Tangential Velocities in the Level Set Framework.	Jean-Philippe Pons,Gerardo Hermosillo,Renaud Keriven,Olivier D. Faugeras	2003	"In this paper, we overcome a major drawback of the levelset framework: the lack of point correspondences. We maintainexplicit backward correspondences from the evolvinginterface to the initial one by advecting the initial point coordinateswith the same speed as the level set function. Ourmethod leads to a system of coupled Eulerian partial differentialequations. We show in a variety of numerical experimentsthat it can handle both normal and tangential velocities,large deformations, shocks, rarefactions and topologicalchanges. Applications are many in computer vision andelsewhere since our method can upgrade virtually any levelset evolution. We complement our work with the design ofnon zero tangential velocities that preserve the relative areaof interface patches; this feature may be crucial in such applicationsas computational geometry, grid generation orunfolding of the organs' surfaces, e.g. brain, in medicalimaging."
ICCV	Adaptive Dynamic Range Imaging: Optical Control of Pixel Exposures Over Space and Time.	Shree K. Nayar,Vlad Branzoi	2003	This paper presents a new approach to imaging thatsignificantly enhances the dynamic range of a camera.The key idea is to adapt the exposure of each pixel onthe image detector, based on the radiance value of thecorresponding scene point. This adaptation is done inthe optical domain, that is, during image formation. Inpractice, this is achieved using a spatial light modulatorwhose transmittance can be varied with high resolutionover space and time. A real-time control algorithm isdeveloped that uses acquired images to automaticallyadjust the transmittance function of the spatial modulator. Each captured image and its corresponding transmittance function are used to compute a very high dynamic range image that is linear in scene radiance.We have implemented a video-rate adaptive dynamicrange camera that consists of a color CCD detector anda controllable liquid crystal light modulator. Experiments have been conducted in scenarios with complexand harsh lighting conditions. The results indicate thatadaptive imaging can have a significant impact on visionapplications such as monitoring, tracking, recognition,and navigation.
ICCV	Variational Stereovision and 3D Scene Flow Estimation with Statistical Similarity Measures.	Jean-Philippe Pons,Renaud Keriven,Olivier D. Faugeras,Gerardo Hermosillo	2003	We present a common variational framework for dense depth recovery and dense three-dimensional motion field estimation from multiple video sequences, which is robustto camera spectral sensitivity differences and illumination changes. For this purpose, we first show that both problems reduce to a generic image matching problem after backprojecting the input images onto suitable surfaces. We then solve this matching problem in the case of statistical similarity criteria that can handle frequently occurring non-affine image intensities dependencies. Our method leads to an efficient and elegant implementation based on fast recursive filters. We obtain good results on real images.
ICCV	"""Perspective Shape from Shading"" and Viscosity Solutions."	Emmanuel Prados,Olivier D. Faugeras	2003	"This article proposes a solution of the Lambertian shapefrom shading (SFS) problem in the case of a pinhole cameramodel (performing a perspective projection). Our approachis based upon the notion of viscosity solutions of Hamilton-Jacobiequations. This approach allows us to naturallydeal with nonsmooth solutions and provides a mathematicalframework for proving correctness of our algorithms.Our work extends previous work in the area in three aspects.First, it models the camera as a pinhole whereasmost authors assume an orthographic projection (see [15]for a panorama of the SFS problem up to 1989 and [29, 17]for a recent survey), thereby extending the applicability ofshape from shading methods to more realistic images. Inparticular it extends the work of [24] and [26]. Second, byadapting the brightness equation to the perspective problem,we obtain a new partial differential equation (PDE).Results about the existence and uniqueness of its solutionare also obtained. Third, it allows us to come up with a newapproximation scheme and a new algorithm for computingnumerical approximations of the ""continuous"" solution aswell as a proof of their convergence toward that solution."
ICCV	Eye Design in the Plenoptic Space of Light Rays.	Jan Neumann,Cornelia Fermüller,Yiannis Aloimonos	2003	Natural eye designs are optimized with regard to the tasksthe eye-carrying organism has to perform for survival. Thisoptimization has been performed by the process of naturalevolution over many millions of years. Every eye capturesa subset of the space of light rays. The information containedin this subset and the accuracy to which the eye canextract the necessary information determines an upper limiton how well an organism can perform a given task. In thiswork we propose a new methodology for camera design. Byinterpreting eyes as sample patterns in light ray space wecan phrase the problem of eye design in a signal processingframework. This allows us to develop mathematical criteriafor optimal eye design, which in turn enables us to build thebest eye for a given task without the trial and error phase ofnatural evolution. The principle is evaluated on the task of3D ego-motion estimation.
ICCV	Automatic Video Summarization by Graph Modeling.	Chong-Wah Ngo,Yu-Fei Ma,HongJiang Zhang	2003	We propose a unified approach for summarization based on the analysis of video structures and video highlights. Our approach emphasizes both the content balance and perceptual quality of a summary. Normalized cut algorithm is employed to globally and optimally partition a video into clusters. A motion attention model based on human perception is employed to compute the perceptual quality of shots and clusters. The clusters, together with the computed attention values, form a temporal graph similar to Markov chain that inherently describes the evolution and perceptual importance of video clusters. In our application, the flow of a temporal graph is utilized to group similar clusters into scenes, while the attention values are used as guidelines to select appropriate sub-shots in scenes for summarization.
ICCV	Phenomenological Eigenfunctions for Image Irradiance.	Peter Nillius,Jan-Olof Eklundh	2003	We present a framework for calculating low-dimensional bases to represent image irradiance from surfaces with isotropic reflectance under arbitrary illumination. By representing the illumination and the bidirectional reflectance distribution function (BRDF) in frequency space, a model for the image irradiance is derived. This model is then reduced in dimensionality by analytically constructing the principal component basis for all images given the variations in both the illumination and the surface material. The principal component basis are constructed in such a way that all the symmetries (Helmholtz reciprocity and isotropy) of the BRDF are preserved in the basis functions. Using the framework we calculate a basis using a database of natural illumination and the CURET database containing BRDFs of real world surface materials.
ICCV	Preemptive RANSAC for Live Structure and Motion Estimation.	David Nistér	2003	A system capable of performing robust live ego-motion estimation for perspective cameras is presented. The system is powered by random sample consensus with preemptive scoring of the motion hypotheses. A general statement of the problem of efficient preemptive scoring is given. Then a theoretical investigation of preemptive scoring under a simple inlier-outlier model is performed. A practical preemption scheme is proposed and it is shown that the preemption is powerful enough to enable robust live structure and motion estimation.
ICCV	The Catchment Feature Model for Multimodal Language Analysis.	Francis K. H. Quek	2003	"The Catchment Feature Model (CFM) addresses two questions in multimodal interaction: how do we bridge video and audio processing with the realities of human multimodal communication, and how information from the different modes may be fused. We discuss the need for our model, motivate the CFM from psycholinguistic research, and present the Model. In contrast to 'whole gesture' recognition, the CFM applies a feature decomposition approach that facilitates cross-modal fusion at the level of discourse planning and conceptualization. We present our experimental framework for CFM-based research, and cite three concrete examples of Catchment Features (CF), and propose new directions of multimodal research based on the model."
ICCV	Obstacle Detection Using Projective Invariant and Vanishing Lines.	Ryuzo Okada,Yasuhiro Taniguchi,Kenji Furukawa,Kazunori Onoguchi	2003	This paper presents a novel method for detecting vehicles as obstacles in various road scenes using a single on-board camera. Vehicles are detected by testing whether the motion of a set of three horizontal line segments, which are always on the vehicles, satisfies the motion constraint of the ground plane or that of the surface plane of the vehicles. The motion constraint of each plane is derived from the projective invariant combined with the vanishing line of the plane that is a prior knowledge of road scenes. The proposed method is implemented into a newly developed on-board LSI. Experimental results for real road scenes under various conditions show the effectiveness of the proposed method.
ICCV	Autocalibration of a Projector-Screen-Camera System: Theory and Algorithm for Screen-to-Camera Homography Estimation.	Takayuki Okatani,Koichiro Deguchi	2003	This paper deals with the autocalibration of a systemthat consists of a planar screen, multiple projectors, and acamera. In the system, either multiple projectors or a singlemoving projector projects patterns on a screen while a stationarycamera placed in front of the screen takes images ofthe patterns. We treat the case in which the patterns that theprojectors project toward space are assumed to be known(i.e., the projectors are calibrated), whereas poses of theprojectors are unknown. Under these conditions, we considerthe problem of estimating screen-to-camera homographyfrom the images alone. This is intended for cases wherethere is no clue on the screen surface that enables directestimation of the screen-to-camera homography. One applicationis a 6DOF input device; poses of a multi-beamprojector freely moving in space are computed from the imagesof beam spots on the screen. The primary contributionof the paper is theoretical results on the uniqueness of solutionsand a noniterative algorithm for the problem. Theeffectiveness of the method is shown by experimental resultson synthetic as well as on real images.
ICCV	Using Temporal Coherence to Build Models of Animals.	Deva Ramanan,David A. Forsyth	2003	This paper describes a system that can build appearance models of animals automatically from a video sequence of the relevant animal with no explicit supervisory information. The video sequence need not have any form of special background. Animals are modeled as a 2D kinematic chain of rectangular segments, where the number of segments and the topology of the chain are unknown. The system detects possible segments, clusters segments whose appearance is coherent over time, and then builds a spatial model of such segment clusters. The resulting representation of the spatial configuration of the animal in each frame can be seen either as a track - in which case the system described should be viewed as a generalized tracker, that is capable of modeling objects while tracking them - or as the source of an appearance model which can be used to build detectors for the particular animal. This is because knowing a video sequence is temporally coherent - i.e. that a particular animal is present through the sequence - is a strong supervisory signal. The method is shown to be successful as a tracker on video sequences of real scenes showing three different animals. For the same reason it is successful as a tracker, the method results in detectors that can be used to find each animal fairly reliably within the Corel collection of images.
ICCV	A Bayesian Network Framework for Relational Shape Matching.	Anand Rangarajan,James M. Coughlan,Alan L. Yuille	2003	A Bayesian network formulation for relational shapematching is presented. The main advantage of the relationalshape matching approach is the obviation ofthe non-rigid spatial mappings used by recent non-rigidmatching approaches. The basic variables that need tobe estimated in the relational shape matching objectivefunction are the global rotation and scale and the localdisplacements and correspondences. The new Bethefree energy approach is used to estimate the pairwisecorrespondences between links of the template graphsand the data. The resulting framework is useful inboth registration and recognition contexts. Results areshown on hand-drawn templates and on 2D transverseT1-weighted MR images.
ICCV	View-invariant Alignment and Matching of Video Sequences.	Cen Rao,Alexei Gritai,Mubarak Shah,Tanveer Fathima Syeda-Mahmood	2003	In this paper, we propose a novel method to establish temporalcorrespondence between the frames of two videos. 3D epipolargeometry is used to eliminate the distortion generated bythe projection from 3D to 2D. Although the fundamental matrixcontains the extrinsic property of the projective geometrybetween views, it is sensitive to noise. Therefore, wepropose the use of a rank constraint of corresponding pointsin two views to measure the similarity between trajectories.This rank constraint shows more robustness and avoids computationof the fundamental matrix. A dynamic programmingapproach using the similarity measurement is proposed to findthe non-linear time-warping function for videos containinghuman activities. In this way, videos of different individualstaken at different times and from distinct viewpoints canbe synchronized. A temporal pyramid of trajectories is appliedto improve the accuracy of the view-invariant dynamictime-warping approach. We show various applications of thisapproach such as video synthesis, human action recognition,and computer aider training. Compared to state-of-the-arttechniques, our method shows a great improvement.
ICCV	Learning a Classification Model for Segmentation.	Xiaofeng Ren,Jitendra Malik	2003	We propose a two-class classification model for grouping. Human segmented natural images are used as positive examples. Negative examples of grouping are constructed by randomly matching human segmentations and images. In a preprocessing stage an image is over segmented into superpixels. We define a variety of features derived from the classical Gestalt cues, including contour, texture, brightness and good continuation. Information-theoretic analysis is applied to evaluate the power of these grouping cues. We train a linear classifier to combine these features. To demonstrate the power of the classification model, a simple algorithm is used to randomly search for good segmentations. Results are shown on a wide range of images.
ICCV	A Non-Iterative Greedy Algorithm for Multi-frame Point Correspondence.	Khurram Shafique,Mubarak Shah	2003	This paper presents a framework for finding point correspondences in monocular image sequences over multiple frames.The general problem of multi-frame point correspondence is NP Hard for three or more frames. A polynomial time algorithm for a restriction of this problem is presented, and is used as the basis of proposed greedy algorithm for the general problem. The greedy nature of the proposed algorithm allows it to be used in real time systems for tracking and surveillance etc. In addition, the proposed algorithm deals with the problems of occlusion, missed detections, and false positives, by using a single non-iterative greedy optimization scheme, and hence, reduces the complexity of the overall algorithm as compared to most existing approaches, where multiple heuristics are used for the same purpose.While most greedy algorithms for point tracking do not allow for entry and exit of points from the scene, this is not a limitation for the proposed algorithm. Experiments with real and synthetic data show that the proposed algorithm outperforms the existing techniques and is applicable in more general settings.
ICCV	Edit Distance From Graph Spectra.	Antonio Robles-Kelly,Edwin R. Hancock	2003	This paper is concerned with computing graph edit distance. One of the criticisms that can be leveled at existing methods for computing graph edit distance is that it lacks the formality and rigour of the computation of string edit distance. Hence, our aim is to convert graphs to string sequences so that standard string edit distance techniques can be used. To do this we use graph spectral seriation method to convert the adjacency matrix into a string or sequence order. We pose the problem of graph-matching as maximum aposteriori probability alignment of the seriation sequences for pairs of graphs. This treatment leads to an expression for the edit costs. We compute the edit distance by finding the sequence of string edit operations which minimise the cost of the path traversing the edit lattice. The edit costs are defined in terms of the a posteriori probability of visiting a site on the lattice. We demonstrate the method with results on a data-set of Delaunay graphs.
ICCV	Efficient, Robust and Accurate Fitting of a 3D Morphable Model.	Sami Romdhani,Thomas Vetter	2003	3D Morphable Models, as a means to generate images of a class of objects and to analyze them, have become increasingly popular. The problematic part of this frameworkis the registration of the model to an image, a.k.a. the fitting. The characteristic features of a fitting algorithm are its efficiency, robustness, accuracy and automation. Many accurate algorithms based on gradient descent techniques exist which are unfortunately short on the other features. Recently, an efficient algorithm called Inverse Compositional Image Alignment (ICIA) algorithm, able to fit 2D images, was introduced. In this paper, we extent this algorithm to fit 3D Morphable Models using a novel mathematical notation which facilitates the formulation of the fitting problem. This formulation enables us to avoid a simplification so far used in the ICIA, being as efficient and leading to improved fitting precision. Additionally, the algorithm is robust without sacrificing its efficiency and accuracy, thereby conforming to three of the four characteristics of a good fitting algorithm.
ICCV	Unsupervised Image Translation.	Rómer Rosales,Kannan Achan,Brendan J. Frey	2003	An interesting and potentially useful vision/graphics task is to render an input image in an enhanced form or also in an unusual style; for example with increased sharpness or with some artistic qualities. In previous work [10, 5],researchers showed that by estimating the mapping from an input image to a registered (aligned) image of the same scene in a different style or resolution, the mapping could be used to render a new input image in that style or resolution. Frequently a registered pair is not available, but instead the user may have only a source image of an unrelated scene that contains the desired style. In this case, the task of inferring the output image is much more difficult since the algorithm must both infer correspondences between features in the input image and the source image, and infer the unknown mapping between the images. We describe a Bayesian technique for inferring the most likely output image. The prior on the output image P ( X )is a patch-based Markov random field obtained from the source image. The likelihood of the input P ( Y|X )is a Bayesian network that can represent different rendering styles. We describe a computationally efficient, probabilistic inference and learning algorithm for inferring the most likely output image and learning the rendering style. We also show that current techniques for image restoration or reconstruction proposed in the vision literature (e.g., image super-resolution or de-noising) and image-based non-photorealistic rendering could be seen as special cases of our model. We demonstrate our technique using several tasks, including rendering a photograph in the artistic style of an unrelated scene, de-noising, and texture transfer.
ICCV	Learning and Inferring Image Segmentations using the GBP Typical Cut Algorithm.	Noam Shental,Assaf Zomet,Tomer Hertz,Yair Weiss	2003	"Significant progress in image segmentation has beenmade by viewing the problem in the framework of graphpartitioning. In particular, spectral clustering methods suchas ""normalized cuts"" (ncuts) can efficiently calculate goodsegmentations using eigenvector calculations. However,spectral methods when applied to images with local connectivityoften oversegment homogenous regions. More importantly,they lack a straightforward probabilistic interpretationwhich makes it difficult to automatically set parametersusing training data.In this paper we revisit the typical cut criterion proposedin [1, 5]. We show that computing the typical cut isequivalent to performing inference in an undirected graphicalmodel. This equivalence allows us to use the powerfulmachinery of graphical models for learning and inferringimage segmentations. For inferring segmentations weshow that the generalized belief propagation (GBP) algorithmcan give excellent results with a runtime that is usuallyfaster than the ncut eigensolver. For learning segmentationswe derive a maximum likelihood learning algorithmto learn affinity matrices from labelled datasets. We illustrateboth learning and inference on challenging real andsynthetic images."
ICCV	Linear Multi-View Reconstruction of Points, Lines, Planes and Cameras using a Reference Plane.	Carsten Rother	2003	This paper presents a new linear method for reconstructingsimultaneously 3D features (points, lines and planes)and cameras from many perspective views by solving a singlelinear system. It assumes that a real or virtual referenceplane is visible in all views. We call it the Direct ReferencePlane (DRP) method. It is well known that the projection relationshipbetween uncalibrated cameras and 3D featuresis non-linear in the absence of a reference plane. With aknown reference plane, points and cameras have a linearrelationship, as shown in [16]. The main contribution ofthis paper is that lines and cameras, as well as, planes andcameras also have a linear relationship. Consequently, all3D features and all cameras can be reconstructed simultaneouslyfrom a single linear system, which handles missingimage measurements naturally. A further contribution is anextensive experimental comparison, using real data, of differentreference plane and non-reference plane reconstructionmethods. For difficult reference plane scenarios, withpoint or line features, the DRP method is superior to allcompared methods. Finally, an extensive list of referenceplane scenarios is presented, which shows the wide applicabilityof the DRP method.
ICCV	A New Paradigm for Recognizing 3-D Object Shapes from Range Data.	Salvador Ruiz-Correa,Linda G. Shapiro,Marina Meila	2003	Most of the work on 3-D object recognition from rangedata has used an alignment-verification approach in whicha specific 3-D object is matched to an exact instance of thesame object in a scene. This approach has been successfullyused in industrial machine vision, but it is not capable ofdealing with the complexities of recognizing classes of similarobjects. This paper undertakes this task by proposingand testing a component-based methodology encompassingthree main ingredients: 1) a new way of learning and extractingshape-class components from surface shape information;2) a new shape representation called a symbolicsurface signature that summarizes the geometric relationshipsamong components; and 3) an abstract representationof shape classes formed by a hierarchy of classifiersthat learn object-class parts and their spatial relationshipsfrom examples.
ICCV	Fast Intensity-based 2D-3D Image Registration of Clinical Data Using Light Fields.	Daniel B. Russakoff,Torsten Rohlfing,Calvin R. Maurer Jr.	2003	Registration of a preoperative CT (3D) image to one or more X-ray projection (2D) images, a special case of the pose estimation problem, has been attempted in a variety of ways with varying degrees of success. Recently, there has been a great deal of interest in intensity-based methods. One of the drawbacks to such methods is the need to create digitally reconstructed radiographs (DRRs) at each step of the optimization process. DRRs are typically generated by ray casting, an operation that requires O(n3) time, where we assume that n is approximately the size (in voxels) of one side of the DRR as well as one side of the CT volume. We address this issue by extending light field rendering techniques from the computer graphics community to generate DRRs instead of conventional rendered images. Using light fields allows most of the computation to be performed in a preprocessing step; after this precomputation, very accurate DRRs can be generated in O(n2) time. Another important issue for 2D-3D registration algorithms is validation. Previously reported 2D-3D registration algorithms were validated using synthetic data or phantoms but not clinical data. We present an intensity-based 2D-3D registration system that generates DRRs using light fields; we validate its performance using clinical data with a known gold standard transformation.
ICCV	Dense Shape Reconstruction of a Moving Object under Arbitrary, Unknown Lighting.	Denis Simakov,Darya Frolova,Ronen Basri	2003	We present a method for shape reconstruction from severalimages of a moving object. The reconstruction is dense(up to image resolution). The method assumes that themotion is known, e.g., by tracking a small number of featurepoints on the object. The object is assumed Lambertian(completely matte), light sources should not be veryclose to the object but otherwise arbitrary, and no knowledgeof lighting conditions is required. An object changesits appearance significantly when it changes its orientationrelative to light sources, causing violation of the commonbrightness constancy assumption. While a lot of effort isdevoted to deal with this violation, we demonstrate howto exploit it to recover 3D structure from 2D images. Wepropose a new correspondence measure that enables pointmatching across views of a moving object. The method hasbeen tested both on computer simulated examples and on areal object.
ICCV	Regression based Bandwidth Selection for Segmentation using Parzen Windows.	Maneesh Kumar Singh,Narendra Ahuja	2003	We consider the problem of segmentation of images that can be modelled as piecewise continuous signals having unknown, non-stationary statistics. We propose a solution to this problem which first uses a regression framework to estimate the image PDF, and then mean-shift to find the modes of this PDF. The segmentation follows from mode identification wherein pixel clusters or image segments are identified with unique modes of the multi-modal PDF. Each pixel is mapped to a mode using a convergent, iterative process. The effectiveness of the approach depends upon the accuracy of the (implicit) estimate of the underlying multi-modal density function and thus on the bandwidth parameters used for its estimate using Parzen windows. Automatic selection of bandwidth parameters is a desired feature of the algorithm. We show that the proposed regression-based model admits a realistic framework to automatically choose bandwidth parameters which minimizes a global error criterion. We validate the theory presented with results on real images.
ICCV	Appearance Sampling for Obtaining A Set of Basis Images for Variable Illumination.	Imari Sato,Takahiro Okabe,Yoichi Sato,Katsushi Ikeuchi	2003	"Previous studies have demonstrated that the appearance ofan object under varying illumination conditions can be representedby a low-dimensional linear subspace. A set ofbasis images spanning such a linear subspace can be obtainedby applying the principal component analysis (PCA)for a large number of images taken under different lightingconditions. While the approaches based on PCA havebeen used successfully for object recognition under varyingillumination conditions, little is known about how many imageswould be required in order to obtain the basis imagescorrectly. In this study, we present a novel method for analyticallyobtaining a set of basis images of an object forarbitrary illumination from input images of the object takenunder a point light source. The main contribution of ourwork is that we show that a set of lighting directions canbe determined for sampling images of an object dependingon the spectrum of the object's BRDF in the angularfrequency domain such that a set of harmonic images canbe obtained analytically based on the sampling theorem onspherical harmonics. In addition, unlike the previously proposedtechniques based on spherical harmonics, our methoddoes not require the 3D shape and reflectance properties ofan object used for rendering harmonics images of the objectsynthetically."
ICCV	Video Google: A Text Retrieval Approach to Object Matching in Videos.	Josef Sivic,Andrew Zisserman	2003	We describe an approach to object and scene retrievalwhich searches for and localizes all the occurrences of auser outlined object in a video. The object is represented bya set of viewpoint invariant region descriptors so that recognitioncan proceed successfully despite changes in viewpoint,illumination and partial occlusion. The temporalcontinuity of the video within a shot is used to track theregions in order to reject unstable regions and reduce theeffects of noise in the descriptors.The analogy with text retrieval is in the implementationwhere matches on descriptors are pre-computed (using vectorquantization), and inverted file systems and documentrankings are used. The result is that retrieval is immediate,returning a ranked list of key frames/shots in the manner ofGoogle.The method is illustrated for matching on two full lengthfeature films.
ICCV	Weighted and Robust Incremental Method for Subspace Learning.	Danijel Skocaj,Ales Leonardis	2003	Visual learning is expected to be a continuous and robustprocess, which treats input images and pixels selectively.In this paper we present a method for subspace learning,which takes these considerations into account. Wepresent an incremental method, which sequentially updatesthe principal subspace considering weighted influence ofindividual images as well as individual pixels within an image.This approach is further extended to enable determinationof consistencies in the input data and imputation of thevalues in inconsistent pixels using the previously acquiredknowledge, resulting in a novel incremental, weighted androbust method for subspace learning.
ICCV	Fast Pose Estimation with Parameter-Sensitive Hashing.	Gregory Shakhnarovich,Paul A. Viola,Trevor Darrell	2003	Example-based methods are effective for parameter estimationproblems when the underlying system is simple orthe dimensionality of the input is low. For complex andhigh-dimensional problems such as pose estimation, thenumber of required examples and the computational complexityrapidly become prohibitively high. We introduce anew algorithm that learns a set of hashing functions that efficientlyindex examples in a way relevant to a particularestimation task. Our algorithm extends locality-sensitivehashing, a recently developed method to find approximateneighbors in time sublinear in the number of examples. Thismethod depends critically on the choice of hash functions;we show how to find the set of hash functions that are optimallyrelevant to a particular estimation problem. Experimentsdemonstrate that the resulting algorithm, which wecall Parameter-Sensitive Hashing, can rapidly and accuratelyestimate the articulated pose of human figures from alarge database of example images.
ICCV	Computing MAP trajectories by representing, propagating and combining PDFs over groups.	Paul Smith,Tom Drummond,Kimon Roussopoulos	2003	This paper addresses the problem of computing the trajectoryof a camera from sparse positional measurementsthat have been obtained from visual localisation, and densedifferential measurements from odometry or inertial sensors.A fast method is presented for fusing these two sourcesof information to obtain the maximum a posteriori estimateof the trajectory. A formalism is introduced for representingprobability density functions over Euclidean transformations,and it is shown how these density functions can bepropagated along the data sequence and how multiple estimatesof a transformation can be combined. A three-passalgorithm is described which makes use of these results toyield the trajectory of the camera.Simulation results are presented which are validatedagainst a physical analogue of the vision problem, and resultsare then shown from sequences of approximately 1,800frames captured from a video camera mounted on a go-kart.Several of these frames are processed using computer visionto obtain estimates of the position of the go-kart. The algorithmfuses these estimates with odometry from the entiresequence in 150 mS to obtain the trajectory of the kart.
ICCV	Image Statistics and Anisotropic Diffusion.	Hanno Scharr,Michael J. Black,Horst W. Haussecker	2003	"Many sensing techniques and image processing applicationsare characterized by noisy, or corrupted, image data.Anisotropic diffusion is a popular, and theoretically wellunderstood, technique for denoising such images. Diffusionapproaches however require the selection of an ""edgestopping"" function, the definition of which is typically adhoc. We exploit and extend recent work on the statisticsof natural images to define principled edge stopping functionsfor different types of imagery. We consider a varietyof anisotropic diffusion schemes and note that they computespatial derivatives at fixed scales from which we estimatethe appropriate algorithm-specific image statistics. Goingbeyond traditional work on image statistics, we also modelthe statistics of the eigenvalues of the local structure tensor.Novel edge-stopping functions are derived from these imagestatistics giving a principled way of formulating anisotropicdiffusion problems in which all edge-stopping parametersare learned from training data."
ICCV	A Theory of Multiplexed Illumination.	Yoav Y. Schechner,Shree K. Nayar,Peter N. Belhumeur	2003	Imaging of objects under variable lighting directions is animportant and frequent practice in computer vision andimage-based rendering. We introduce an approach that significantlyimproves the quality of such images. Traditionalmethods for acquiring images under variable illuminationdirections use only a single light source per acquired image.In contrast, our approach is based on a multiplexing principle,in which multiple light sources illuminate the objectsimultaneously from different directions. Thus, the objectirradiance is much higher. The acquired images are thencomputationally demultiplexed. The number of image acquisitionsis the same as in the single-source method. Theapproach is useful for imaging dim object areas. We givethe optimal code by which the illumination should be multiplexedto obtain the highest quality output. For n imagescorresponding to n light sources, the noise is reduced by\sqrt n/2 relative to the signal. This noise reduction translatesto a faster acquisition time or an increase in density of illuminationdirection samples. It also enables one to use lightingwith high directional resolution using practical setups,as we demonstrate in our experiments.
ICCV	Tales of Shape and Radiance in Multi-view Stereo.	Stefano Soatto,Anthony J. Yezzi,Hailin Jin	2003	To what extent can three-dimensional shape and radiancebe inferred from a collection of images? Can the two be estimatedseparately while retaining optimality? How shouldthe optimality criterion be computed? When is it necessaryto employ an explicit model of the reflectance properties ofa scene? In this paper we introduce a separation principlefor shape and radiance estimation that applies to Lambertianscenes and holds for any choice of norm. When thescene is not Lambertian, however, shape cannot be decoupledfrom radiance, and therefore matching image-to-imageis not possible directly. We employ a rank constraint onthe radiance tensor, which is commonly used in computergraphics, and construct a novel cost functional whose minimizationleads to an estimate of both shape and radiancefor non-Lambertian objects, which we validate experimentally.
ICCV	The Beltrami Flow over Implicit Manifolds.	Nir A. Sochen,Rachid Deriche,Lucero Lopez Perez	2003	In many medical computer vision tasks the relevant data isattached to a specific tissue such as the colon or the cortex.This situation calls for regularization techniques whichare defined over surfaces.We introduce in this paper theBeltrami flow over implicit manifolds.This new regularizationtechnique overcomes the over-smoothing of the L2flow and the staircasing effects of the L1 flow, that wererecently suggested via the harmonic map methods.The keyof our approach is first to clarify the link between the intrinsic Polyakov action and the implicit Harmonic energyfunctional and then use the geometrical understanding ofthe Beltrami Flow to generalize it to images on implicitlydefined non flat surfaces.It is shown that once again theBeltrami flow interpolates between the L2 and L1 flows onnon-flat surfaces.The implementation scheme of this flowis presented and various experimental results obtained on aset of various real images illustrate the performances of theapproach as well as the differences with the harmonic mapflows.This extension of the Beltrami flow to the case of nonflat surfaces opens new perspectives in the regularization ofnoisy data defined on manifolds.
ICCV	Model-Based Multiple View Reconstruction of People.	Jonathan Starck,Adrian Hilton	2003	This paper presents a framework to reconstruct a scenecaptured in multiple camera views based on a prior modelof the scene geometry. The framework is applied to thecapture of animated models of people. A multiple camerastudio is used to simultaneously capture a moving personfrom multiple viewpoints. A humanoid computer graphicsmodel is animated to match the pose at each time frame.Constrained optimisation is then used to recover the multipleview correspondence from silhouette, stereo and featurecues, updating the geometry and appearance of the model.The key contribution of this paper is a model-based computervision framework for the reconstruction of shape andappearance from multiple views. This is compared to currentmodel-free approaches for multiple view scene capture.The technique demonstrates improved scene reconstructionin the presence of visual ambiguities and providesthe means to capture a dynamic scene with a consistentmodel that is instrumented with an animation structure toedit the scene dynamics or to synthesise new content.
ICCV	Minimally-Supervised Classification using Multiple Observation Sets.	Chris Stauffer	2003	This paper discusses building complex classifiers from a single labeled example and vast number of unlabeled observation sets, each derived from observation of a single processor object. When data can be measured by observation, it is often plentiful and it is often possible to make more than one observation of the state of a process or object. This paper discusses how to exploit the variability across such sets of observations of the same object to estimate class labels for unlabeled examples given a minimal number of labeled examples. In contrast to similar semi-supervised classification procedures that define the likelihood that two observations share a label as a function of the embedded distance between the two observations, this method uses the Naive Bayes estimate of how often the two observations did result from the same observed process. Exploiting this additional source of information in an iterative estimation procedure can generalize complex classification models from single labeled observations. Some examples involving classification of tracked objects in a low-dimensional feature space given thousands of unlabeled observation sets are used to illustrate the effectiveness of this method.
ICCV	Spectral Partitioning for Structure from Motion.	Drew Steedly,Irfan A. Essa,Frank Dellaert	2003	We propose a spectral partitioning approach for large-scaleoptimization problems, specifically structure from motion.In structure from motion, partitioning methods reduce theproblem into smaller and better conditioned subproblemswhich can be efficiently optimized.Our partitioning methoduses only the Hessian of the reprojection error and its eigenvector.We show that partitioned systems that preserve theeigenvectors corresponding to small eigenvalues result inlower residual error when optimized.We create partitionsby clustering the entries of the eigenvectors of the Hessiancorresponding to small eigenvalues.This is a more generaltechnique than relying on domain knowledge and heuristicssuch as bottom-up structure from motion approaches.Simultaneously,it takes advantage of more information thangeneric matrix partitioning algorithms.
ICCV	Filtering Using a Tree-Based Estimator.	Bjoern Stenger,Arasanathan Thayananthan,Philip H. S. Torr,Roberto Cipolla	2003	Within this paper a new framework for Bayesian tracking ispresented, which approximates the posterior distribution atmultiple resolutions. We propose a tree-based representationof the distribution, where the leaves define a partition ofthe state space with piecewise constant density. The advantageof this representation is that regions with low probabilitymass can be rapidly discarded in a hierarchical search,and the distribution can be approximated to arbitrary precision.We demonstrate the effectiveness of the technique byusing it for tracking 3D articulated and non-rigid motionin front of cluttered background. More specifically, we areinterested in estimating the joint angles, position and orientationof a 3D hand model in order to drive an avatar.
ICCV	Dense Matching of Multiple Wide-baseline Views.	Christoph Strecha,Tinne Tuytelaars,Luc J. Van Gool	2003	This paper describes a PDE-based method for densedepth extraction from multiple wide-baseline images. Emphasislies on the usage of only a small amount of images.The integration of these multiple wide-baseline views isguided by the relative confidence that the system has in thematching to different views. This weighting is fine-grainedin that it is determined for every pixel at every iteration.Reliable information spreads fast at the expense of less reliabledata, both in terms of spatial communications withina view and in terms of information exchange between theviews. Changes in intensity between images can be handledin a similar fine grained fashion.
ICCV	Video Input Driven Animation (VIDA).	Meng Sun,Allan D. Jepson,Eugene Fiume	2003	There are many challenges associated with the integration of synthetic and real imagery. One particularly difficult problem is the automatic extraction of salient parameters of natural phenomena in real video footage for subsequent application to synthetic objects. Can we ensure that the hair and clothing of a synthetic actor placed in a meadow of swaying grass will move consistently with the wind that moved that grass? The video footage can be seen as a controller for the motion of synthetic features, a concept we call video input driven animation (VIDA). We propose a schema that analyzes an input video sequence, extracts parameters from the motion of objects in the video, and uses this information to drive the motion of synthetic objects. To validate the principles of VIDA, we approximate the inverse problem to harmonic oscillation, which we use to extract parameters of wind and of regular water waves. We observe the effect of wind on a tree in a video, estimate wind speed parameters from its motion, and then use this to make synthetic objects move. We also extract water elevation parameters from the observed motion of boats and apply the resulting water waves to synthetic boats.
ICCV	Separating Reflection Components of Textured Surfaces using a Single Image.	Robby T. Tan,Katsushi Ikeuchi	2003	"The presence of highlights, which in dielectric inhomogeneousobjects are linear combination of specular and diffusereflection components, is inevitable. A number of methodshave been developed to separate these reflection components.To our knowledge, all methods that use a singleinput image require explicit color segmentation to deal withmulticolored surfaces. Unfortunately, for complex texturedimages, current color segmentation algorithms are stillproblematic to segment correctly. Consequently, a methodwithout explicit color segmentation becomes indispensable,and this paper presents such a method. The method is basedsolely on colors, particularly chromaticity, without requiringany geometrical parameter information. One of the basicideas is to compare the intensity logarithmic differentiationof specular-free images and input images iteratively.The specular-free image is a pseudo-code of diffuse componentsthat can be generated by shifting a pixel's intensityand chromaticity nonlinearly while retaining its hue. Allprocesses in the method are done locally, involving a maximumof only two pixels. The experimental results on naturalimages show that the proposed method is accurate and robustunder known scene illumination chromaticity. Unlikethe existing methods that use a single image, our methodis effective for textured objects with complex multicoloredscenes."
ICCV	Highlight Removal by Illumination-Constrained Inpainting.	Ping Tan,Stephen Lin,Long Quan,Heung-Yeung Shum	2003	We present a single-image highlight removal method that incorporates illumination-based constraints into image inpainting. Unlike occluded image regions filled by traditional inpainting, highlight pixels contain some useful information for guiding the inpainting process. Constraints provided by observed pixel colors, highlight color analysis and illumination color uniformity are employed in our method to improve estimation of the underlying diffuse color. The inclusion of these illumination constraints allows for better recovery of shading and textures by inpainting. Experimental results are given to demonstrate the performance of our method.
ICCV	Face Sketch Synthesis and Recognition.	Xiaoou Tang,Xiaogang Wang	2003	In this paper, we propose a novel face photo retrievalsystem using sketch drawings. By transforming a photoimage into a sketch, we reduce the difference betweenphoto and sketch significantly, thus allow effectivematching between the two. To improve the synthesisperformance, we separate shape and texture informationin a face photo, and conduct transformation on themrespectively. Finally a Bayesian classifier is used torecognize the probing sketch from the synthesizedpseudo-sketches. Experiments on a data set containing606 people clearly demonstrate the efficacy of thealgorithm.
ICCV	A New Perspective [on] Shape-from-Shading.	Ariel Tankus,Nir A. Sochen,Yehezkel Yeshurun	2003	Shape-from-Shading (SfS) is a fundamental problem inComputer Vision. The vast majority of research in this fieldhave assumed orthography as its projection model. Thispaper re-examines the basis of SfS, the image irradianceequation, under an assumption of perspective projection.The paper also shows that the perspective image irradianceequation depends merely on the natural logarithm of thedepth function (and not on the depth function itself), and assuch it is invariant to scale changes of the depth function.We then suggest a simple reconstruction algorithm basedon the perspective formula, and compare it to existing orthographicSfS algorithms. This simple algorithm obtainedlower error rates than legacy SfS algorithms, and equatedwith and sometimes surpassed state-of-the-art algorithms.These findings lend support to the assumption that transitionto a more realistic set of assumptions improves reconstructionsignificantly.
ICCV	Comparison of Graph Cuts with Belief Propagation for Stereo, using Identical MRF Parameters.	Marshall F. Tappen,William T. Freeman	2003	"Recent stereo algorithms have achieved impressive resultsby modelling the disparity image as a Markov RandomField (MRF). An important component of an MRF-basedapproach is the inference algorithm used to find the mostlikely setting of each node in the MRF. Algorithms havebeen proposed which use Graph Cuts or Belief Propagationfor inference. These stereo algorithms differ in both theinference algorithm used and the formulation of the MRF.It is unknown whether to attribute the responsibility for differencesin performance to the MRF or the inference algorithm.We address this through controlled experiments bycomparing the Belief Propagation algorithm and the GraphCuts algorithm on the same MRF's, which have been createdfor calculating stereo disparities. We find that the labellingsproduced by the two algorithms are comparable.The solutions produced by Graph Cuts have a lower energythan those produced with Belief Propagation, but this doesnot necessarily lead to increased performance relative tothe ground-truth."
ICCV	Surface Reconstruction from Feature Based Stereo.	Camillo J. Taylor	2003	This paper describes an approach to recovering surface models of complex scenes from the quasi-sparse data returned by a feature based stereo system. The method can be used to merge stereo results obtained from different viewpoints into a single coherent surface mesh. The technique proceeds by exploiting the free space theorem which provides a principled mechanism for reasoning about the structure of the scene based on quasi-sparse correspondences in multiple image. Effective methods for overcoming the difficulties posed by missing features and outliers are discussed. Results obtained by applying this approach to actual images are presented.
ICCV	Machine Learning and Multiscale Methods in the Identification of Bivalve Larvae.	Sanjay Tiwari,Scott Gallager	2003	This paper describes a novel application of support vector machines and multiscale texture and color invariants to a problem in biological oceanography: the identification of 6 species of bivalve larvae. Our data consists of polarized color images of scallop and other bivalve larvae (between 2 and 17 days old) collected from the ocean by a shipboard optical imaging system of our design. Larvae of scallops, clams, and oysters are small (100 microns) with few distinguishing features when observed under standard light microscopy. However, the use of polarized light with a full wave retardation plate produces a vivid color, bi-refringence pattern. The patterns display very subtle differences between species, often not discernable to human observers. We show that a soft-margin support vector machine with Gaussian RBF kernel is a good discriminator on a feature set extracted from Gabor wavelet transforms and color distribution angles of each image. By constraining the Gabor center frequencies to be low, the resulting system can attain classification accuracy in excess of 90% for vertically oriented images, and in excess of 80% for randomly oriented images.
ICCV	Entropy-of-likelihood Feature Selection for Image Correspondence.	Matthew Toews,Tal Arbel	2003	"Feature points for image correspondence are often selectedaccording to subjective criteria (e.g. edge density,nostrils). In this paper, we present a general, non-subjectivecriterion for selecting informative feature points, based onthe correspondence model itself. We describe the approachwithin the framework of the Bayesian Markov random field(MRF) model, where the degree of feature point informationis encoded by the entropy of the likelihood term. We proposethat feature selection according to minimum entropy-of-likelihood (EOL) is less likely to lead to correspondence ambiguity, thus improving the optimization process in termsof speed and quality of solution. Experimental resultsdemonstrate the criterion's ability to select optimal featurespoints in a wide variety of image contexts (e.g. objects,faces). Comparison with the automatic Kanade-Lucas-Tomasifeature selection criterion shows correspondence tobe significantly faster with feature points selected accordingto minimum EOL in difficult correspondence problems."
ICCV	3D Tracking = Classification + Interpolation.	Carlo Tomasi,Slav Petrov,Arvind Sastry	2003	Hand gestures are examples of fast and complex motions.Computers fail to track these in fast video, but sleight ofhand fools humans as well: what happens too quickly wejust cannot see. We show a 3D tracker for these types ofmotions that relies on the recognition of familiar configurationsin 2D images (classification), and fills the gapsin-between (interpolation). We illustrate this idea with experimentson hand motions similar to finger spelling. Thepenalty for a recognition failure is often small: if two configurationsare confused, they are often similar to eachother, and the illusion works well enough, for instance, todrive a graphics animation of the moving hand. We contributeadvances in both feature design and classifier training:our image features are invariant to image scale, translation,and rotation, and we propose a classification methodthat combines VQPCA with discrimination trees.
ICCV	Context-based vision system for place and object recognition.	Antonio Torralba,Kevin P. Murphy,William T. Freeman,Mark A. Rubin	2003	While navigating in an environment, a vision system has to be able to recognize where it is and what the main objects in the scene are. In this paper we present a context-based vision system for place and object recognition. The goal is to identify familiar locations (e.g., office 610, conference room 941, Main Street), to categorize new environments (office, corridor, street) and to use that information to provide contextual priors for object recognition (e.g., tables are more likely in an office than a street). We present a low-dimensional global image representation that provides relevant information for place recognition and categorization, and show how such contextual information introduces strong priors that simplify object recognition. We have trained the system to recognize over 60 locations (indoors and outdoors) and to suggest the presence and locations of more than 20 different object types. The algorithm has been integrated into a mobile system that provides real-time feedback to the user.
ICCV	Variational Frameworks for DT-MRI Estimation, Regularization and Visualization.	David Tschumperlé,Rachid Deriche	2003	We address three crucial issues encountered in DT-MRI (Diffusion Tensor Magnetic Resonance Imaging) : diffusion tensor Estimation, Regularization and fiber bundle Visualization. We first review related algorithms existing in the literature and propose then alternative variational formalisms that lead to new and improved schemes, thanks to the preservation of important tensor constraints (positivity, symmetry). We illustrate how our complete DT-MRI processing pipeline can be successfully used to construct and draw fiber bundles in the white matter of the brain, from a set of noisy raw MRI images.
ICCV	Image Parsing: Unifying Segmentation, Detection, and Recognition.	Zhuowen Tu,Xiangrong Chen,Alan L. Yuille,Song Chun Zhu	2003	We propose a general framework for parsing images into regions and objects. In this framework, the detection and recognition of objects proceed simultaneously with image segmentation in a competitive and cooperative manner. We illustrate our approach on natural images of complex city scenes where the objects of primary interest are faces and text. This method makes use of bottom-up proposals combined with top-down generative models using the Data Driven Markov Chain Monte Carlo (DDMCMC) algorithm which is guaranteed to converge to the optimal estimate asymptotically. More precisely, we define generative models for faces, text, and generic regions- e.g. shading, texture, and clutter. These models are activated by bottom-up proposals. The proposals for faces and text are learnt using a probabilistic version of AdaBoost. The DDMCMC combines reversible jump and diffusion dynamics to enable the generative models to explain the input images in a competitive and cooperative manner. Our experiments illustrate the advantages and importance of combining bottom-up and top-down models and of performing segmentation and object detection/recognition simultaneously.
ICCV	Plane-based Calibration Algorithm for Multi-camera Systems via Factorization of Homography Matrices.	Toshio Ueshiba,Fumiaki Tomita	2003	A new calibration algorithm for multi-camera systemsusing a planar reference pattern is proposed. The algorithmis an extension of Sturm-Maybank-Zhang style plane-basedcalibration technique for use with multiple cameras. Rigiddisplacements between the cameras are recovered as well asthe intrinsic parameters only by capturing with the camerasa model plane with known reference points placed at threeor more locations. Thus the algorithm yields a simple calibrationmeans for stereo vision systems with an arbitrarynumber of cameras while maintaining the handiness andflexibility of the original method. The algorithm is based onfactorization of homography matrices between the modeland image planes into the camera and plane parameters.To compensate for the indetermination of scaling factors,each homography matrix is rescaled by a double eigenvalueof a planar homology defined by two views and two modelplanes. The obtained parameters are finally refined by anon-linear maximum likelihood estimation (MLE) process.The validity of the proposed technique was verified throughsimulation and experiments with real data.
ICCV	Maintaining Multi-Modality through Mixture Tracking.	Jaco Vermaak,Arnaud Doucet,Patrick Pérez	2003	In recent years particle filters have become a tremendouslypopular tool to perform tracking for non-linearand/or non-Gaussian models. This is due to their simplicity,generality and success over a wide range of challengingapplications. Particle filters, and Monte Carlo methodsin general, are however poor at consistently maintainingthe multi-modality of the target distributions that may arisedue to ambiguity or the presence of multiple objects. Toaddress this shortcoming this paper proposes to model thetarget distribution as a non-parametric mixture model, andpresents the general tracking recursion in this case. It isshown how a Monte Carlo implementation of the generalrecursion leads to a mixture of particle filters that interactonly in the computation of the mixture weights, thus leadingto an efficient numerical algorithm, where all the resultspertaining to standard particle filters apply. The ability ofthe new method to maintain posterior multi-modality is illustratedon a synthetic example and a real world trackingproblem involving the tracking of football players in a videosequence.
ICCV	Object Recognition with Informative Features and Linear Classification.	Michel Vidal-Naquet,Shimon Ullman	2003	In this paper we show that efficient object recognition canbe obtained by combining informative features with linear classification. The results demonstrate the superiority of informative class-specific features, as compared with generic type features such as wavelets, for the task of object recognition. We show that information rich features can reach optimal performance with simple linear separation rules, while generic feature based classifiers require more complex classification schemes. This is significant because efficient and optimal methods have been developed for spaces that allow linear separation. To compare different strategies for feature extraction, we trained and compared classifiers working in feature spaces of the same low dimensionality, using two feature types (image fragments vs. wavelets) and two classification rules (linear hyperplane and a Bayesian Network). The results show that by maximizing the individual information of the features, it is possible to obtain efficient classification by a simple linear separating rule, aswell as more efficient learning.
ICCV	Detecting Pedestrians Using Patterns of Motion and Appearance.	Paul A. Viola,Michael J. Jones,Daniel Snow	2003	This paper describes a pedestrian detection system that integratesimage intensity information with motion information.We use a detection style algorithm that scans a detectorover two consecutive frames of a video sequence. Thedetector is trained (using AdaBoost) to take advantage ofboth motion and appearance information to detect a walkingperson. Past approaches have built detectors based onmotion information or detectors based on appearance information,but ours is the first to combine both sources ofinformation in a single detector. The implementation describedruns at about 4 frames/second, detects pedestriansat very small scales (as small as 20x15 pixels), and has avery low false positive rate.Our approach builds on the detection work of Viola andJones. Novel contributions of this paper include: i) developmentof a representation of image motion which is extremelyefficient, and ii) implementation of a state of theart pedestrian detection system which operates on low resolutionimages under difficult conditions (such as rain andsnow).
ICCV	Recognition with Local Features: the Kernel Recipe.	Christian Wallraven,Barbara Caputo,Arnulf B. A. Graf	2003	Recent developments in computer vision have shown that local features can provide efficient representations suitable for robust object recognition. Support Vector Machines have been established as powerful learning algorithms with good generalization capabilities. In this paper, we combine these two approaches and propose a general kernel method for recognition with local features. We show that the proposed kernel satisfies the Mercer condition and that it is suitable for many established local feature frameworks. Large-scale recognition results are presented on three different databases, which demonstrate that SVMs with the proposed kernel perform better than standard matching techniques on local features. In addition, experiments on noisy and occluded images show that local feature representations significantly outperform global approaches.
ICCV	Facial Expression Decomposition.	Hongcheng Wang,Narendra Ahuja	2003	In this paper, we propose a novel approach for facialexpression decomposition - Higher-Order Singular ValueDecomposition (HOSVD), a natural generalization ofmatrix SVD. We learn the expression subspace and personsubspace from a corpus of images showing seven basicfacial expressions, rather than resort to expert-coded facialexpression parameters as in [3]. We propose a simultaneousface and facial expression recognition algorithm,which can classify the given image into one of the sevenbasic facial expression categories, and then other facialexpressions of the new person can be synthesized using thelearned expression subspace model. The contributions ofthis work lie mainly in two aspects. First, we propose a newHOSVD based approach to model the mapping betweenpersons and expressions, used for facial expression synthesisfor a new person. Second, we realize simultaneous faceand facial expression recognition as a result of facialexpression decomposition. Experimental results are presentedthat illustrate the capability of the person subspaceand expression subspace in both synthesis and recognitiontasks. As a quantitative measure of the quality of synthesis,we propose using Gradient Minimum Square Error (GMSE)which measures the gradient difference between the originaland synthesized images.
ICCV	A Novel Approach For Texture Shape Recovery.	Jing Wang,Kristin J. Dana	2003	In vision and graphics, there is a sustained interest incapturing accurate 3D shape with various scanning devices.However, the resulting geometric representation isonly part of the story. Surface texture of real objects isalso an important component of the representation and fine-scalesurface geometry such as surface markings, roughness,and imprints, are essential in highly realistic renderingand accurate prediction. We present a novel approachfor measuring the fine-scale surface shape of specular surfacesusing a curved mirror to view multiple angles in asingle image. A distinguishing aspect of our method is thatit is designed for specular surfaces, unlike many methods(e.g. laser scanning) which cannot handle highly specularobjects. Also, the spatial resolution is very high so that itcan resolve very small surface details that are beyond theresolution of standard devices. Furthermore, our approachincorporates the simultaneous use of a bidirectional texturemeasurement method, so that spatially varying bidirectionalreflectance is measured at the same time as surfaceshape.
ICCV	Landmark-based Shape Deformation with Topology-Preserving Constraints.	Song Wang,Jim Xiuquan Ji,Zhi-Pei Liang	2003	This paper presents a novel approach for landmark-basedshape deformation, in which fitting error and shapedifference are formulated into a support vector machine(SVM) regression problem. To well describe nonrigid shapedeformation, this paper measures the shape difference usinga thin-plate spline model. The proposed approach iscapable of preserving the topology of the template shape inthe deformation. This property is achieved by inserting aset of additional points and imposing a set of linear equalityand/or inequality constraints. The underlying optimizationproblem is solved using a quadratic programming algorithm.The proposed method has been tested using practicaldata in the context of shape-based image segmentation.Some relevant practical issues, such as missing detectedlandmarks and selection of the regularization parameter arealso briefly discussed.
ICCV	Fusion of Static and Dynamic Body Biometrics for Gait Recognition.	Liang Wang,Huazhong Ning,Tieniu Tan,Weiming Hu	2003	Human identification at a distance has recently gainedgrowing interest from computer vision researchers. Thispaper aims to propose a visual recognition algorithmbased upon fusion of static and dynamic body biometrics.For each sequence involving a walking figure, posechanges of the segmented moving silhouettes arerepresented as an associated sequence of complex vectorconfigurations, and are then analyzed using the Procrustesshape analysis method to obtain a compact appearancerepresentation, called static information of body. Also, amodel-based approach is presented under a Condensationframework to track the walker and to recover joint-angletrajectories of lower limbs, called dynamic information ofgait. Both static and dynamic cues are respectively used forrecognition using the nearest exemplar classifier. They arealso effectively fused on decision level using differentcombination rules to improve the performance of bothidentification and verification. Experimental results on adataset including 20 subjects demonstrate the validity ofthe proposed algorithm.
ICCV	Variable Bandwidth QMDPE and Its Application in Robust Optical Flow Estimation.	Hanzi Wang,David Suter	2003	Robust estimators, such as Least Median of Squared (LMedS) Residuals, M-estimators, the Least Trimmed Squares (LTS) etc., have been employed to estimate optical flow from image sequences in recent years. However, these robust estimators have a breakdown point of no more than 50%. In this paper, we propose a novel robust estimator, called variable bandwidth Quick Maximum Density Power Estimator (vbQMDPE),which can tolerate more than 50% outliers. We apply the novel proposed estimator to robust optical flow estimation. Our method yields better results than most other recently proposed methods, and it has the potential to better handle multiple motion effects.
ICCV	Eye Gaze Estimation from a Single Image of One Eye.	Jian-Gang Wang,Eric Sung,Ronda Venkateswarlu	2003	"In this paper, we present a novel approach, called the""one-circle"" algorithm, for measuring the eye gaze using a monocular image that zooms in on only one eye of a person. Observing that the iris contour is a circle, we estimate the normal direction of this iris circle, considered as the eye gaze, from its elliptical image. From basic projective geometry, an ellipse can be back-projected into space onto two circles of different orientations. However, by using an anthropometric property of the eyeball, the correct solution can be disambiguated. This allows us to obtain a higher resolution image of the iris with a zoom-in camera and thereby achieving higher accuracies in the estimation. The robustness of our gaze determination approach was verified statistically by the extensive experiments on synthetic and real image data. The two key contributions in this paper are that we show the possibility of finding the unique eye gaze direction from a single image of one eye and that one can obtain better accuracy as a consequence of this."
ICCV	Unified Subspace Analysis for Face Recognition.	Xiaogang Wang,Xiaoou Tang	2003	We propose a face difference model that decomposesface difference into three components, intrinsicdifference, transformation difference, and noise. Usingthe face difference model and a detailed subspace analysison the three components we develop a unified frameworkfor subspace analysis. Using this framework we discoverthe inherent relationship among different subspacemethods and their unique contributions to the extractionof discriminating information from the face difference.This eventually leads to the construction of a 3Dparameter space that uses three subspace dimensions asaxis. Within this parameter space, we develop a unifiedsubspace analysis method that achieves better recognitionperformance than the standard subspace methods on over2000 face images from the FERET database.
ICCV	Joint Region Tracking with Switching Hypothesized Measurements.	Yang Wang,Tele Tan,Kia-Fock Loe	2003	This paper proposes a switching hypothesized measurements (SHM) model supporting multimodal probability distributions and presents the application ofthe model in handling potential variability in visual environments when tracking multiple objects jointly. For a set of occlusion hypotheses, a frame is measured once under each hypothesis, resulting in a set of measurements at each time instant. A computationally efficient SHM filter is derived for online joint region tracking. Both occlusion relationships and states of the objects are recursively estimated from the history of hypothesized measurements. The reference image is updated adaptively to deal with appearance changes of the objects. The SHM model is generally applicable to various dynamic processes with multiple alternative measurement methods.
ICCV	Cumulative Residual Entropy, A New Measure of Information & its Application to Image Alignment.	Fei Wang,Baba C. Vemuri,Murali Rao,Yunmei Chen	2003	"In this paper we use the cumulative distribution of a random variable to define the information content in it and use it to develop a novel measure of information that parallels Shannon entropy, which we dub cumulative residual entropy (CRE). The key features of CRE may be summarized as, (1)its definition is valid in both the continuous and discrete domains,(2) it is mathematically more general than the Shannon entropy and (3) its computation from sample data is easy and these computations converge asymptotically to the true values. We define the cross-CRE (CCRE) between two random variables and apply it to solve the uni- & multi-modal image alignment problem for parameterized (rigid, affine and projective) transformations. The key strengths of the CCRE over using the now popular mutual information method (based on Shannon's entropy) are that the former has significantly larger noise immunity and a much larger convergence range over the field of parameterized transformations. These strengths of CCRE are demonstrated via experiments on synthesized and real image data."
ICCV	Modeling Textured Motion : Particle, Wave and Sketch.	Yizhou Wang,Song Chun Zhu	2003	In this paper, we present a generative model for textured motion phenomena, such as falling snow, wavy river and dancing grass, etc. Firstly, we represent an image as a linear superposition of image bases selected from a generic and over-complete dictionary. The dictionary contains Gabor bases for point/particle elements and Fourier bases for wave-elements. These bases compete to explain the input images. The transform from a raw image to a base or a token representation leads to large dimension reduction. Secondly, we introduce a unified motion equation to characterize the motion of these bases and the interactions between waves and particles, e.g. a ball floating on water. We use statistical learning algorithm to identify the structure of moving objects and their trajectories automatically. Then novel sequences can be synthesized easily from the motion and image models. Thirdly, we replace the dictionary of Gabor and Fourier bases with symbolic sketches (also bases). With the same image and motion model, we can render realistic and stylish cartoon animation. In our view, cartoon and sketch are symbolic visualization of the inner representation for visual perception. The success of the cartoon animation, in turn, suggests that our image and motion models capture the essence of visual perception of textured motion.
ICCV	Color Edge Detection by Photometric Quasi-Invariants.	Joost van de Weijer,Theo Gevers,Jan-Mark Geusebroek	2003	Photometric invariance is used in many computer vision applications.The advantage of photometric invariance is therobustness against shadows, shading, and illumination conditions.However, the drawbacks of photometric invarianceis the loss of discriminative power and the inherent instabilitiescaused by the non-linear transformations to computethe invariants.In this paper, we propose a new class of derivativeswhich we refer to as photometric quasi-invariants. Thesequasi-invariants share with full invariants the nice propertythat they are robust against photometric edges, such asshadows or specular edges. Further, these quasi-invariantsdo not have the inherent instabilities of full photometric invariants.We will apply these quasi-invariant derivativesin the context of photometric invariant edge detection andclassification. Experiments show that the quasi-invariantderivatives are stable and they significantly outperform thefull invariant derivatives in discriminative power.
ICCV	Capturing Subtle Facial Motions in 3D Face Tracking.	Zhen Wen,Thomas S. Huang	2003	"Facial motions produce not only facial feature points motions,but also subtle appearance changes such as wrinklesand shading changes. These subtle changes are importantyet difficult issues for both analysis (tracking) and synthesis(animation). Previous approaches were mostly basedon models learned from extensive training appearance examples.However, the space of all possible facial motionappearance is huge. Thus, it is not feasible to collect samplescovering all possible variations due to lighting conditions,individualities, and head poses. Therefore, it is difficultto adapt such models to new conditions. In this paper,we present an adaptive technique for analyzing subtle facialappearance changes. We propose a new ratio-imagebased appearance feature, which is independent of a person'sface albedo. This feature is used to track face appearancevariations based on exemplars. To adapt the exemplarappearance model to new people and lighting conditions,we develop an online EM-based algorithm. Experimentsshow that the proposed method improves classification resultsin a facial expression recognition task, where a varietyof people and lighting conditions are involved."
ICCV	Combinatorial Constraints on Multiple Projections of a Set of Points.	Tomás Werner	2003	Multiple projections of a scene cannot be arbitrary, the allowedconfigurations being given by matching constraints.This paper presents new matching constraints on multipleprojections of a rigid point set by uncalibrated cameras, obtainedby formulation in the oriented projective rather thanprojective geometry. They follow from consistency of orientationsof camera rays and from the fact that the scene is theaffine rather that projective space. For their non-parametricnature, we call them combinatorial. The constraints are derivedin a unified theoretical framework using the theory oforiented matroids. For example, we present constraints on4 point correspondences for 2D camera resectioning, on 3correspondences in two 1D cameras, and on 4 correspondencesin two 2D cameras.
ICCV	Scene Modeling Based on Constraint System Decomposition Techniques.	Marta Wilczkowiak,Gilles Trombettoni,Christophe Jermann,Peter F. Sturm,Edmond Boyer	2003	We present a new approach to 3D scene modeling basedon geometric constraints. Contrary to the existing methods,we can quickly obtain 3D scene models that respectthe given constraints exactly. Our system can describe alarge variety of linear and non-linear constraints in a flexibleway.To deal with the constraints, we decided to exploit theproperties of the GPDOF algorithm developed in the ConstraintProgramming community [12]. The approach isbased on a dictionary of so-called r-methods, based on theoremsof geometry, which can solve a subset of geometricconstraints in a very efficient way. GPDOF is used to find,in polynomial-time, a reduced parameterization of a scene,and to decompose the equation system, induced by constraints,into a sequence of r-methods. We have validatedour approach in reconstructing, from images, 3D modelsof buildings based on linear and quadratic geometric constraints.
ICCV	A Sparse Probabilistic Learning Algorithm for Real-Time Tracking.	Oliver M. C. Williams,Andrew Blake,Roberto Cipolla	2003	"This paper addresses the problem of applying powerful pattern recognition algorithms based on kernels to efficient visual tracking. Recently Avidan [1] has shown that object recognizers using kernel-SVMs can be elegantly adapted to localization by means of spatial perturbation of the SVM, using optic flow. Whereas Avidan's SVM applies to each frame of a video independently of other frames, the benefits of temporal fusion of data are well known. This issue is addressed here by using a fully probabilistic 'Relevance Vector Machine' (RVM) to generate observations with Gaussian distributions that can be fused over time. To improve performance further, rather than adapting a recognizer, webuild a localizer directly using the regression form of the RVM. A classification SVM is used in tandem, for object verification, and this provides the capability of automatic initialization and recovery. The approach is demonstrated in real-time face and vehicle tracking systems. The 'sparsity' of the RVMs means that only a fraction of CPU time is required to track at frame rate. Tracker output is demonstrated in a camera management task in which zoom and pan are controlled inresponse to speaker/vehicle position and orientation, over an extended period. The advantages of temporal fusion inthis system are demonstrated."
ICCV	Feature Selection for Unsupervised and Supervised Inference: the Emergence of Sparsity in a Weighted-based Approach.	Lior Wolf,Amnon Shashua	2003	"The problem of selecting a subset of relevant features in a potentially overwhelming quantity of data is classic and found in many branches of science including - examples in computer vision, text processing and more recently bio-informatics are abundant. In this work we present a definition of '""relevancy"" based on spectral properties of the Affinity (or Laplacian) of the features' measurement matrix. The feature selection process is then based on a continuous ranking of the features defined by a least-squares optimization process. A remarkable property of the feature relevance function is that sparse solutions for the ranking values naturally emerge as a result of a ""biased non-negativity"" of a key matrix in the process. As a result, a simple least-squares optimization process converges onto a sparse solution, i.e., a selection of a subset of features which form a local maxima over the relevance function. The feature selection algorithm can be embedded in both unsupervised and supervised inference problems and empirical evidence show that the feature selections typically achieve high accuracy even when only a small fraction of the features are relevant."
ICCV	Tracking Articulated Body by Dynamic Markov Network.	Ying Wu,Gang Hua,Ting Yu	2003	A new method for visual tracking of articulated objectsis presented. Analyzing articulated motion is challengingbecause the dimensionality increase potentially demandstremendous increase of computation. To ease this problem,we propose an approach that analyzes subparts locallywhile reinforcing the structural constraints at the meantime. The computational model of the proposed approachis based on a dynamic Markov network, a generative modelwhich characterizes the dynamics and the image observationsof each individual subpart as well as the motion constraintsamong different subparts. Probabilistic variationalanalysis of the model reveals a mean field approximationto the posterior densities of each subparts given visual evidence,and provides a computationally efficient way forsuch a difficult Bayesian inference problem. In addition,we design mean field Monte Carlo (MFMC) algorithms, inwhich a set of low dimensional particle filters interact witheach other and solve the high dimensional problem collaboratively.Extensive experiments on tracking human bodyparts demonstrate the effectiveness, significance and computationalefficiency of the proposed method.
ICCV	Two-Frame Wide Baseline Matching.	Jiangjian Xiao,Mubarak Shah	2003	This paper describes a novel approach to automatically recover corresponding feature points and epipolar geometry over two wide baseline frames. Our contributions consist of several aspects: First, the use of an affine invariant feature, edge-corner, is introduced to provide a robust and consistent matching primitives. Second, based on SVD decomposition of affine matrix, the affine matching space between two corners can be approximately divided into two independent spaces by rotation angle and scaling factor. Employing this property, a two-stage affine matching algorithm is designed to obtain robust matches over two frames. Third, using the epipolar geometry estimated by these matches, more corresponding feature points are determined. Based on these robust correspondences, the fundamental matrix is refined, and a series of virtual views of the scene are synthesized. Finally, several experiments are presented to illustrate that a number of robust correspondences can be stably determined for two wide baseline images under significant camera motions with illumination changes, occlusions, and self-similarities. After testing a number of examples and comparing with the existing methods, the experimental results strongly demonstrate that our matching method outperforms the state-of-art algorithms for all of the test cases.
ICCV	Boosting Chain Learning for Object Detection.	Rong Xiao,Long Zhu,HongJiang Zhang	2003	"A general classification framework, called boostingchain, is proposed for learning boosting cascade. In thisframework, a ""chain"" structure is introduced to integratehistorical knowledge into successive boosting learning.Moreover, a linear optimization scheme is proposed toaddress the problems of redundancy in boosting learningand threshold adjusting in cascade coupling. By thismeans, the resulting classifier consists of fewer weakclassifiers yet achieves lower error rates than boostingcascade in both training and test. Experimentalcomparisons of boosting chain and boosting cascade areprovided through a face detection problem. Thepromising results clearly demonstrate the effectivenessmade by boosting chain."
ICCV	Tracking Articulated Hand Motion with Eigen Dynamics Analysis.	Hanning Zhou,Thomas S. Huang	2003	This paper introduces the concept of eigen-dynamics andproposes an eigen dynamics analysis (EDA) method to learnthe dynamics of natural hand motion from labelled sets ofmotion captured with a data glove. The result is parameterizedwith a high-order stochastic linear dynamic system(LDS) consisting of five lower-order LDS. Each correspondingto one eigen-dynamics. Based on the EDA model, weconstruct a dynamic Bayesian network (DBN) to analyzethe generative process of a image sequence of natural handmotion. Using the DBN, a hand tracking system is implemented.Experiments on both synthesized and real-worlddata demonstrate the robustness and effectiveness of thesetechniques.
ICCV	A Background Layer Model for Object Tracking Through Occlusion.	Yue Zhou,Hai Tao	2003	Motion layer estimation has recently emerged as apromising object tracking method. In this paper, we extendprevious research on layer-based tracker by introducingthe concept of background occluding layers and explicitlyinferring depth ordering of foreground layers. Thebackground occluding layers lie in front of, behind, and inbetween foreground layers. Each pixel in the backgroundregions belongs to one of these layers and occludes all theforeground layers behind it. Together with the foregroundordering, the complete information necessary for reliablytracking objects through occlusion is included in ourrepresentation. An MAP estimation framework isdeveloped to simultaneously update the motion layerparameters, the ordering parameters, and the backgroundoccluding layers. Experimental results show that undervarious conditions with occlusion, including situationswith moving objects undergoing complex motions orhaving complex interactions, our tracking algorithm isable to handle many difficult tracking tasks reliably.
ICCV	Ranking Prior Likelihood Distributions for Bayesian Shape Localization Framework.	Shuicheng Yan,Mingjing Li,HongJiang Zhang,QianSheng Cheng	2003	In this paper, we formulate the shape localization problem in the Bayesian framework. In the learning stage, we propose the Constrained Rank Boost approach to model the likelihood of local features associated with the keypoints of an object, like face, while preserve the prior ranking order between the ground truth position of a keypoint and its neighbors; in the inferring stage, a simple efficient iterative algorithm is proposed to uncover the MAP shape by locally modeling the likelihood distribution around each key point via our proposed variational Locally Weighted Learning (VLWL) method. Our proposed framework has the following benefits: 1) compared to the classical PCA models, the likelihood presented by the ranking prior likelihood model has more discriminating power as to the optimal position and its neighbors, especially in the problem with ambiguity between the optimal positions and their neighbors; 2) the VLWL method guarantees that the posterior probability of the derived shape increases monotonously; and 3) the above two methods are both based on accurate probability formulation, which spontaneously leads to a robust confidence measure for the discovered shape. Moreover, we present a theoretical analysis for the convergence of the Constrained Rank-Boost. Extensive experiments compared with the Active Shape Models demonstrate the accuracy, robustness, and stability of our proposed framework.
ICCV	Automatically Labeling Video Data Using Multi-class Active Learning.	Rong Yan,Jie Yang,Alexander G. Hauptmann	2003	Labeling video data is an essential prerequisite for many vision applications that depend on training data, such as visual information retrieval, object recognition, and human activity modeling. However, manually creating labels is not only time-consuming but also subject to human errors, and eventually, becomes impossible for a very large amount of data (e.g. 24/7 surveillance video). To minimize the human effort in labeling, we propose a unified multi-class active learning approach for automatically labeling video data. The contributions of this paper include extending active learning from binary classes to multiple classes and evaluating several practical sample selection strategies. The experimental results show that the proposed approach works effectively even with a significantly reduced amount of labeled data. The best sample selection strategy can achieve more than a 50% error reduction over random sample selection.
ICCV	Binocular Helmholtz Stereopsis.	Todd Zickler,Jeffrey Ho,David J. Kriegman,Jean Ponce,Peter N. Belhumeur	2003	Helmholtz stereopsis has been introduced recently as a surfacereconstruction technique that does not assume a modelof surface reflectance. In the reported formulation, correspondencewas established using a rank constraint, necessitatingat least three viewpoints and three pairs of images.Here, it is revealed that the fundamental Helmholtz stereopsisconstraint defines a nonlinear partial differential equation,which can be solved using only two images. It is shownthat, unlike conventional stereo, binocular Helmholtz stereopsisis able to establish correspondence (and thereby recoversurface depth) for objects having an arbitrary andunknown BRDF and in textureless regions (i.e., regions ofconstant or slowly varying BRDF). An implementation andexperimental results validate the method for specular surfaceswith and without texture.
ICCV	Improved Fast Gauss Transform and Efficient Kernel Density Estimation.	Changjiang Yang,Ramani Duraiswami,Nail A. Gumerov,Larry S. Davis	2003	Evaluating sums of multivariate Gaussians is a common computational task in computer vision and pattern recognition, including in the general and powerful kernel density estimation technique. The quadratic computational complexity of the summation is a significant barrier to the scalability of this algorithm to practical applications. The fast Gauss transform (FGT) has successfully accelerated the kernel density estimation to linear running time for low-dimensional problems. Unfortunately, the cost of a direct extension of the FGT to higher-dimensional problems grows exponentially with dimension, making it impractical for dimensions above 3. We develop an improved fast Gauss transform to efficiently estimate sums of Gaussians in higher dimensions, where a new multivariate expansion scheme and an adaptive space subdivision technique dramatically improve the performance. The improved FGT has been applied to the mean shift algorithm achieving linear computational complexity. Experimental results demonstrate the efficiency and effectiveness of our algorithm.
ICCV	Counting People in Crowds with a Real-Time Network of Simple Image Sensors.	Danny B. Yang,Héctor H. González-Baños,Leonidas J. Guibas	2003	"Estimating the number of people in a crowded environment is a central task in civilian surveillance. Most vision-based counting techniques depend on detecting individuals in order to count, an unrealistic proposition in crowded settings. We propose an alternative approach that directly estimates the number of people. In our system, groups of image sensors segment foreground objects from the background, aggregate the resulting silhouettes over a network, and compute a planar projection of the scene's visual hull. We introduce a geometric algorithm that calculates bounds on the number of persons in each region of the projection, after phantom regions have been eliminated. The computational requirements scale well with the number of sensors and the number of people, and only limited amounts of data are transmitted over the network. Because of these properties, our system runs in real-time and can be deployed as an untethered wireless sensor network. We describe the major components of our system, and report preliminary experiments with our first prototype implementation."
ICCV	Dealing with Textureless Regions and Specular Highlights - A Progressive Space Carving Scheme Using a Novel Photo-consistency Measure.	Ruigang Yang,Marc Pollefeys,Greg Welch	2003	We present two extensions to the Space Carving framework. The first is a progressive scheme to better reconstruct surfaces lacking sufficient textures. The second is a novel photo-consistency measure that is valid for both specular and diffuse surfaces, under unknown lighting conditions.
ICCV	Geometric Segmentation of Perspective Images Based on Symmetry Groups.	Allen Y. Yang,Shankar Rao,Kun Huang,Wei Hong,Yi Ma	2003	Symmetry is an effective geometric cue to facilitate conventionalsegmentation techniques on images of man-madeenvironment. Based on three fundamental principles thatsummarize the relations between symmetry and perspectiveimaging, namely, structure from symmetry, symmetry hypothesistesting, and global symmetry testing, we developa prototype system which is able to automatically segmentsymmetric objects in space from single 2-D perspective images.The result of such a segmentation is a hierarchy ofgeometric primitives, called symmetry cells and complexes,whose 3-D structure and pose are fully recovered. Such ageometrically meaningful segmentation may greatly facilitateapplications such as feature matching and robot navigation.
ICCV	Assessing Accuracy Factors in Deformable 2D/3D Medical Image Registration Using a Statistical Pelvis Model.	Jianhua Yao,Russell H. Taylor	2003	Deformable 2D-3D medical image registration is anessential technique in Computer Integrated Surgery (CIS)to fuse 3D pre-operative data with 2D intra-operativedata. Several factors may affect the accuracy of 2D-3Dregistration, including the number of 2D views, the anglebetween views, the view angle relative to anatomical objects,the co-registration error between views, the imagenoise, and the image distortion. In this paper, we investigateand assess the relationship between these factors andthe accuracy of 2D-3D registration. We proposed a deformable2D-3D registration method based on a statisticalmodel. We conducted experiments using a hemi-pelvismodel and simulated x-ray images. Some discussions areprovided on how to improve the accuracy of 2D-3D registrationbased on our assessment.
ICCV	Reinforcement Learning for Combining Relevance Feedback Techniques.	Peng-Yeng Yin,Bir Bhanu,Kuang-Cheng Chang,Anlei Dong	2003	"Relevance feedback (RF) is an interactive process which refines the retrievals by utilizing user's feedback history. Most researchers strive to develop new RF techniques and ignore the advantages of existing ones. In this paper, we propose an image relevance reinforcement learning (IRRL) model for integrating existing RF techniques. Various integration schemes are presented and a long-term shared memory is used to exploit the retrieval experience from multiple users. Also, a concept digesting method is proposed to reduce the complexity of storage demand. The experimental results manifest that the integration of multiple RF approaches gives better retrieval performance than using one RF technique alone, and that the sharing of relevance knowledge between multiple query sessions also provides significant contributions for improvement. Further, the storage demand is significantly reduced by the concept digesting technique. This shows the scalability of the proposed model against a growing-size database."
ICCV	Catadioptric Camera Calibration Using Geometric Invariants.	Xianghua Ying,Zhanyi Hu	2003	Central catadioptric cameras are imaging devices thatuse mirrors to enhance the field of view while preservinga single effective viewpoint. In this paper, we propose anovel method for the calibration of central catadioptriccameras using geometric invariants. Lines in space areprojected into conics in the catadioptric image plane aswell as spheres in space. We proved that the projection ofa line can provide three invariants whereas the projectionof a sphere can provide two. From these invariants,constraint equations for the intrinsic parameters ofcatadioptric camera are derived. Therefore, there are twovariants of this novel method. The first one uses theprojections of lines and the second one uses theprojections of spheres. In general, the projections of twolines or three spheres are sufficient to achieve thecatadioptric camera calibration. One importantobservation in this paper is that the method based on theprojections of spheres is more robust and has higheraccuracy than that using the projections of lines. Theperformances of our method are demonstrated by theresults of simulations and experiments with real images.
ICCV	Multiclass Spectral Clustering.	Stella X. Yu,Jianbo Shi	2003	We propose a principled account on multiclass spectral clustering. Given a discrete clustering formulation, we first solve a relaxed continuous optimization problem by eigen-decomposition. We clarify the role of eigenvectors as a generator of all optimal solutions through orthonormal transforms. We then solve an optimal discretization problem, which seeks a discrete solution closest to the continuous optima. The discretization is efficiently computed in an iterative fashion using singular value decomposition and non-maximum suppression. The resulting discrete solutions are nearly global-optimal. Our method is robust to random initialization and converges faster than other clustering methods. Experiments on real image segmentation are reported.
ICCV	Shape and Motion under Varying Illumination: Unifying Structure from Motion, Photometric Stereo, and Multi-view Stereo.	Li Zhang,Brian Curless,Aaron Hertzmann,Steven M. Seitz	2003	This paper presents an algorithm for computing optical flow,shape, motion, lighting, and albedo from an image sequenceof a rigidly-moving Lambertian object under distant illumination.The problem is formulated in a manner that subsumesstructure from motion, multi-view stereo, and photometricstereo as special cases. The algorithm utilizes bothspatial and temporal intensity variation as cues: the formerconstrains flow and the latter constrains surface orientation;combining both cues enables dense reconstruction ofboth textured and texture-less surfaces. The algorithm worksby iteratively estimating affine camera parameters, illumination,shape, and albedo in an alternating fashion. Results aredemonstrated on videos of hand-held objects moving in frontof a fixed light and camera.
ICCV	Tracking Objects Using Density Matching and Shape Priors.	Tao Zhang,Daniel Freedman	2003	We present a novel method for tracking objects by combiningdensity matching with shape priors. Density matchingis a tracking method which operates by maximizing theBhattacharyya similarity measure between the photometricdistribution from an estimated image region and a modelphotometric distribution. Such trackers can be expressed asPDE-based curve evolutions, which can be implemented usinglevel sets. Shape priors can be combined with this level-setimplementation of density matching by representing theshape priors as a series of level sets; a variational approachallows for a natural, parametrization-independentshape term to be derived. Experimental results on real imagesequences are shown.
ICCV	Facial Expression Understanding in Image Sequences Using Dynamic and Active Visual Information Fusion.	Yongmian Zhang,Qiang Ji	2003	This paper explores the use of multisensory information fusiontechnique with Dynamic Bayesian networks (DBNs)for modeling and understanding the temporal behaviors offacial expressions in image sequences. Our approach tothe facial expression understanding lies in a probabilisticframework by integrating the DBNs with the facial actionunits (AUs) from psychological view. The DBNs provide acoherent and unified hierarchical probabilistic frameworkto represent spatial and temporal information related to facialexpressions, and to actively select the most informativevisual cues from the available information to minimize theambiguity in recognition. The recognition of facial expressionsis accomplished by fusing not only from the currentvisual observations, but also from the previous visual evidences.Consequently, the recognition becomes more robustand accurate through modeling the temporal behavior of facialexpressions. Experimental results demonstrate that ourapproach is more admissible for facial expression analysisin image sequences.
ICCV	Segmenting Foreground Objects from a Dynamic Textured Background via a Robust Kalman Filter.	Jing Zhong,Stan Sclaroff	2003	The algorithm presented in this paper aims to segment the foreground objects in video (e.g., people) given time-varying, textured backgrounds. Examples of time-varying backgrounds include waves on water, clouds moving, trees waving in the wind, automobile traffic, moving crowds, escalators, etc. We have developed a novel foreground-background segmentation algorithm that explicitly accounts for the non-stationary nature and clutter-like appearance of many dynamic textures. The dynamic texture is modeled byan Autoregressive Moving Average Model (ARMA). A robust Kalman filter algorithm iteratively estimates the intrinsic appearance of the dynamic texture, as well as the regions of the foreground objects. Preliminary experiments with this method have demonstrated promising results.
ICCV	Conditional Feature Sensitivity: A Unifying View on Active Recognition and Feature Selection.	Xiang Sean Zhou,Dorin Comaniciu,Arun Krishnan	2003	"The objective of active recognition is to iteratively collectthe next ""best"" measurements (e.g., camera angles orviewpoints), to maximally reduce ambiguities in recognition.However, existing work largely overlooked featureinteraction issues. Feature selection, on the other hand,focuses on the selection of a subset of measurements for agiven classification task, but is not context sensitive (i.e.,the decision does not depend on the current input). Thispaper proposes a unified perspective through conditionalfeature sensitivity analysis, taking into account both currentcontext and feature interactions. Based on differentrepresentations of the contextual uncertainties, we presentthree treatment models and exploit their joint power fordealing with complex feature interactions. Synthetic examplesare used to systematically test the validity of theproposed models. A practical application in medical domainis illustrated using an echocardiography databasewith more than 2000 video segments with both subjective(from experts) and objective validations."
VAST	Portals: Aiding Navigation in Virtual Museums.	Ioannis Kotziampasis,Nathan Sidwell,Alan Chalmers	2003	Portals: Aiding Navigation in Virtual Museums.
VAST	On the Interactive 3D Reconstruction of Iberian Vessels.	Francisco J. Melero,Juan Carlos Torres,Alejandro León	2003	On the Interactive 3D Reconstruction of Iberian Vessels.
VAST	High Fidelity Lighting of Knossos.	Ioannis Roussos,Alan Chalmers	2003	High Fidelity Lighting of Knossos.
VAST	Digital Reunification of the Parthenon and its Sculptures.	Jessi Stumpfel,Chris Tchou,Nathan Yun,Philippe Martinez,Tim Hawkins,Andrew Jones,Brian Emerson,Paul E. Debevec	2003	Digital Reunification of the Parthenon and its Sculptures.
VAST	Computer Aided Analysis of Underdrawings in Infrared Reflectograms.	Paul Kammerer,Ernestine Zolda,Robert Sablatnig	2003	Computer Aided Analysis of Underdrawings in Infrared Reflectograms.
VAST	Design and Application of an Augmented Reality System for continuous, context-sensitive guided tours of indoor and outdoor cultural sites and museums.	Vassilios Vlahakis,T. Pliakas,Athanasios M. Demiris,Nikolaos Ioannidis	2003	Design and Application of an Augmented Reality System for continuous, context-sensitive guided tours of indoor and outdoor cultural sites and museums.
VAST	Presenting an Archaeological Site in the Virtual Showcase.	Florian Ledermann,Dieter Schmalstieg	2003	Presenting an Archaeological Site in the Virtual Showcase.
VAST	Scholarly documentation through digital storytelling: opening the tomb of Ramsses II .	Kevin Cain	2003	Scholarly documentation through digital storytelling: opening the tomb of Ramsses II .
VAST	"The Dissemination & Promotion of Cultural Heritage Sites to People 'on the move' Employing Digital TV."	Emmanuel Tsekleves,John Cosmas	2003	"The Dissemination & Promotion of Cultural Heritage Sites to People 'on the move' Employing Digital TV."
VAST	Web-based visualization of virtual archaeological sites.	Markus Grabner,Ralph Wozelka,Manish Mirchandani,Konrad Schindler	2003	Web-based visualization of virtual archaeological sites.
VAST	"VCLab's Tools for 3D range data processing."	Marco Callieri,Paolo Cignoni,Fabio Ganovelli,Claudio Montani,Paolo Pingi,Roberto Scopigno	2003	"VCLab's Tools for 3D range data processing."
VAST	Providing Multimedia Tools for Recording, Reconstruction, Visualisation and Database Storage/Access of Archaeological Excavations.	John Cosmas,T. Itegaki,Damian Green,N. Joseph,Luc J. Van Gool,Alexey Zalesny,Desi Vanrintel,Franz Leberl,Markus Grabner,Konrad Schindler,Konrad F. Karner,Michael Gervautz,Stefan Hynst,Marc Waelkens,Maarten Vergauwen,Marc Pollefeys,Kurt Cornelis,Tijl Vereenooghe,Robert Sablatnig,Martin Kampel,P. Axell,E. Meyns	2003	Providing Multimedia Tools for Recording, Reconstruction, Visualisation and Database Storage/Access of Archaeological Excavations.
VAST	Genuine Archaeology and Real Computer Graphics .	Duncan Brown	2003	Genuine Archaeology and Real Computer Graphics .
VAST	VAST 2003: The 4th International Symposium on Virtual Reality, Archaeology and Intelligent Cultural Heritage, Brighton, United Kingdom, 2003	David B. Arnold,Alan Chalmers,Franco Niccolucci	2003	VAST 2003: The 4th International Symposium on Virtual Reality, Archaeology and Intelligent Cultural Heritage, Brighton, United Kingdom, 2003
VAST	Enhancing Virtual Reality Walkthroughs of Archaeological Sites.	Georgios Papaioannou,Athanasios Gaitatzes,Dimitrios Christopoulos	2003	Enhancing Virtual Reality Walkthroughs of Archaeological Sites.
VAST	Managing the real with the virtual: A role for digital media recording in archaeological fieldwork.	Michael Ashley López	2003	Managing the real with the virtual: A role for digital media recording in archaeological fieldwork.
VAST	Animating the Dance Archives.	Lars Wilke,Tom Calvert,Rhonda Ryman,Ilene Fox	2003	Animating the Dance Archives.
VAST	Multi-Platform Skeletal Visualisation and Reproduction in Stereolithography.	B. Magnanti,A. Nicholls,R. Sayle	2003	Multi-Platform Skeletal Visualisation and Reproduction in Stereolithography.
VAST	Modelling the Walled City of Nicosia.	M. Dikaiakou,A. Efthymiou,Yiorgos Chrysanthou	2003	Modelling the Walled City of Nicosia.
VAST	New Approaches to Efficient Rendering of Complex Reconstructed Environments.	Sven Havemann,Dieter W. Fellner,A. M. Day,David B. Arnold	2003	New Approaches to Efficient Rendering of Complex Reconstructed Environments.
VAST	Interactive virtual mummy dissections in a historical anatomic theatre.	Stefan Seipel,M. Lindkvist,Lars Winkler Pettersson	2003	Interactive virtual mummy dissections in a historical anatomic theatre.
VAST	Monticello Through the Window.	Nathaniel Williams,Chad Hantak,Kok-Lim Low,John Thomas,Kurtis Keller,Lars S. Nyland,David P. Luebke,Anselmo Lastra	2003	Monticello Through the Window.
VAST	Photorealism and Non-Photorealism in Virtual Heritage Representation.	Maria Roussou,George Drettakis	2003	Photorealism and Non-Photorealism in Virtual Heritage Representation.
VAST	Visualisation of Highly Textured Surfaces.	Sabry F. El-Hakim,Lorenzo Gonzo,Michel Picard,Stefano Girardi,Andrea Simoni,Eric Paquet,Herna L. Viktor,Claus Brenner	2003	Visualisation of Highly Textured Surfaces.
VAST	Virtual Vessel Reconstruction from a Fragment?s Profile.	Martin Kampel,Francisco J. Melero	2003	Virtual Vessel Reconstruction from a Fragment?s Profile.
IEEE Visualization	Hybrid Segmentation and Exploration of the Human Lungs.	Dirk Bartz,Dirk Mayer,Jan Fischer,Sebastian Ley,Ángel del Río,Steffi Thust,Claus Peter Heussel,Hans-Ulrich Kauczor,Wolfgang Straßer	2003	Segmentation of the tracheo-bronchial tree of the lungs is notoriously difficult. This is due to the fact that the small size of some of the anatomical structures are subject to partial volume effects. Furthermore, the limited intensity contrast between the participating materials (air, blood, and tissue) increases the segmentation difficulties. In this paper, we propose a hybrid segmentation method which is based on a pipeline of three segmentation stages to extract the lower airways down to the seventh generation of the bronchi. User interaction is limited to the specification of a seed point inside the easily detectable trachea at the upper end of the lower airways. Similarly, the complementary vascular tree of the lungs can be segmented. Furthermore, we modified our virtual endoscopy system to visualize the vascular and airway system of the lungs along with other features, such as lung tumors.
IEEE Visualization	Visualization of Steep Breaking Waves and Thin Spray Sheets around a Ship.	Paul Adams,Douglas Dommermuth	2003	The simulation of breaking of waves, the formation of thin spray sheets, and the entrainment of air around the next generation of naval surface combatants is an ongoing 3-year Department of Defense (DoD) Challenge Project. The goal of this project is a validated computation capability to model the full hydrodynamics around a surface combatant including all of the processes that affect mission and performance. Visualization of these large-scale simulations is paramount to understanding the complex physics involved. These simulations produce enormous data sets with both surface and volumetric qualities. Wave breaking, spray sheets, and air entrainment can be visualized using isosurfaces of scalar data. Visualization of quantities such as the vorticity field also provides insight into the dynamics of droplet and bubble formation. This paper documents the techniques used, results obtained, and lessons learned from the visualization of the hydrodynamics of naval vessels.
IEEE Visualization	Conveying Shape and Features with Image-Based Relighting.	David Akers,Frank Losasso,Jeff Klingner,Maneesh Agrawala,John Rick,Pat Hanrahan	2003	Hand-crafted illustrations are often more effective than photographs for conveying the shape and important features of an object, but they require expertise and time to produce. We describe an image compositing system and user interface that allow an artist to quickly and easily create technical illustrations from a set of photographs of an object taken from the same point of view under variable lighting conditions. Our system uses a novel compositing process in which images are combined using spatially-varying light mattes, enabling the final lighting in each area of the composite to be manipulated independently. We describe an interface that provides for the painting of local lighting effects (e.g. shadows, highlights, and tangential lighting to reveal texture) directly onto the composite. We survey some of the techniques used in illustration and lighting design to convey the shape and features of objects and describe how our system can be used to apply these techniques.
IEEE Visualization	Psychophysical Scaling of a Cardiovascular Information Display.	Robert Albert,Noah Syroid,Yinqi Zhang,James Agutter,Frank Drews,David L. Strayer,George Hutchinson,Dwayne R. Westenskow	2003	"A new method was developed to increase the saliency of changing variables in a cardiovascular visualization for use by anesthesiologists in the operating room (OR). Clinically meaningful changes in patient physiology were identified and then mapped to the inherent psychophysical properties of the visualization. A long history of psychophysical research has provided an understanding of the parameters within which the human information processing system is able to detect changes in the size, shape and color of visual objects [Gescheider 1976; Spence 1990; Baird 1970]. These detection thresholds are known as just noticeable differences (JNDs) which characterize the amount of change in an objectýs attribute that is recognizable 50% of the time. A prototype version of the display has been demonstrated to facilitate anesthesiologist's performance while reducing cognitive workload during simulated cardiac events [Agutter et al. 2002]. In order to further improve the utility of the new cardiovascular visualization, the clinically relevant changes in cardiovascular variables are mapped to noticeable perceptual changes in the representational elements of the display. The results of the method described in this paper are used to merge information from the psychophysical properties of the cardiovascular visualization, with clinically relevant changes in the patient's cardiovascular physiology as measured by the clinical meaningfulness questionnaire. The result of this combination will create a visualization that is sensitive to changes in the cardiovascular health of the patient and communicates this information to the user in a meaningful, salient and intuitive manner."
IEEE Visualization	Space Efficient Fast Isosurface Extraction for Large Datasets.	Udeepta Bordoloi,Han-Wei Shen	2003	In this paper, we present a space efficient algorithm for speeding up isosurface extraction. Even though there exist algorithms that can achieve optimal search performance to identify isosurface cells, they prove impractical for large datasets due to a high storage overhead. With the dual goals of achieving fast isosurface extraction and simultaneously reducing the space requirement, we introduce an algorithm based on transform coding to compress the interval information of the cells in a dataset. Compression is achieved by first transforming the cell intervals (minima, maxima) into a form which allows more efficient compaction. It is followed by a dataset optimized non-uniform quantization stage. The compressed data is stored in a data structure that allows fast searches in the compression domain, thus eliminating the need to retrieve the original representation of intervals at run-time. The space requirement of our search data structure is the mandatory cost of storing every cell id once, plus an overhead for quantization information. The overhead is typically in the order of a few hundredths of the dataset size.
IEEE Visualization	A Multi-Resolution Data Structure for 2-Dimensional Morse Functions.	Peer-Timo Bremer,Herbert Edelsbrunner,Bernd Hamann,Valerio Pascucci	2003	A Multi-Resolution Data Structure for 2-Dimensional Morse Functions.
IEEE Visualization	Counting Cases in Marching Cubes: Toward a Generic Algorithm for Producing Substitopes.	David C. Banks,Stephen A. Linton	2003	"We describe how to count the cases that arise in a family of visualization techniques, including Marching Cubes, Sweeping Simplices, Contour Meshing, Interval Volumes, and Separating Surfaces. Counting the cases is the first step toward developing a generic visualization algorithm to produce substitopes (geometric substitutions of polytopes). We demonstrate the method using a software system (""GAP"") for computational group theory. The case-counts are organized into a table that provides a taxonomy of members of the family; numbers in the table are derived from actual lists of cases, which are computed by our methods. The calculation confirms previously reported case-counts for large dimensions that are too large to check by hand, and predicts the number of cases that will arise in algorithms that have not yet been invented."
IEEE Visualization	Fairing scalar fields by variational modeling of contours.	Martin Bertram	2003	Volume rendering and isosurface extraction from three-dimensional scalar fields are mostly based on piecewise trilinear representations. In regions of high geometric complexity such visualization methods often exhibit artefacts, due to trilinear interpolation. In this work, we present an iterative fairing method for scalar fields interpolating function values associated with grid points while smoothing the contours inside the grid cells based on variational principles. We present a local fairing method providing a piecewise bicubic representation of two-dimensional scalar fields. Our algorithm generalizes to the trivariate case and can be used to increase the resolution of data sets either locally or globally, reducing interpolation artefacts. In contrast to filtering methods, our algorithm does not reduce geometric detail supported by the data.
IEEE Visualization	Interoperability of Visualization Software and Data Models is NOT an Achievable Goal.	E. Wes Bethel,Greg Abram,John Shalf,Randy Frank,James P. Ahrens,Steve Parker,Nagiza F. Samatova,Mark Miller	2003	Interoperability of Visualization Software and Data Models is NOT an Achievable Goal.
IEEE Visualization	Out-of-Core Isosurface Extraction of Time-Varying Irregular Grids.	Yi-Jen Chiang	2003	Out-of-Core Isosurface Extraction of Time-Varying Irregular Grids.
IEEE Visualization	Planet-Sized Batched Dynamic Adaptive Meshes (P-BDAM).	Paolo Cignoni,Fabio Ganovelli,Enrico Gobbetti,Fabio Marton,Federico Ponchio,Roberto Scopigno	2003	We describe an efficient technique for out-of-core management and interactive rendering of planet sized textured terrain surfaces. The technique, called P-Batched Dynamic Adaptive Meshes (P-BDAM), extends the BDAM approach by using as basic primitive a general triangulation of points on a displaced triangle. The proposed framework introduces several advances with respect to the state of the art: thanks to a batched host-to-graphics communication model, we outperform current adaptive tessellation solutions in terms of rendering speed; we guarantee overall geometric continuity, exploiting programmable graphics hardware to cope with the accuracy issues introduced by single precision floating points; we exploit a compressed out of core representation and speculative prefetching for hiding disk latency during rendering of out-of-core data; we efficiently construct high quality simplified representations with a novel distributed out of core simplification algorithm working on a standard PC network.
IEEE Visualization	Hierarchical Clustering for Unstructured Volumetric Scalar Fields.	Christopher S. Co,Bjørn Heckel,Hans Hagen,Bernd Hamann,Kenneth I. Joy	2003	"We present a method to represent unstructured scalar fields at multiple levels of detail. Using a parallelizable classification algorithm to build a cluster hierarchy, we generate a multiresolution representation of a given volumetric scalar data set. The method uses principal component analysis (PCA) for cluster generation and a fitting technique based on radial basis functions (RBFs). Once the cluster hierarchy has been generated, we utilize a variety of techniques for extracting different levels of detail. The main strength of this work is its generality. Regardless of grid type, this method can be applied to any discrete scalar field representation, even one given as a ""point cloud."""
IEEE Visualization	Accelerating Large Data Analysis by Exploiting Regularities.	David Ellsworth,Patrick J. Moran	2003	We present techniques for discovering and exploiting regularity in large curvilinear data sets. The data can be based on a single mesh or a mesh composed of multiple submeshes (also known as zones). Multi-zone data are typical in Computational Fluid Dynamics (CFD) simulations. Regularities include axis-aligned rectilinear and cylindrical meshes as well as cases where one zone is equivalent to a rigid-body transformation of another. Our algorithms can also discover rigid-body motion of meshes in time-series data. Next, we describe a data model where we can utilize the results from the discovery process in order to accelerate large data visualizations. Where possible, we replace general curvilinear zones with rectilinear or cylindrical zones. In rigid-body motion cases we replace a time-series of meshes with a transformed mesh object where a reference mesh is dynamically transformed based on a given time value in order to satisfy geometry requests, on demand. The data model enables us to make these substitutions and dynamic transformations transparently with respect to the visualization algorithms. We present results with large data sets where we combine our mesh replacement and transformation techniques with out-of-core paging in order to achieve analysis speedups ranging from 1.5 to 2.
IEEE Visualization	Monte Carlo Volume Rendering.	Balázs Csébfalvi,László Szirmay-Kalos	2003	Monte Carlo Volume Rendering.
IEEE Visualization	Video Visualization.	Gareth Daniel,Min Chen	2003	"Video data, generated by the entertainment industry, security and traffic cameras, video conferencing systems, video emails, and so on, is perhaps most time-consuming to process by human beings. In this paper, we present a novel methodology for ""summarizing"" video sequences using volume visualization techniques. We outline a system pipeline for capturing videos, extracting features, volume rendering video and feature data, and creating video visualization. We discuss a collection of image comparison metrics, including the linear dependence detector, for constructing ""relative"" and ""absolute"" difference volumes that represent the magnitude of variation between video frames. We describe the use of a few volume visualization techniques, including volume scene graphs and spatial transfer functions, for creating video visualization. In particular, we present a stream-based technique for processing and directly rendering video data in real time. With the aid of several examples, we demonstrate the effectiveness of using video visualization to convey meaningful information contained in video sequences."
IEEE Visualization	Clifford Convolution And Pattern Matching On Vector Fields.	Julia Ebling,Gerik Scheuermann	2003	The goal of this paper is to define a convolution operation which transfers image processing and pattern matching to vector fields from flow visualization. For this, a multiplication of vectors is necessary. Clifford algebra provides such a multiplication of vectors. We define a Clifford convolution on vector fields with uniform grids. The Clifford convolution works with multivector filter masks. Scalar and vector masks can be easily converted to multivector fields. So, filter masks from image processing on scalar fields can be applied as well as vector and multivector masks. Furthermore, a method for pattern matching with Clifford convolution on vector fields is described. The method is independent of the direction of the structures. This provides an automatic approach to feature detection. The features can be visualized using any known method like glyphs, isosurfaces or streamlines. The features are defined by filter masks instead of analytical properties and thus the approach is more intuitive.
IEEE Visualization	Visibility Culling Using Plenoptic Opacity Functions for Large Data Visualization.	Jinzhu Gao,Jian Huang,Han-Wei Shen,James Arthur Kohl	2003	Visibility Culling Using Plenoptic Opacity Functions for Large Data Visualization.
IEEE Visualization	Visualization, Optimization, and Business Strategy: A Case Study.	Donna L. Gresh,Eugene I. Kelton	2003	"We describe a visualization application intended for operational use in formulating business strategy in the customer service arena. The visualization capability provided in this application implicitly allows the user to better formulate the objective function for large optimization runs which act to minimize costs based on certain input parameters. Visualization is necessary because many of the inputs to the optimization runs are themselves strategic business decisions which are not pre-ordained. Both information visualization presentations and three-dimensional visualizations are included to help users better understand the cost/benefit tradeoffs of these strategic business decisions. Here, visualization explicitly provides value not possible algorithmically, as the perceived benefit of different combinations of service level does not have an a priori mathematical formulation. Thus we take advantage of the fundamental power of visualization, bringing the user's intuition and pattern recognition skills into the solution, while simultaneously taking advantage of the strength of algorithmic approaches to quickly and accurately find an optimal solution to a well-defined problem."
IEEE Visualization	Which Comes First, Utility or Usability?	Georges G. Grinstein,Alfred Kobsa,Catherine Plaisant,Ben Shneiderman,John T. Stasko	2003	Which Comes First, Utility or Usability?
IEEE Visualization	High-Quality Two-Level Volume Rendering of Segmented Data Sets on Consumer Graphics Hardware.	Markus Hadwiger,Christoph Berger,Helwig Hauser	2003	One of the most important goals in volume rendering is to be able to visually separate and selectively enable specific objects of interest contained in a single volumetric data set, which can be approached by using explicit segmentation information. We show how segmented data sets can be rendered interactively on current consumer graphics hardware with high image quality and pixel-resolution filtering of object boundaries. In order to enhance object perception, we employ different levels of object distinction. First, each object can be assigned an individual transfer function, multiple of which can be applied in a single rendering pass. Second, different rendering modes such as direct volume rendering, iso-surfacing, and non-photorealistic techniques can be selected for each object. A minimal number of rendering passes is achieved by processing sets of objects that share the same rendering mode in a single pass. Third, local compositing modes such as alpha blending and MIP can be selected for each object in addition to a single global mode, thus enabling high-quality two-level volume rendering on GPUs.
IEEE Visualization	LightKit: A lighting system for effective visualization.	Michael Halle,Jeanette C. Meng	2003	"LightKit is a system for lighting three-dimensional synthetic scenes. LightKit simplifies the task of producing visually pleasing, easily interpretable images for visualization while making it harder to produce results where the scene illumination distracts from the visualization process. LightKit is based on lighting designs developed by artists and photographers and shown in previous studies to enhance shape perception. A key light provides natural overhead illumination of the scene, augmented by fill, head, and back lights. By default, lights are attached to a normalized, subject-centric, camera-relative coordinate frame to ensure consistent lighting independent of camera location or orientation. This system allows all lights to be positioned by specifying just six parameters. The intensity of each light is specified as a ratio to the key light intensity, allowing the scene's brightness to be adjusted using a single parameter. The color of each light is specified by a single normalized color parameter called warmth that is based on color temperature of natural sources. LightKit's default values for light position, intensity, and color are chosen to produce good results for a variety of scenes. LightKit is designed to work with both hardware graphics systems and, potentially, higher quality off-line rendering systems. We provide examples of images created using a LightKit implementation within the VTK visualization toolkit software framework."
IEEE Visualization	Visualizing Industrial CT Volume Data for Nondestructive Testing Applications.	Runzhen Huang,Kwan-Liu Ma,Patrick S. McCormick,William Ward	2003	This paper describes a set of techniques developed for the visualization of high-resolution volume data generated from industrial computed tomography for nondestructive testing (NDT)applications. Because the data are typically noisy and contain fine features, direct volume rendering methods do not always give us satisfactory results. We have coupled region growing techniques and a 2D histogram interface to facilitate volumetric feature extraction. The new interface allows the user to conveniently identify, separate or composite, and compare features in the data. To lower the cost of segmentation, we show how partial region growing results can suggest a reasonably good classification function for the rendering of the whole volume. The NDT applications that we work on demand visualization tasks including not only feature extraction and visual inspection, but also modeling and measurement of concealed structures in volumetric objects. An efficient filtering and modeling process for generating surface representation of extracted features is also introduced. Four CT data sets for preliminary NDT are used to demonstrate the effectiveness of the new visualization strategy that we have developed.
IEEE Visualization	Adaptive Design of a Global Opacity Transfer Function for Direct Volume Rendering of Ultrasound Data.	Dieter Hönigmann,Johannes Ruisz,Christoph Haider	2003	While there are a couple of transfer function design approaches for CT and MRI data, direct volume rendering of ultrasound data still relies on manual adjustment of an inflexible piecewise linear opacity transfer function (OTF) on a trial-and-error basis. The main challenge of automatically designing an OTF for visualization of sonographic data is the low signal-to-noise ratio in combination with real time data acquisition at frame rates up to 25 volumes per second. In this paper, we present an efficient solution of this task. Our approach is based on the evaluation of tube cores, i.e., collections of voxels gathered by traversing the volume in rendering direction. We use information about the probable position of an interface between tissue of different echogenicity to adaptively design an OTF in a multiplicative way. We show the appropriateness of our approach by examples, deliberately on data sets of moderate quality arising frequently in clinical settings.
IEEE Visualization	Hierarchical Splatting of Scattered Data.	Matthias Hopf,Thomas Ertl	2003	Numerical particle simulations and astronomical observations create huge data sets containing uncorrelated 3D points of varying size. These data sets cannot be visualized interactively by simply rendering millions of colored points for each frame. Therefore, in many visualization applications a scalar density corresponding to the point distribution is resampled on a regular grid for direct volume rendering. However, many fine details are usually lost for voxel resolutions which still allow interactive visualization on standard workstations. Since no surface geometry is associated with our data sets, the recently introduced point-based rendering algorithms cannot be applied as well. In this paper we propose to accelerate the visualization of scattered point data by a hierarchical data structure based on a PCA clustering procedure. By traversing this structure for each frame we can trade-off rendering speed vs. image quality. Our scheme also reduces memory consumption by using quantized relative coordinates and it allows for fast sorting of semi-transparent clusters. We analyze various software and hardware implementations of our renderer and demonstrate that we can now visualize data sets with tens of millions of points interactively with sub-pixel screen space error on current PC graphics hardware employing advanced vertex shader functionality.
IEEE Visualization	A Constraint-Based Technique for Haptic Volume Exploration.	Milan Ikits,J. Dean Brederson,Charles D. Hansen,Christopher R. Johnson	2003	We present a haptic rendering technique that uses directional constraints to facilitate enhanced exploration modes for volumetric datasets. The algorithm restricts user motion in certain directions by incrementally moving a proxy point along the axes of a local reference frame. Reaction forces are generated by a spring coupler between the proxy and the data probe, which can be tuned to the capabilities of the haptic interface. Secondary haptic effects including field forces, friction, and texture can be easily incorporated to convey information about additional characteristics of the data. We illustrate the technique with two examples: displaying fiber orientation in heart muscle layers and exploring diffusion tensor fiber tracts in brain white matter tissue. Initial evaluation of the approach indicates that haptic constraints provide an intuitive means for displaying directional information in volume data.
IEEE Visualization	Large Mesh Simplification using Processing Sequences.	Martin Isenburg,Peter Lindstrom,Stefan Gumhold,Jack Snoeyink	2003	In this paper we show how out-of-core mesh processing techniques can be adapted to perform their computations based on the new processing sequence paradigm [Isenburg and Gumhold 2003; Isenburg et al. 2003], using mesh simplification as an example. We believe that this processing concept will also prove useful for other tasks, such as parameterization, remeshing, or smoothing, for which currently only in-core solutions exist. A processing sequence represents a mesh as a particular inter-leaved ordering of indexed triangles and vertices. This representation allows streaming very large meshes through main memory while maintaining information about the visitation status of edges and vertices. At any time, only a small portion of the mesh is kept in-core, with the bulk of the mesh data residing on disk. Mesh access is restricted to a fixed traversal order, but full connectivity and geometry information is available for the active elements of the traversal. This provides seamless and highly efficient out-of-core access to very large meshes for algorithms that can adapt their computations to this fixed ordering. The two abstractions that are naturally supported by this representation are boundary-based and buffer-based processing. We illustrate both abstractions by adapting two different simplification methods to perform their computation using a prototype of our mesh processing sequence API. Both algorithms benefit from using processing sequences in terms of improved quality, more efficient execution, and smaller memory footprints.
IEEE Visualization	The Visualization Market: Open Source vs. Commercial Approaches.	Jeremy Jaech,Stephen C. North,Mike Peery,Will Schroeder,Jim Thomas	2003	The Visualization Market: Open Source vs. Commercial Approaches.
IEEE Visualization	Appearance-Preserving View-Dependent Visualization.	Justin Jang,William Ribarsky,Christopher D. Shaw,Peter Wonka	2003	In this paper a new quadric-based view-dependent simplification scheme is presented. The scheme provides a method to connect mesh simplification controlled by a quadric error metric with a level-of-detail hierarchy that is accessed continuously and efficiently based on current view parameters. A variety of methods for determining the screen-space metric for the view calculation are implemented and evaluated, including an appearance-preserving method that has both geometry- and texture-preserving aspects. Results are presented and compared for a variety of models.
IEEE Visualization	Volume Tracking Using Higher Dimensional Isocontouring.	Guangfeng Ji,Han-Wei Shen,Rephael Wenger	2003	Tracking and visualizing local features from a time-varying volumetric data allows the user to focus on selected regions of interest, both in space and time, which can lead to a better understanding of the underlying dynamics. In this paper, we present an efficient algorithm to track time-varying isosurfaces and interval volumes using isosurfacing in higher dimensions. Instead of extracting the data features such as isosurfaces or interval volumes separately from multiple time steps and computing the spatial correspondence between those features, our algorithm extracts the correspondence directly from the higher dimensional geometry and thus can more efficiently follow the user selected local features in time. In addition, by analyzing the resulting higher dimensional geometry, it becomes easier to detect important topological events and the corresponding critical time steps for the selected features. With our algorithm, the user can interact with the underlying time-varying data more easily. The computation cost for performing time-varying volume tracking is also minimized.
IEEE Visualization	Visualizing Spatial and Temporal Variability in Coastal Observatories.	Walter Jiménez,Wagner Toledo Corrêa,Cláudio T. Silva,António M. Baptista	2003	In this paper, we describe a set of 3D and 4D visualization tools and techniques for CORIE, a complex environmental observation and forecasting system (EOFS) for the Columbia River. The Columbia River, a complex and highly variable estuary, is the target of numerous cross-disciplinary ecosystem research projects and is at the heart of multiple sustainable development issues with long reaching implications for the Pacific Northwest. However, there has been until recently no comprehensive and objective system available for modeling this environment, and as a consequence, researchers and agencies have had inadequate tools for evaluating the effects of natural resource management decisions. CORIE was designed to address this gap and is a major step towards the vision of a scalable, multi-use, real-time EOFS. Although CORIE already had a rich set of visualization tools, most of them produced 2D visualizations and did not allow for interactive visualization. Our work adds advanced interactive 3D tools to CORIE, which can be used for further inspection of the simulated and measured data.
IEEE Visualization	Interactive Protein Manipulation.	Oliver Kreylos,Nelson L. Max,Bernd Hamann,Silvia N. Crivelli,E. Wes Bethel	2003	"We describe an interactive visualization and modeling program for the creation of protein structures ""from scratch."" The input to our program is an amino acid sequence - decoded from a gene - and a sequence of predicted secondary structure types for each amino acid - provided by external structure prediction programs. Our program can be used in the set-up phase of a protein structure prediction process; the structures created with it serve as input for a subsequent global internal energy minimization, or another method of protein structure prediction. Our program supports basic visualization methods for protein structures, interactive manipulation based on inverse kinematics, and visualization guides to aid a user in creating ""good"" initial structures."
IEEE Visualization	Advanced Curved Planar Reformation: Flattening of Vascular Structures.	Armin Kanitsar,Rainer Wegenkittl,Dominik Fleischmann,Eduard Gröller	2003	Traditional volume visualization techniques may provide incomplete clinical information needed for applications in medical visualization. In the area of vascular visualization important features such as the lumen of a diseased vessel segment may not be visible. Curved Planar Reformation (CPR) has proven to be an acceptable practical solution. Existing CPR techniques, however, still have diagnostically relevant limitations. In this paper we introduce two advanced methods for efficient vessel visualization, based on the concept of CPR. Both methods benefit from relaxation of spatial coherence in favor of improved feature perception. We present a new technique to visualize the interior of a vessel in a single image. A vessel is re-sampled along a spiral around its central axis. The helical spiral depicts the vessel volume. Furthermore, a method to display an entire vascular tree without mutually occluding vessels is presented. Minimal rotations at the bifurcations avoid occlusions. For each viewing direction the entire vessel structure is visible.
IEEE Visualization	Acceleration Techniques for GPU-based Volume Rendering.	Jens Krüger,Rüdiger Westermann	2003	Nowadays, direct volume rendering via 3D textures has positioned itself as an efficient tool for the display and visual analysis of volumetric scalar fields. It is commonly accepted, that for reasonably sized data sets appropriate quality at interactive rates can be achieved by means of this technique. However, despite these benefits one important issue has received little attention throughout the ongoing discussion of texture based volume rendering: the integration of acceleration techniques to reduce per-fragment operations. In this paper, we address the integration of early ray termination and empty-space skipping into texture based volume rendering on graphical processing units (GPU). Therefore, we describe volume ray-casting on programmable graphics hardware as an alternative to object-order approaches. We exploit the early z-test to terminate fragment processing once sufficient opacity has been accumulated, and to skip empty space along the rays of sight. We demonstrate performance gains up to a factor of 3 for typical renditions of volumetric data sets on the ATI 9700 graphics card.
IEEE Visualization	Image Space Based Visualization of Unsteady Flow on Surfaces.	Robert S. Laramee,Bruno Jobard,Helwig Hauser	2003	We present a novel technique for direct visualization of unsteady flow on surfaces from computational fluid dynamics. The method generates dense representations of time-dependent vector fields with high spatio-temporal correlation using both Lagrangian-Eulerian Advection and Image Based Flow Visualization as its foundation. While the 3D vector fields are associated with arbitrary triangular surface meshes, the generation and advection of texture properties is confined to image space. Frame rates of up to 20 frames per second are realized by exploiting graphics card hardware. We apply this algorithm to unsteady flow on boundary surfaces of, large, complex meshes from computational fluid dynamics composed of more than 250,000 polygons, dynamic meshes with time-dependent geometry and topology, as well as medical data.
IEEE Visualization	Curvature-Based Transfer Functions for Direct Volume Rendering: Methods and Applications.	Gordon L. Kindlmann,Ross T. Whitaker,Tolga Tasdizen,Torsten Möller	2003	Direct volume rendering of scalar fields uses a transfer function to map locally measured data properties to opacities and colors. The domain of the transfer function is typically the one-dimensional space of scalar data values. This paper advances the use of curvature information in multi-dimensional transfer functions, with a methodology for computing high-quality curvature measurements. The proposed methodology combines an implicit formulation of curvature with convolution-based reconstruction of the field. We give concrete guidelines for implementing the methodology, and illustrate the importance of choosing accurate filters for computing derivatives with convolution. Curvature-based transfer functions are shown to extend the expressivity and utility of volume rendering through contributions in three different application areas: non-photorealistic volume rendering, surface smoothing via anisotropic diffusion, and visualization of isosurface uncertainty.
IEEE Visualization	Gaussian Transfer Functions for Multi-Field Volume Visualization.	Joe Kniss,Simon Premoze,Milan Ikits,Aaron E. Lefohn,Charles D. Hansen,Emil Praun	2003	Volume rendering is a flexible technique for visualizing dense 3D volumetric datasets. A central element of volume rendering is the conversion between data values and observable quantities such as color and opacity. This process is usually realized through the use of transfer functions that are precomputed and stored in lookup tables. For multidimensional transfer functions applied to multivariate data, these lookup tables become prohibitively large. We propose the direct evaluation of a particular type of transfer functions based on a sum of Gaussians. Because of their simple form (in terms of number of parameters), these functions and their analytic integrals along line segments can be evaluated efficiently on current graphics hardware, obviating the need for precomputed lookup tables. We have adopted these transfer functions because they are well suited for classification based on a unique combination of multiple data values that localize features in the transfer function domain. We apply this technique to the visualization of several multivariate datasets (CT, cryosection) that are difficult to classify and render accurately at interactive rates using traditional approaches.
IEEE Visualization	Interactive Deformation and Visualization of Level Set Surfaces Using Graphics Hardware.	Aaron E. Lefohn,Joe Michael Kniss,Charles D. Hansen,Ross T. Whitaker	2003	Deformable isosurfaces, implemented with level-set methods, have demonstrated a great potential in visualization for applications such as segmentation, surface processing, and surface reconstruction. Their usefulness has been limited, however, by their high computational cost and and reliance on significant parameter tuning. This paper presents a solution to these challenges by describing graphics processor (GPU) based algorithms for solving and visualizing level-set solutions at interactive rates. Our efficient GPU-based solution relies on packing the level-set isosurface data into a dynamic, sparse texture format. As the level set moves, this sparse data structure is updated via a novel GPU to CPU message passing scheme. When the level-set solver is integrated with a real-time volume renderer operating on the same packed format, a user can visualize and steer the deformable level-set surface as it evolves. In addition, the resulting isosurface can serve as a region-of-interest specifier for the volume renderer. This paper demonstrates the capabilities of this technology for interactive volume visualization and segmentation.
IEEE Visualization	Interactive 3D Visualization Of Rigid Body Systems.	Zoltan Konyha,Kresimir Matkovic,Helwig Hauser	2003	Simulation of rigid body dynamics has been an field of active research for quite some time. However, the presentation of simulation results has received far less attention so far. We present an interactive and intuitive 3D visualization framework for rigid body simulation data. We introduce various glyphs representing vector attributes such as force and velocity as well as angular attributes including angular velocity and torque. We have integrated our visualization method into an application developed at one of the leading companies in automotive engine design and simulation. We apply our principles to visualization of chain and belt driven timing drives in engines.
IEEE Visualization	Chameleon: An interactive texture-based rendering framework for visualizing three-dimensional vector fields.	Guo-Shi Li,Udeepta Bordoloi,Han-Wei Shen	2003	In this paper we present an interactive texture-based technique for visualizing three-dimensional vector fields. The goal of the algorithm is to provide a general volume rendering framework allowing the user to compute three-dimensional flow textures interactively, and to modify the appearance of the visualization on the fly. To achieve our goal, we decouple the visualization pipeline into two disjoint stages. First, streamlines are generated from the 3D vector data. Various geometric properties of the streamlines are extracted and converted into a volumetric form using a hardware-assisted slice sweeping algorithm. In the second phase of the algorithm, the attributes stored in the volume are used as texture coordinates to look up an appearance texture to generate both informative and aesthetic representations of the underlying vector field. Users can change the input textures and instantaneously visualize the rendering results. With our algorithm, visualizations with enhanced structural perception using various visual cues can be rendered in real time. A myriad of existing geometry-based and texture-based visualization techniques can also be emulated.
IEEE Visualization	Empty Space Skipping and Occlusion Clipping for Texture-based Volume Rendering.	Wei Li,Klaus Mueller,Arie E. Kaufman	2003	We propose methods to accelerate texture-based volume rendering by skipping invisible voxels. We partition the volume into sub-volumes, each containing voxels with similar properties. Sub-volumes composed of only voxels mapped to empty by the transfer function are skipped. To render the adaptively partitioned sub-volumes in visibility order, we reorganize them into an orthogonal BSP tree. We also present an algorithm that computes incrementally the intersection of the volume with the slicing planes, which avoids the overhead of the intersection and texture coordinates computation introduced by the partitioning. Rendering with empty space skipping is 2 to 5 times faster than without it. To skip occluded voxels, we introduce the concept of orthogonal opacity map, that simplifies the transformation between the volume coordinates and the opacity map coordinates, which is intensively used for occlusion detection. The map is updated efficiently by the GPU. The sub-volumes are then culled and clipped against the opacity map. We also present a method that adaptively adjusts the optimal number of the opacity map updates. With occlusion clipping, about 60% of non-empty voxels can be skipped and an additional 80% speedup on average is gained for iso-surface-like rendering.
IEEE Visualization	Do I Really See a Bone?	Raghu Machiraju,Chris Johnson,Terry S. Yoo,Roger Crawfis,David S. Ebert,Don Stredney	2003	Do I Really See a Bone?
IEEE Visualization	MC*: Star Functions for Marching Cubes.	Gregory M. Nielson	2003	We describe a modification of the widely used marching cubes method that leads to the useful property that the resulting isosurfaces are locally single valued functions. This implies that conventional interpolation and approximation methods can be used to locally represent the surface. These representations can be used for computing approximations of local surface properties. We utilize this possibility in order to develop algorithms for locally approximating Gaussian and mean curvature, methods for constrained smoothing of isosurface, and techniques for the parameterization of isosurfaces.
IEEE Visualization	Using Deformations for Browsing Volumetric Data.	Michael J. McGuffin,Liviu Tancau,Ravin Balakrishnan	2003	"Many traditional techniques for ""looking inside"" volumetric data involve removing portions of the data, for example using various cutting tools, to reveal the interior. This allows the user to see hidden parts of the data, but has the disadvantage of removing potentially important surrounding contextual information. We explore an alternate strategy for browsing that uses deformations, where the user can cut into and open up, spread apart, or peel away parts of the volume in real time, making the interior visible while still retaining surrounding context. We consider various deformation strategies and present a number of interaction techniques based on different metaphors. Our designs pay special attention to the semantic layers that might compose a volume (e.g. the skin, muscle, bone in a scan of a human). Users can apply deformations to only selected layers, or apply a given deformation to a different degree to each layer, making browsing more flexible and facilitating the visualization of relationships between layers. Our interaction techniques are controlled with direct, ""in place"" manipulation, using pop-up menus and 3D widgets, to avoid the divided attention and awkwardness that would come with panels of traditional widgets. Initial user feedback indicates that our techniques are valuable, especially for showing portions of the data spatially situated in context with surrounding data."
IEEE Visualization	Visualization of Noisy and Biased Volume Data Using First and Second Order Derivative Techniques.	Marc P. Persoon,Iwo Serlie,Frits H. Post,Roel Truyen,Frans Vos	2003	The quality of volume visualization depends strongly on the quality of the underlying data. In virtual colonoscopy, CT data should be acquired at a low radiation dose that results in a low signal-to-noise ratio. Alternatively, MRI data is acquired without ionizing radiation, but suffers from noise and bias (global signal fluctuations). Current volume visualization techniques often do not produce good results with noisy or biased data. This paper describes methods for volume visualization that deal with these imperfections. The techniques are based on specially adapted edge detectors using first and second order derivative filters. The filtering is integrated into the the visualization process. The first order derivative method results in good quality images but suffers from localization bias. The second order method has better surface localization, especially in highly curved areas. It guarantees minimal detail smoothing resulting in a better visualization of polyps.
IEEE Visualization	Holographic Video Display of Time-Series Volumetric Medical Data.	Wendy Plesniak,Michael Halle,Steven D. Pieper,William M. Wells III,Marianna Jakab,Dominik S. Meier,Stephen A. Benton,Charles R. G. Guttmann,Ron Kikinis	2003	We describe an animated electro-holographic visualization of brain lesions due to the progression of multiple sclerosis. A research case study is used which documents the expression of visible brain lesions in a series of magnetic resonance imaging (MRI) volumes collected over the interval of one year. Some of the salient information resident within this data is described, and the motivation for using a dynamic spatial display to explore its spatial and temporal characteristics is stated. We provide a brief overview of spatial displays in medical imaging applications, and then describe our experimental visualization pipeline, from the processing of MRI datasets, through model construction, computer graphic rendering, and hologram encoding. The utility, strengths and shortcomings of the electro-holographic visualization are described and future improvements are suggested.
IEEE Visualization	Producing High Quality Visualizations of Large-Scale Simulations.	Voicu Popescu,Christoph M. Hoffmann,Sami Kilic,Mete Sozen,Scott Meador	2003	This paper describes the work of a team of researchers in computer graphics, geometric computing, and civil engineering to produce a visualization of the September 2001 attack on the Pentagon. The immediate motivation for the project was to understand the behavior of the building under the impact. The longer term motivation was to establish a path for producing high-quality visualizations of large scale simulations. The first challenge was managing the enormous complexity of the scene to fit within the limits of state-of-the art simulation software systems and supercomputing resources. The second challenge was to integrate the simulation results into a high-quality visualization. To meet this challenge, we implemented a custom importer that simplifies and loads the massive simulation data in a commercial animation system. The surrounding scene is modeled using image-based techniques and is also imported in the animation system where the visualization is produced. A specific issue for us was to federate the simulation and the animation systems, both commercial systems not under our control and following internally different conceptualizations of geometry and animation. This had to be done such that scalability was achieved. The reusable link created between the two systems allows communicating the results to non-specialists and the public at large, as well as facilitating communication in teams with members having diverse technical backgrounds.
IEEE Visualization	A Framework for Sample-based Rendering with O-buffers.	Huamin Qu,Arie E. Kaufman,Ran Shao,Ankush Kumar	2003	We present an innovative modeling and rendering primitive, called the O-buffer, for sample-based graphics, such as images, volumes, and points. The 2D or 3D O-buffer is in essence a conventional image or a volume, respectively, except that samples are not restricted to a regular grid. A sample position in the O-buffer is recorded as an offset to the nearest grid point of a regular base grid (hence the name O-buffer). The offset is typically quantized for compact representation and efficient rendering. The O-buffer emancipates pixels and voxels from the regular grids and can greatly improve the modeling power of images and volumes. It is a semi-regular structure which lends itself to efficient construction and rendering. Image quality can be improved by storing more spatial information with samples and by avoiding multiple resamplings and delaying reconstruction to the final rendering stage. Using O-buffers, more accurate multi-resolution representations can be developed for images and volumes. It can also be exploited to represent and render unstructured primitives, such as points, particles, curvilinear or irregular volumes. The O-buffer is therefore a uniform representation for a variety of graphics primitives and supports mixing them in the same scene. We demonstrate the effectiveness of the O-buffer with hierarchical O-buffers, layered depth O-buffers, and hybrid volume rendering with O-buffers.
IEEE Visualization	Information and Scientific Visualization: Separate but Equal or Happy Together at Last.	Theresa-Marie Rhyne,Melanie Tory,Tamara Munzner,Matthew O. Ward,Chris Johnson,David H. Laidlaw	2003	Information and Scientific Visualization: Separate but Equal or Happy Together at Last.
IEEE Visualization	Visually Accurate Multi-Field Weather Visualization.	Kirk Riley,David S. Ebert,Charles D. Hansen,Jason J. Levit	2003	Weather visualization is a difficult problem because it comprises volumetric multi-field data and traditional surface-based approaches obscure details of the complex three-dimensional structure of cloud dynamics. Therefore, visually accurate volumetric multi-field visualization of storm scale and cloud scale data is needed to effectively and efficiently communicate vital information to weather forecasters, improving storm forecasting, atmospheric dynamics models, and weather spotter training. We have developed a new approach to multi-field visualization that uses field specific, physically-based opacity, transmission, and lighting calculations per-field for the accurate visualization of storm and cloud scale weather data. Our approach extends traditional transfer function approaches to multi-field data and to volumetric illumination and scattering.
IEEE Visualization	Visualization of Volume Data with Quadratic Super Splines.	Christian Rössl,Frank Zeilfelder,Günther Nürnberger,Hans-Peter Seidel	2003	We develop a new approach to reconstruct non-discrete models from gridded volume samples. As a model, we use quadratic trivariate super splines on a uniform tetrahedral partition ¿. The approximating splines are determined in a natural and completely symmetric way by averaging local data samples, such that appropriate smoothness conditions are automatically satisfied. On each tetra-hedron of ¿, the quasi-interpolating spline is a polynomial of total degree two which provides several advantages including efficient computation, evaluation and visualization of the model. We apply Bernstein-Bézier techniques well-known in CAGD to compute and evaluate the trivariate spline and its gradient. With this approach the volume data can be visualized efficiently e.g. with isosurface ray-casting. Along an arbitrary ray the splines are univariate, piecewise quadratics and thus the exact intersection for a prescribed isovalue can be easily determined in an analytic and exact way. Our results confirm the efficiency of the quasi-interpolating method and demonstrate high visual quality for rendered isosurfaces.
IEEE Visualization	Exploring Curved Anatomic Structures with Surface Sections.	Laurent Saroul,Sebastian Gerlach,Roger D. Hersch	2003	The extraction of planar sections from volume images is the most commonly used technique for inspecting and visualizing anatomic structures.We propose to generalize the concept of planar section to the extraction of curved cross-sections (free form surfaces). Compared with planar slices, curved cross-sections may easily follow the trajectory of tubular structures and organs such as the aorta or the colon.They may be extracted from a 3D volume, displayed as a 3D view and possibly flattened.Flattening of curved cross-sections allows to inspect spatially complex relationship between anatomic structures and their neighbourhood.They also allow to carry out measurements along a specific orientation.For the purpose of facilitating the interactive specification of free form surfaces, users may navigate in real time within the body and select the slices on which the surface control points will be positioned.Immediate feedback is provided by displaying boundary curves as cylindrical markers within a 3D view composed of anatomic organs, planar slices and possibly free form surface sections.Extraction of curved surface sections is an additional service that is available online as a Java applet (http://visiblehuman.epfl.ch).It may be used as an advanced tool for exploring and teaching anatomy.
IEEE Visualization	Compression Domain Volume Rendering.	Jens Schneider,Rüdiger Westermann	2003	A survey of graphics developers on the issue of texture mapping hardware for volume rendering would most likely find that the vast majority of them view limited texture memory as one of the most serious drawbacks of an otherwise fine technology. In this paper, we propose a compression scheme for static and time-varying volumetric data sets based on vector quantization that allows us to circumvent this limitation. We describe a hierarchical quantization scheme that is based on a multiresolution covariance analysis of the original field. This allows for the efficient encoding of large-scale data sets, yet providing a mechanism to exploit temporal coherence in non-stationary fields. We show, that decoding and rendering the compressed data stream can be done on the graphics chip using programmable hardware. In this way, data transfer between the CPU and the graphics processing unit (GPU) can be minimized thus enabling flexible and memory efficient real-time rendering options. We demonstrate the effectiveness of our approach by demonstrating interactive renditions of Gigabyte data sets at reasonable fidelity on commodity graphics hardware.
IEEE Visualization	Feature Space Analysis of Unstructured Meshes.	Ariel Shamir	2003	Unstructured meshes are often used in simulations and imaging applications. They provide advanced flexibility in modeling abilities but are more difficult to manipulate and analyze than regular data. This work provides a novel approach for the analysis of unstructured meshes using feature-space clustering and feature-detection. Analyzing and revealing underlying structures in data involve operators on both spatial and functional domains. Slicing concentrates more on the spatial domain, while iso-surfacing or volume-rendering concentrate more on the functional domain. Nevertheless, many times it is the combination of the two domains which provides real insight on the structure of the data. In this work a combined feature-space is defined on top of unstructured meshes in order to search for structure in the data. A point in feature-space includes the spatial coordinates of the point in the mesh domain and all chosen attributes defined on the mesh. A distance measures between points in feature-space is defined enabling the utilization of clustering using the mean shift procedure (previously used for images) on unstructured meshes. Feature space analysis is shown to be useful for feature-extraction, for data exploration and partitioning.
IEEE Visualization	Fast Volume Segmentation With Simultaneous Visualization Using Programmable Graphics Hardware.	Anthony Sherbondy,Michael Houston,Sandy Napel	2003	Segmentation of structures from measured volume data, such as anatomy in medical imaging, is a challenging data-dependent task. In this paper, we present a segmentation method that leverages the parallel processing capabilities of modern programmable graphics hardware in order to run significantly faster than previous methods. In addition, collocating the algorithm computation with the visualization on the graphics hardware circumvents the need to transfer data across the system bus, allowing for faster visualization and interaction. This algorithm is unique in that it utilizes sophisticated graphics hardware functionality (i.e., floating point precision, render to texture, computational masking, and fragment programs) to enable fast segmentation and interactive visualization.
IEEE Visualization	Signed Distance Transform Using Graphics Hardware.	Christian Sigg,Ronald Peikert,Markus H. Gross	2003	This paper presents a signed distance transform algorithm using graphics hardware, which computes the scalar valued function of the Euclidean distance to a given manifold of co-dimension one. If this manifold is closed and orientable, the distance has a negative sign on one side of the manifold and a positive sign on the other. Triangle meshes are considered for the representation of a two-dimensional manifold and the distance function is sampled on a regular Cartesian grid. In order to achieve linear complexity in the number of grid points, to each primitive we assign a simple poly-hedron enclosing its Voronoi cell. Voronoi cells are known to contain exactly all points that lay closest to its corresponding primitive. Thus, the distance to the primitive only has to be computed for grid points inside its polyhedron. Although Voronoi cells partition space, the polyhedrons enclosing these cells do overlap. In regions where these overlaps occur, the minimum of all computed distances is assigned to a grid point. In order to speed up computations, points inside each polyhedron are determined by scan conversion of grid slices using graphics hardware. For this task, a fragment program is used to perform the nonlinear interpolation and minimization of distance values.
IEEE Visualization	Vicinity Shading for Enhanced Perception of Volumetric Data.	A. James Stewart	2003	This paper presents a shading model for volumetric data which enhances the perception of surfaces within the volume. The model incorporates uniform diffuse illumination, which arrives equally from all directions at each surface point in the volume. This illumination is attenuated by occlusions in the local vicinity of the surface point, resulting in shadows in depressions and crevices. Experiments by other authors have shown that perception of a surface is superior under uniform diffuse lighting, compared to illumination from point source lighting.
IEEE Visualization	Extraction of topologically simple isosurfaces from volume datasets.	Andrzej Szymczak,James Vanderhyde	2003	There are numerous algorithms in graphics and visualization whose performance is known to decay as the topological complexity of the input increases. On the other hand, the standard pipeline for 3D geometry acquisition often produces 3D models that are topologically more complex than their real forms. We present a simple and efficient algorithm that allows us to simplify the topology of an iso-surface by altering the values of some number of voxels. Its utility and performance are demonstrated on several examples, including signed distance functions from polygonal models and CT scans.
IEEE Visualization	Shape Simplification Based on the Medial Axis Transform.	Roger C. Tam,Wolfgang Heidrich	2003	We present a new algorithm for simplifying the shape of 3D objects by manipulating their medial axis transform (MAT). From an unorganized set of boundary points, our algorithm computes the MAT, decomposes the axis into parts, then selectively removes a subset of these parts in order to reduce the complexity of the overall shape. The result is a simplified MAT that can be used for a variety of shape operations. In addition, a polygonal surface of the resulting shape can be directly generated from the filtered MAT using a robust surface reconstruction method. The algorithm presented is shown to have a number of advantages over other existing approaches.
IEEE Visualization	Visibility based methods and assessment for detail-recovery.	Marco Tarini,Paolo Cignoni,Roberto Scopigno	2003	"In this paper we propose a new method for the creation of normal maps for recovering the detail on simplified meshes and a set of objective techniques to metrically evaluate the quality of different recovering techniques. The proposed techniques, that automatically produces a normal-map texture for a simple 3D model that ""imitates"" the high frequency detail originally present in a second, much higher resolution one, is based on the computation of per-texel visibility and self-occlusion information. This information is used to define a point-to-point correspondence between simplified and hires meshes. Moreover, we introduce a number of criteria for measuring the quality (visual or otherwise) of a given mapping method, and provide efficient algorithms to implement them. Lastly, we apply them to rate different mapping methods, including the widely used ones and the new one proposed here."
IEEE Visualization	3D IBFV: Hardware-Accelerated 3D Flow Visualization.	Alexandru Telea,Jarke J. van Wijk	2003	We present a hardware-accelerated method for visualizing 3D flow fields. The method is based on insertion, advection, and decay of dye. To this aim, we extend the texture-based IBFV technique presented in [van Wijk 2001] for 2D flow visualization in two main directions. First, we decompose the 3D flow visualization problem in a series of 2D instances of the mentioned IBFV technique. This makes our method benefit from the hardware acceleration the original IBFV technique introduced. Secondly, we extend the concept of advected gray value (or color) noise by introducing opacity (or matter) noise. This allows us to produce sparse 3D noise pattern advections, thus address the occlusion problem inherent to 3D flow visualization. Overall, the presented method delivers interactively animated 3D flow, uses only standard OpenGL 1.1 calls and 2D textures, and is simple to understand and implement.
IEEE Visualization	A Visual Exploration Process for the Analysis of Internet Routing Data.	Soon Tee Teoh,Kwan-Liu Ma,Shyhtsun Felix Wu	2003	The Internet pervades many aspects of our lives and is becoming indispensable to critical functions in areas such as commerce, government, production and general information dissemination. To maintain the stability and efficiency of the Internet, every effort must be made to protect it against various forms of attacks, malicious uses, and errors. A key component in the Internet security effort is the routine examination of Internet routing data, which unfortunately can be too large and complicated to browse directly. We have developed an interactive visualization process which proves to be very effective for the analysis of Internet routing data. In this application paper, we show how each step in the visualization process helps direct the analysis and glean insights from the data. These insights include the discovery of patterns, detection of faults and abnormal events, understanding of event correlations, formation of causation hypotheses, and classification of anomalies. We also discuss lessons learned in our visual analysis study.
IEEE Visualization	Saddle Connectors - An Approach to Visualizing the Topological Skeleton of Complex 3D Vector Fields.	Holger Theisel,Tino Weinkauf,Hans-Christian Hege,Hans-Peter Seidel	2003	One of the reasons that topological methods have a limited popularity for the visualization of complex 3D flow fields is the fact that such topological structures contain a number of separating stream surfaces. Since these stream surfaces tend to hide each other as well as other topological features, for complex 3D topologies the visualizations become cluttered and hardly interpretable. This paper proposes to use particular stream lines called saddle connectors instead of separating stream surfaces and to depict single surfaces only on user demand. We discuss properties and computational issues of saddle connectors and apply these methods to complex flow data. We show that the use of saddle connectors makes topological skeletons available as a valuable visualization tool even for topologically complex 3D flow data.
IEEE Visualization	Mental Registration of 2D and 3D Visualizations (An Empirical Study).	Melanie Tory	2003	2D and 3D views are used together in many visualization domains, such as medical imaging, flow visualization, oceanographic visualization, and computer aided design (CAD). Combining these views into one display can be done by: (1) orientation icon (i.e., separate windows), (2) in-place methods (e.g., clip and cutting planes), and (3) a new method called ExoVis. How 2D and 3D views are displayed affects ease of mental registration (understanding the spatial relationship between views), an important factor influencing user performance. This paper compares the above methods in terms of their ability to support mental registration. Empirical results show that mental registration is significantly easier with in-place displays than with ExoVis, and significantly easier with ExoVis than with orientation icons. Different mental transformation strategies can explain this result. The results suggest that ExoVis may be a better alternative to orientation icons when in-place displays are not appropriate (e.g, when in-place methods hide data or cut the 3D view into several pieces).
IEEE Visualization	A Novel Interface for Higher-Dimensional Classification of Volume Data.	Fan-Yin Tzeng,Eric B. Lum,Kwan-Liu Ma	2003	In the traditional volume visualization paradigm, the user specifies a transfer function that assigns each scalar value to a color and opacity by defining an opacity and a color map function.The transfer function has two limitations.First, the user must define curves based on histogram and value rather than seeing and working with the volume itself.Second, the transfer function is inflexiblein classifying regions of interest, where values at a voxel such as intensity and gradient are used to differentiate material, not taking into account additional properties such as texture and position. We describe an intuitive user interface for specifying the classification functions that consists of the users painting directly on sample slices of the volume.These painted regions are used to automatically define high-dimensional classification functions that can be implemented in hardware for interactive rendering.The classification of the volume is iteratively improved as the user paints samples, allowing intuitive and efficient viewing of materials of interest.
IEEE Visualization	Effectively Visualizing Multi-Valued Flow Data Using Color and Texture.	Timothy Urness,Victoria Interrante,Ivan Marusic,Ellen Longmire,Bharathram Ganapathisubramani	2003	"In this paper we offer several new insights and techniques for effectively using color and texture to simultaneously convey information about multiple 2D scalar and vector distributions, in a way that facilitates allowing each distribution to be understood both individually and in the context of one or more of the other distributions. Specifically,we introduce the concepts of: - 'color weaving' for simultaneously representing information about multiple co-located color encoded distributions, and - 'texture stitching' for achieving more spatially accurate multi-frequency line integral convolution representations of combined scalar and vector distributions. The target application for our research is the definition, detection and visualization of regions of interest in a turbulent boundary layer flow at moderate Reynolds number. In this work, we examine and analyze streamwise-spanwise planes of three-component velocity vectors with the goal of identifying and characterizing spatially organized packets of hairpin vortices."
IEEE Visualization	Feature-Sensitive Subdivision and Isosurface Reconstruction.	Gokul Varadhan,Shankar Krishnan,Young J. Kim,Dinesh Manocha	2003	We present improved subdivision and isosurface reconstruction algorithms for polygonizing implicit surfaces and performing accurate geometric operations. Our improved reconstruction algorithm uses directed distance fields [Kobbelt et al. 2001] to detect multiple intersections along an edge, separates them into components and reconstructs an isosurface locally within each component using the dual contouring algorithm [Ju et al. 2002]. It can reconstruct thin features without creating handles and results in improved surface extraction from volumetric data. Our subdivision algorithm takes into account sharp features that arise from intersecting surfaces or Boolean operations and generates an adaptive grid such that each voxel has at most one sharp feature. The subdivision algorithm is combined with our improved reconstruction algorithm to compute accurate polygonization of Boolean combinations or offsets of complex primitives that faithfully reconstruct the sharp features. We have applied these algorithms to polygonize complex CAD models designed using thousands of Boolean operations on curved primitives.
IEEE Visualization	Hardware-Based Nonlinear Filtering and Segmentation using High-Level Shading Languages.	Ivan Viola,Armin Kanitsar,Eduard Gröller	2003	Non-linear filtering is an important task for volume analysis.This paper presents hardware-based implementations of various non-linear filters for volume smoothing with edge preservation.TheCg high-level shading language is used in combination with latest PC consumer graphics hardware.Filtering is divided into pervertex and per-fragment stages.In both stages we propose techniques to increase the filtering performance.The vertex program pre-computes texture coordinates in order to address all contributing input samples of the operator mask.Thus additional computations are avoided in the fragment program.The presented fragment programs preserve cache coherence, exploit 4D vector arithmetic, and internal fixed point arithmetic to increase performance.We show the applicability of non-linear filters as part of a GPU-based segmentation pipeline.The resulting binary mask is compressed and decompressed in the graphics memory on-the-fly.
IEEE Visualization	Real-Time Refinement and Simplification of Adaptive Triangular Meshes.	Vasily Volkov,Ling Li	2003	In this paper we present a generic method for incremental mesh adaptation based on hierarchy of semi-regular meshes. Our method supports any refinement rule mapping vertices onto vertices such as 1-to-4 split or \sqrt3-subdivision. Resulting adaptive mesh has subdivision connectivity and hence good aspect ratio of triangles. Hierarchic representation of the mesh allows incremental local refinement and simplification operations exploiting frame-to-frame coherence. We also present an out-of-core storage layout scheme designed for semi-regular meshes of arbitrary subdivision connectivity. It provides high cache coherency in the data retrieval and relies on the interleaved storage of resolution levels and maintaining good geometrical proximity within each level. The efficiency of the proposed method is demonstrated with applications in physically-based cloth simulation, real-time terrain visualization and procedural modeling.
IEEE Visualization	Quasi-Static Approximation for 6 Degrees-of-Freedom Haptic Rendering.	Ming Wan,William A. McNeely	2003	In this paper, we propose a quasi-static approximation (QSA) approach to simulate the movement of the movable object in 6-degrees-of-freedom (DOF) haptic rendering. In our QSA approach, we solve for static equilibrium during each haptic time step, ignoring any dynamical properties such as inertia. The major contribution of this approach is to overcome the computational instability problem in overly stiff systems arising from numerical integration of second-order differential equations in previous dynamic models. Our primary experimental results on both simulated aircraft geometry and a large-scale real-world aircraft engine showed that our QSA approach was capable of maintaining the 1000Hz haptic refresh rate with more robust collision avoidance and more reliable force and torque feedback.
IEEE Visualization	Hardware-Based Ray Casting for Tetrahedral Meshes.	Manfred Weiler,Martin Kraus,Markus Merz,Thomas Ertl	2003	We present the first implementation of a volume ray casting algorithm for tetrahedral meshes running on off-the-shelf programmable graphics hardware. Our implementation avoids the memory transfer bottleneck of the graphics bus since the complete mesh data is stored in the local memory of the graphics adapter and all computations, in particular ray traversal and ray integration, are performed by the graphics processing unit. Analogously to other ray casting algorithms, our algorithm does not require an expensive cell sorting. Provided that the graphics adapter offers enough texture memory, our implementation performs comparable to the fastest published volume rendering algorithms for unstructured meshes. Our approach works with cyclic and/or non-convex meshes and supports early ray termination. Accurate ray integration is guaranteed by applying pre-integrated volume rendering. In order to achieve almost interactive modifications of transfer functions, we propose a new method for computing three-dimensional pre-integration tables.
IEEE Visualization	A Texture-Based Framework for Spacetime-Coherent Visualization of Time-Dependent Vector Fields.	Daniel Weiskopf,Gordon Erlebacher,Thomas Ertl	2003	We propose Unsteady Flow Advection-Convolution (UFAC) as a novel visualization approach for unsteady flows. It performs time evolution governed by pathlines, but builds spatial correlation according to instantaneous streamlines whose spatial extent is controlled by the flow unsteadiness. UFAC is derived from a generic framework that provides spacetime-coherent dense representations of time dependent-vector fields by a two-step process: 1) construction of continuous trajectories in spacetime for temporal coherence, and 2) convolution along another set of paths through the above spacetime for spatially correlated patterns. Within the framework, known visualization techniques-such as Lagrangian-Eulerian Advection, Image-Based Flow Visualization, Unsteady Flow LIC, and Dynamic LIC-can be reproduced, often with better image quality, higher performance, or increased flexibility of the visualization style. Finally, we present a texture-based discretization of the framework and its interactive implementation on graphics hardware, which allows the user to gradually balance visualization speed against quality.
IEEE Visualization	A Frequency-Sensitive Point Hierarchy for Images and Volumes.	Tomihisa Welsh,Klaus Mueller	2003	This paper introduces a method for converting an image or volume sampled on a regular grid into a space-efficient irregular point hierarchy. The conversion process retains the original frequency characteristics of the dataset by matching the spatial distribution of sample points with the required frequency. To achieve good blending, the spherical points commonly used in volume rendering are generalized to ellipsoidal point primitives. A family of multiresolution, oriented Gabor wavelets provide the frequency-space analysis of the dataset. The outcome of this frequency analysis is the reduced set of points, in which the sampling rate is decreased in originally oversampled areas. During rendering, the traversal of the hierarchy can be controlled by any suitable error metric or quality criteria. The local level of refinement is also sensitive to the transfer function. Areas with density ranges mapped to high transfer function variability are rendered at higher point resolution than others. Our decomposition is flexible and can be used for iso-surface rendering, alpha compositing and X-ray rendering of volumes. We demonstrate our hierarchy with an interactive splatting volume renderer, in which the traversal of the point hierarchy for rendering is modulated by a user-specified frame rate.
IEEE Visualization	Image Based Flow Visualization for Curved Surfaces.	Jarke J. van Wijk	2003	A new method for the synthesis of dense, vector-field aligned textures on curved surfaces is presented, called IBFVS. The method is based on Image Based Flow Visualization (IBFV). In IBFV two-dimensional animated textures are produced by defining each frame of a flow animation as a blend between a warped version of the previous image and a number of filtered white noise images. We produce flow aligned texture on arbitrary three-dimensional triangle meshes in the same spirit as the original method: Texture is generated directly in image space. We show that IBFVS is efficient and effective. High performance (typically fifty frames or more per second) is achieved by exploiting graphics hardware. Also, IBFVS can easily be implemented and a variety of effects can be achieved. Applications are flow visualization and surface rendering. Specifically, we show how to visualize the wind field on the earth and how to render a dirty bronze bunny.
IEEE Visualization	High Dimensional Direct Rendering of Time-Varying Volumetric Data.	Jonathan Woodring,Chaoli Wang,Han-Wei Shen	2003	We present an alternative method for viewing time-varying volumetric data. We consider such data as a four-dimensional data field, rather than considering space and time as separate entities. If we treat the data in this manner, we can apply high dimensional slicing and projection techniques to generate an image hyperplane. The user is provided with an intuitive user interface to specify arbitrary hyperplanes in 4D, which can be displayed with standard volume rendering techniques. From the volume specification, we are able to extract arbitrary hyperslices, combine slices together into a hyperprojection volume, or apply a 4D raycasting method to generate the same results. In combination with appropriate integration operators and transfer functions, we are able to extract and present different space-time features to the user.
IEEE Visualization	Visualization Experiences and Issues in Deep Space Exploration.	John Wright,Scott Burleigh,Makoto Maruya,Scott Maxwell,René Pischel	2003	Visualization tools play a key role in the exploration of outer space. Since it is difficult and expensive to send humans to other planets, immersive visualization of such hostile environments is as close as we will get for some time. Visualization is also used in a variety of supporting roles for deep space missions, from simulation and rehearsal of planned operations to analysis of spacecraft state to analysis of science data returned from a variety of instruments. The panelists will discuss their experiences in collecting data in deep space, transmitting it to Earth, processing and visualizing it here, and using the visualization to drive the continued mission. This closes the loop, making missions more responsive to their environment, particularly in-situ operations on planetary surfaces and within planetary atmospheres.
IEEE Visualization	Piecewise C1 Continuous Surface Reconstruction of Noisy Point Cloud via Local Implicit Quadric Regression.	Hui Xie,Jianning Wang,Jing Hua,Hong Qin,Arie E. Kaufman	2003	"This paper addresses the problem of surface reconstruction of highly noisy point clouds. The surfaces to be reconstructed are assumed to be 2-manifolds of piecewise C 1 continuity, with isolated small irregular regions of high curvature, sophisticated local topology or abrupt burst of noise. At each sample point, a quadric field is locally fitted via a modified moving least squares method. These locally fitted quadric fields are then blended together to produce a pseudo-signed distance field using Shepard's method. We introduce a prioritized front growing scheme in the process of local quadrics fitting. Flatter surface areas tend to grow faster. The already fitted regions will subsequently guide the fitting of those irregular regions in their neighborhood."
IEEE Visualization	Interactive View-Dependent Rendering with Conservative Occlusion Culling in Complex Environments.	Sung-Eui Yoon,Brian Salomon,Dinesh Manocha	2003	This paper presents a novel algorithm combining view-dependent rendering and conservative occlusion culling for interactive display of complex environments. A vertex hierarchy of the entire scene is decomposed into a cluster hierarchy through a novel clustering and partitioning algorithm. The cluster hierarchy is then used for view-frustum and occlusion culling. Using hardware accelerated occlusion queries and frame-to-frame coherence, a potentially visible set of clusters is computed. An active vertex front and face list is computed from the visible clusters and rendered using vertex arrays. The integrated algorithm has been implemented on a Pentium IV PC with a NVIDIA GeForce 4 graphics card and applied in two complex environments composed of millions of triangles. The resulting system can render these environments at interactive rates with little loss in image quality and minimal popping artifacts.
IEEE Visualization	Voxels on Fire.	Ye Zhao,Xiaoming Wei,Zhe Fan,Arie E. Kaufman,Hong Qin	2003	We introduce a method for the animation of fire propagation and the burning consumption of objects represented as volumetric data sets. Our method uses a volumetric fire propagation model based on an enhanced distance field. It can simulate the spreading of multiple fire fronts over a specified isosurface without actually having to create that isosurface. The distance field is generated from a specific shell volume that rapidly creates narrow spatial bands around the virtual surface of any given isovalue. The complete distance field is then obtained by propagation from the initial bands. At each step multiple fire fronts can evolve simultaneously on the volumetric object. The flames of the fire are constructed from streams of particles whose movement is regulated by a velocity field generated with the hardware-accelerated Lattice Boltzmann Model (LBM). The LBM provides a physically-based simulation of the air flow around the burning object. The object voxels and the splats associated with the flame particles are rendered in the same pipeline so that the volume data with its external and internal structures can be displayed along with the fire.
IEEE Visualization	HyperLIC.	Xiaoqiang Zheng,Alex Pang	2003	We introduce a new method for visualizing symmetric tensor fields. The technique produces images and animations reminiscent of line integral convolution (LIC). The technique is also slightly related to hyperstreamlines in that it is used to visualize tensor fields. However, the similarity ends there. HyperLIC uses a multi-pass approach to show the anisotropic properties in a 2D or 3D tensor field. We demonstrate this technique using data sets from computational fluid dynamics as well as diffusion-tensor MRI.
IEEE Visualization	Heart-Muscle Fiber Reconstruction from Diffusion Tensor MRI.	Leonid Zhukov,Alan H. Barr	2003	In this paper we use advanced tensor visualization techniques to study 3D diffusion tensor MRI data of a heart. We use scalar and tensor glyph visualization methods to investigate the data and apply a moving least squares (MLS) fiber tracing method to recover and visualize the helical structure and the orientation of the heart muscle fibers.
IEEE Visualization	14th IEEE Visualization 2003 Conference (VIS 2003), 19-24 October 2003, Seattle, WA, USA	Greg Turk,Jarke J. van Wijk,Robert J. Moorhead II	2003	14th IEEE Visualization 2003 Conference (VIS 2003), 19-24 October 2003, Seattle, WA, USA
Information Visualization	Informed spatial decisions through coordinated views.	Natalia V. Andrienko,Gennady L. Andrienko	2003	According to a commonly accepted view, the process of decision making comprises three major phases: intelligence (situation analysis and problem recognition), design (finding possible variants of problem solution), and choice (evaluation of the options and selection of the most appropriate ones). It is widely recognised that exploratory data visualisation is very helpful during the first phase of the decision-making process, while the other phases require different software tools. In particular, the choice phase is typically supported by various computational methods that find appropriate trade-offs among multiple conflicting criteria taking into account user-specified priorities. Visualisation plays a limited role: in the best case, it is used to represent the final results of the computations. We argue that conscious, well-substantiated choice requires a more extensive use of exploratory visualisation facilities, which need to be properly coordinated with the computational multicriteria decision support methods. Extremely important is a high degree of user interactivity, which allows the user to probe the robustness and quality of computationally derived solutions. We suggest several mechanisms for linking and coordinating visual exploratory tools with two types of computational methods differing in the sort of output they produce. We demonstrate the use of this dynamic link with an example of a real spatially related decision problem.
Information Visualization	GeneVis: simulation and visualization of genetic networks.	Charles A. H. Baker,M. Sheelagh T. Carpendale,Przemyslaw Prusinkiewicz,Michael G. Surette	2003	GeneVis simulates genetic networks and visualizes the process of this simulation interactively, providing a visual environment for exploring the dynamics of genetic regulatory networks. The visualization environment supports several representational modes, which include: an individual protein representation, a protein concentration representation, and a network structure representation. The individual protein representation shows the activities of the individual proteins. The protein concentration representation illustrates the relative spread and concentrations of the different proteins in the simulation. The network structure representation depicts the genetic network dependencies that are present in the simulation. GeneVis includes several interactive viewing tools. These include animated transitions from the individual protein representation to the protein concentration representation and from the individual protein representation to the network structure representation. Three types of lenses are used to provide different views within a representation: fuzzy lenses, base pair lenses, and the network structure ring lens. With a fuzzy lens an alternate representation can be viewed in a selected region. The base pair lenses allow users to reposition genes for better viewing or to minimize interference during the simulation. The ring lens provides detail-in-context viewing of individual levels in the genetic network structure representation.
Information Visualization	Social diffusion patterns in three-dimensional virtual worlds.	Katy Börner,Shashikant Penumarthy	2003	Social diffusion patterns in three-dimensional virtual worlds.
Information Visualization	A model and software system for coordinated and multiple views in exploratory visualization.	Nadia Boukhelifa,Peter J. Rodgers	2003	This paper describes a model for expressing coordination in multiple view visualization systems. We present the model and describe a prototype implementation that illustrates the features of the model. Current visualization systems tend to have an informal and inconsistent approach to coordination. Our model takes a formal approach to describing widely used coordination concepts. The model is based on views sharing abstract objects such as the visualization parameters of the dataflow model. Additionaly, this paper describes how current coordinations in exploratory visualization work and how novel coordinations can be constructed using our model.
Information Visualization	Visual unrolling of network evolution and the analysis of dynamic discourse?	Ulrik Brandes,Steven R. Corman	2003	We introduce a method for visualizing evolving networks. In addition to the intermediate states of the network, it conveys the nature of change between states by unrolling the dynamics of the network. Each modification is shown in a separate layer of a three-dimensional representation, where the stack of layers corresponds to a time line of the evolution. We focus on networks of dynamic discourse as the driving application, but the method extends to any type of networks evolving in similar ways.
Information Visualization	On improved projection techniques to support visual exploration of multi-dimensional data sets.	Eduardo Tejada,Rosane Minghim,Luis Gustavo Nonato	2003	On improved projection techniques to support visual exploration of multi-dimensional data sets.
Information Visualization	Articles in this issue.	Chaomei Chen	2003	Articles in this issue.
Information Visualization	Editorial.	Chaomei Chen	2003	Editorial.
Information Visualization	"Guest Editor's Introduction: Special issue on Selected InfoVis 2002 Papers."	Pak Chung Wong	2003	"Guest Editor's Introduction: Special issue on Selected InfoVis 2002 Papers."
Information Visualization	Using the self-organizing map as a visualization tool in financial benchmarking.	Tomas Eklund,Barbro Back,Hannu Vanharanta,Ari Visa	2003	Using the self-organizing map as a visualization tool in financial benchmarking.
Information Visualization	InterRing: a visual interface for navigating and manipulating hierarchies.	Jing Yang,Matthew O. Ward,Elke A. Rundensteiner,Anilkumar Patro	2003	Radial, space-filling (RSF) techniques for hierarchy visualization have several advantages over traditional node-link diagrams, including the ability to use the display space efficiently while effectively conveying the hierarchy structure. Several RSF systems and tools have been developed to date, each with varying degrees of support for interactive operations such as selection and navigation. In this paper, we describe what we believe to be a complete set of desirable operations on hierarchical structures. We then present InterRing, an RSF hierarchy visualization system that supports a significantly more extensive set of these operations than prior systems. In particular, InterRing supports multifocus distortions, interactive hierarchy reconfiguration, and both semiautomated and manual selection. We show the power and utility of these and other operations, and describe our ongoing efforts to evaluate their effectiveness and usability.
Information Visualization	Coordinating computational and visual approaches for interactive feature selection and multivariate clustering.	Diansheng Guo	2003	Unknown (and unexpected) multivariate patterns lurking in high-dimensional datasets are often very hard to find. This paper describes a human-centered exploration environment, which incorporates a coordinated suite of computational and visualization methods to explore high-dimensional data for uncovering patterns in multivariate spaces. Specifically, it includes: (1) an interactive feature selection method for identifying potentially interesting, multidimensional subspaces from a high-dimensional data space, (2) an interactive, hierarchical clustering method for searching multivariate clusters of arbitrary shape, and (3) a suite of coordinated visualization and computational components centered around the above two methods to facilitate a human-led exploration. The implemented system is used to analyze a cancer dataset and shows that it is efficient and effective for discovering unknown and unexpected multivariate patterns from high-dimensional data.
Information Visualization	Beamtrees: compact visualization of large hierarchies.	Frank van Ham,Jarke J. van Wijk	2003	Beamtrees are a new method for the visualization of large hierarchical data sets, such as directory structures and organization structures. Nodes are shown as stacked circular beams such that both the hierarchical structure as well as the size of nodes are depicted. The dimensions of beams are calculated using a variation of the treemap algorithm. Both a two-dimensional and a three-dimensional variant are presented. A small user study indicated that beamtrees are significantly more effective than nested treemaps and cushion treemaps for the extraction of global hierarchical information.
Information Visualization	A visualization model of interactive knowledge discovery systems and its implementations.	Jianchao Han,Xiaohua Hu,Nick Cercone	2003	We briefly introduce an interactive visualization model, RuleViz, for knowledge discovery and data mining, which consists of five components: data preparation and visualization, interactive data reduction, data preprocessing, pattern discovery, and pattern visualization. With this model, the implementation issues are considered and three implementation paradigms, including image-based paradigm, algorithm-embedded paradigm, and interaction-driven paradigm, are discussed. We implement an interactive Visualization system, AViz, which discovers 3D numerical association rules from large data sets based on the image-based paradigm. The framework of the AViz system is presented and each component is explored. To discretize numerical attributes, three approaches, including equal-sized, bin-packing-based equal-depth, and interaction-based approaches are proposed, and the algorithm for mining and visualizing numerical association rules is developed. Our experimental result on a census data set is illustrated, which shows that the AViz system is useful and helpful for discovering and visualizing numerical association rules.
Information Visualization	Data-mining and Uncertain Reasoning: An Integrated Approach (Book Review).	Xiaohua Hu	2003	Data-mining and Uncertain Reasoning: An Integrated Approach (Book Review).
Information Visualization	Network intrusion visualization with NIVA, an intrusion detection visual and haptic analyzer.	Craig Scott,Kofi Nyarko,Tanya Capers,Jumoke Ladeji-Osias	2003	The rapid growth of malicious activities on worldwide communication networks, such as the Internet, has highlighted the need for efficient intrusion detection systems. The efficiency of traditional intrusion detection systems is limited, in part, by their inability to relay effectively relevant information due to their lack of interactive/immersive technologies. In this paper, we explore several network visualization techniques geared toward intrusion detection on small- and large-scale networks. We also examine the use of haptics in network intrusion visualization. By incorporating concepts from electromagnetics, fluid dynamics, and gravitational theory, we show that haptic technologies can provide another dimension of information critical to the efficient visualization of network intrusion data. Furthermore, we explore the applicability of these visualization techniques in conjunction with commercial network intrusion detectors. Finally, we present a network intrusion visualization application with haptic integration, NIVA, which allows the analyst to interactively investigate as well as efficiently detect structured attacks across time and space using advanced interactive three-dimensional displays.
Information Visualization	Visualizing geographic information: VisualPoints vs. CartoDraw.	Daniel A. Keim,Stephen C. North,Christian Panse,Jörn Schneidewind	2003	Cartograms are a well-known technique for showing geography-related statistical information, such as population demographics and epidemiological data. The basic idea is to distort a map by resizing its regions according to a statistical parameter, but in a way that keeps the map recognizable. In this paper, we deal with the problem of making continuous cartograms that strictly retain the topology of the input mesh. We compare two algorithms that solve the continuous cartogram problem. The first one uses an iterative relocation of vertices based on scanlines. This algorithm explicitly accounts for induced shape error. The second one is based on the Gridfit technique, which uses pixel-based distortion based on a quadtree-like data structure. The basic idea is to insert pixels, the number of which corresponds to a statistical parameter, into the data structure and distort the pixels such that every pixel obtains a unique, nonoverlapping position. Relocation of vertices of the map are positioned using the same distortion. We discuss the results obtained from both methods, compare their shape and area trade-offs as well as their efficiency, and show results from different applications.
Information Visualization	Metric 3D structure in visualizations.	Mats Lind,Geoffrey P. Bingham,Camilla Forsell	2003	A large body of results on the characteristics of human spatial vision suggests that space perception is distorted. Recent studies indicate that the geometry of visual space is best understood as affine. If this is the case, it has far-reaching implications on how 3D visualizations can be successfully employed. For instance, all attempts to build visualization systems where users are expected to discover relations based on Euclidean distances or shapes will be ineffective. Since visualization can, and sometimes does, employ all possible types of depth information and because the results from vision research usually concentrates on one or two such types, three experiments were performed under near-optimal viewing conditions. The aim of the experiments was two-fold: to test whether the earlier findings generalize to shape perception under near-optimal viewing conditions and to get a sense of the size of the error under such conditions. The results show that the findings do generalize and that the errors are large. The implications of these results for successful visualizations are discussed.
Information Visualization	Dendrogram seriation using simulated annealing.	Steven A. Morris,Benyam Asnake,Gary G. Yen	2003	"Seriation is the ordering of the leaves of a dendrogram, such that leaves representing similar items are placed near each other according to some metric, within the constraints of the cluster tree. Such ordering greatly aids the interpretation of the relations represented by the dendrogram and reduces visual misinterpretation caused by unrelated items from different subtrees being placed near each other during random ordering. The seriation method presented here uses simulated annealing to find an approximately optimal dendrogram ordering by minimizing a penalty function. The method employs a 'similarity weighted distance' penalty function that tends to avoid artifacts introduced by the traveling salesman problem algorithms commonly used for dendrogram seriation. Examples are given showing the effectiveness of the method in presenting dendrograms of the structure of a social network, and additional examples show an application for interpreting the structure of a network of journal papers covering the subject of anthrax research."
Information Visualization	Fast multidimensional scaling through sampling, springs and interpolation.	Alistair Morrison,Greg Ross,Matthew Chalmers	2003	"The term 'proximity data' refers to data sets within which it is possible to assess the similarity of pairs of objects. Multidimensional scaling (MDS) is applied to such data and attempts to map high-dimensional objects onto low-dimensional space through the preservation of these similarity relations. Standard MOS techniques have in the past suffered from high computational complexity and, as such, could not feasibly be applied to data sets over a few thousand objects in size. Through a novel hybrid approach based upon stochastic sampling, interpolation and spring models, we have designed an algorithm running in O(N√N). Using Chalmers' 1996 O(N2) spring model as a benchmark for the evaluation of our technique, we compare layout quality and run times using sets of synthetic and real data. Our algorithm executes significantly faster than Chalmers' 1996 algorithm, while producing superior layouts. In reducing complexity and run time, we allow the visualisation of data sets of previously infeasible size. Our results indicate that our method is a solid foundation for interactive and visual exploration of data."
Information Visualization	Space-optimized tree: a connection+enclosure approach for the visualization of large hierarchies.	Quang Vinh Nguyen,Mao Lin Huang	2003	This paper describes a new approach, space-optimized tree, for the visualization and navigation of tree-structured relational data. This technique can be used especially for the display of very large hierarchies in a two-dimensional space. We discuss the advantages and limitations of current techniques of tree visualization. Our strategy is to optimize the drawing of trees in a geometrical plane and maximize the utilization of display space by allowing more nodes and links to be displayed at a limited screen resolution. Space-optimized tree is a connection+enclosure visualization approach that recursively positions children of a subtree into polygon areas and still uses a node-link diagram to present the entire hierarchical structure. To be able to handle the navigation of large hierarchies, we use a new hybrid viewing technique that combines two viewing methods, the modified semantic zooming and a focus+context technique. While the semantic zooming technique can enlarge a particular viewing area by filtering out the rest of tree structure from the visualization, the focus+context technique allows the user to interactively focus, view and browse the entire visual structure with a reasonable high-density display.
Information Visualization	Co-Citation count vs correlation for influence network visualization.	Steven Noel,Chee-Hung Henry Chu,Vijay Raghavan	2003	Co-Citation count vs correlation for influence network visualization.
Information Visualization	A framework for augmenting the visualization of dynamic raster surfaces.	Sanjay Rana,Jason Dykes	2003	Animated sequences of raster images that represent continuously varying surfaces, such as a temporal series of an evolving landform or an attribute series of socio-economic variation, are often used in an attempt to gain insight from ordered sequences of raster spatial data. Despite their aesthetic appeal and condensed nature, such representations are limited in terms of their suitability for prompting ideas and offering insight due to their poor information delivery and the lack of the levels of interactivity that are required to support visualization. Cartographic techniques aim to assist users of geographic information through processes of abstraction, by selecting, simplifying, smoothing and exaggerating when representing an underlying spatial data set graphically. Here we suggest a number of transformations and abstractions that take advantage of these techniques in a specific context-that of addressing the limitations associated with using animated raster surfaces for visualization, and propose them in the context of a framework that can be used to inform practice. The five techniques proposed are spatial and attribute smoothing, temporal interpolation, transformation of the surfaces into a network of morphometric features, the use of a graphic lag or fading and the employment of techniques for conditional interactivity that are appropriate for visualization. These efforts allow us to generate graphical environments that support visualization when using animated sequences of images representing continuous surfaces and are analogous to traditional cartographic techniques, namely, smoothing and exaggeration, simplification, enhancement and the various issues of design. By developing a framework for considering cartography in support of visualization from this particular type of data and phenomenon we aim to highlight the utility of a generically cartographic approach to information visualization. A number of particular techniques originating from computer science and conventional cartography are used in an application of the framework. A suitably interactive software tool is offered for evaluation-to establish the results of applying the framework and demonstrate ways in which we may augment the visualization of dynamic raster surfaces through animation and more generally aim to offer opportunity for insight through cartographic design.
Information Visualization	"Guest editor's introduction: special issue on coordinated and multiple views in exploratory visualization."	Jonathan C. Roberts	2003	"Guest editor's introduction: special issue on coordinated and multiple views in exploratory visualization."
Information Visualization	Application of information visualization techniques to the design of a mathematical mindtool: a usability study.	Kamran Sedig,Sonja Rowhani,Jim Morey,Hai-Ning Liang	2003	Application of information visualization techniques to the design of a mathematical mindtool: a usability study.
Information Visualization	A visual workspace for constructing hybrid multidimensional scaling algorithms and coordinating multiple views.	Greg Ross,Matthew Chalmers	2003	Data can be distinguished according to volume, variable types and distribution, and each of these characteristics imposes constraints upon the choice of applicable algorithms for their visualisation. This has led to an abundance of often disparate algorithmic techniques. Previous work has shown that a hybrid algorithmic approach can be successful in addressing the impact of data volume on the feasibility of multidimensional scaling (MDS). This paper presents a system and framework in which a user can easily explore algorithms as well as their hybrid conjunctions and the data flowing through them. Visual programming and a novel algorithmic architecture let the user semiautomatically define data-flows and the coordination of multiple views of algorithmic and visualisation components. We propose that our approach has two main benefits: significant improvements in run times of MDS algorithms can be achieved, and intermediate views of the data and the visualisation program structure can provide greater insight and control over the visualisation process.
IEEE Trans. Vis. Comput. Graph.	Computing and Rendering Point Set Surfaces.	Marc Alexa,Johannes Behr,Daniel Cohen-Or,Shachar Fleishman,David Levin,Cláudio T. Silva	2003	We advocate the use of point sets to represent shapes. We provide a definition of a smooth manifold surface from a set of points close to the original surface. The definition is based on local maps from differential geometry, which are approximated by the method of moving least squares (MLS). The computation of points on the surface is local, which results in an out-of-core technique that can handle any point set. We show that the approximation error is bounded and present tools to increase or decrease the density of the points, thus allowing an adjustment of the spacing among the points to control the error. To display the point set surface, we introduce a novel point rendering technique. The idea is to evaluate the local maps according to the image resolution. This results in high quality shading effects and smooth silhouettes at interactive frame rates.
IEEE Trans. Vis. Comput. Graph.	Fast Multiresolution Image Operations in the Wavelet Domain.	Iddo Drori,Dani Lischinski	2003	Fast Multiresolution Image Operations in the Wavelet Domain.
IEEE Trans. Vis. Comput. Graph.	Semantic Representation and Correspondence for State-Based Motion Transition.	Golam Ashraf,Kok Cheong Wong	2003	Consistent transition algorithms preserve salient source motion features by establishing feature-based correspondence between motions and accordingly warping them before interpolation. These processes are commonly dubbed as preprocessing in motion transition literature. Current transition methods suffer from a lack of economical and generic preprocessing algorithms. Classical computer vision methods for human motion classification and correspondence are too computationally intensive for computer animation. This paper proposes an analytical framework that combines low-level kinematics analysis and high-level knowledge-based analysis to create states that provide coherent snapshots of body-parts active during the motion. These states are then corresponded via a globally optimal search tree algorithm. The framework proposed here is intuitive, controllable, and delivers results in near real-time. The validity and performance of the proposed system are tangibly proven with extensive experiments.
IEEE Trans. Vis. Comput. Graph.	"Editor's Note."	David S. Ebert	2003	reg
IEEE Trans. Vis. Comput. Graph.	"Editor's Note."	David S. Ebert	2003	"Editor's Note."
IEEE Trans. Vis. Comput. Graph.	Image-Based Techniques in a Hybrid Collision Detector.	George Baciu,Wingo Sai-Keung Wong	2003	Most collision detection methods developed so far are based on geometrical object-space interference tests. While this remains the basic mode of investigation for geometric algorithms, the requirements for interactive rates and complex geometry predominate in commercial applications. In this article, we propose a new mode of collision detection based on an image-space approach. This approach breaks the object-space collision detection bottleneck by distributing the computational load onto the hardware graphics pipeline. The image-space approach, in conjuction with efficient bounding-box strategies in the object-space, has the potential to handle complex object interactions at interactive rates.
IEEE Trans. Vis. Comput. Graph.	Deriving a Particle System from Continuum Mechanics for the Animation of Deformable Objects.	Olaf Etzmuß,Joachim Gross,Wolfgang Straßer	2003	Mass-spring and particle systems have been widely employed in computer graphics to model deformable objects because they allow fast numerical solutions. In this work, we establish a link between these discrete models and classical mathematical elasticity. It turns out that discrete systems can be derived from a continuum model by a finite difference formulation and approximate classical continuum models unless the deformations are large. In this work, we present the derivation of a particle system from a continuum model, compare it to the models of classical elasticity theory, and assess its accuracy. In this way, we gain insight into the way discrete systems work and we are able to specify the correct scaling when the discretization is changed. Physical material parameters that describe materials in continuum mechanics are also used in the derived particle system.
IEEE Trans. Vis. Comput. Graph.	Material Interface Reconstruction.	Kathleen S. Bonnell,Mark A. Duchaineau,Daniel Schikore,Bernd Hamann,Kenneth I. Joy	2003	This paper presents an algorithm for material interface reconstruction for data sets where fractional material information is given as a percentage for each element of the underlying grid. The reconstruction problem is transformed to a problem that analyzes a dual grid, where each vertex in the dual grid has an associated barycentric coordinate tuple that represents the fraction of each material present. Material boundaries are constructed by analyzing the barycentric coordinate tuples of a tetrahedron in material space and calculating intersections with Voronoi cells that represent the regions where one material dominates. These intersections are used to calculate intersections in the Euclidean coordinates of the tetrahedron. By triangulating these intersection points, one creates the material boundary. The algorithm can treat data sets containing any number of materials. The algorithm can also create nonmanifold boundary surfaces if necessary. By clipping the generated material boundaries against the original cells, one can examine the error in the algorithm. Error analysis shows that the algorithm preserves volume fractions within an error range of 0.5 percent per material.
IEEE Trans. Vis. Comput. Graph.	Communicating Centrality in Policy Network Drawings.	Ulrik Brandes,Patrick Kenis,Dorothea Wagner	2003	We introduce a network visualization technique that supports an analytical method applied in the social sciences. Policy network analysis is an approach to study policy making structures, processes, and outcomes, thereby concentrating on relations between policy actors. An important operational concept for the analysis of policy networks is the notion of centrality, i.e., the distinction of actors according to their importance in a relational structure. We integrate this measure in a layout model for networks by mapping structural to geometric centrality. Thus, centrality values and network data can be presented simultaneously and explored interactively.
IEEE Trans. Vis. Comput. Graph.	Texture Synthesis for 3D Shape Representation.	Gabriele Gorla,Victoria Interrante,Guillermo Sapiro	2003	"Considerable evidence suggests that a viewer's perception of the 3D shape of a polygonally-defined object can be significantly affected (either masked or enhanced) by the presence of a surface texture pattern. However, investigations into the specific mechanisms of texture's effect on shape perception are still ongoing and the question of how to design and apply a texture pattern to a surface in order to best facilitate shape perception remains open. Recently, we have suggested that, for anisotropic texture patterns, the accuracy of shape judgments may be significantly affected by the orientation of the surface texture pattern anisotropy with respect to the principal directions of curvature over the surface. However, it has been difficult, until this time, to conduct controlled studies specifically investigating the effect of texture orientation on shape perception because there has been no simple and reliable method for texturing an arbitrary doubly curved surface with a specified input pattern such that the dominant orientation of the pattern everywhere follows a predefined directional vector field over the surface, while seams and projective distortion of the pattern are avoided. In this paper, we present a straightforward and highly efficient method for achieving such a texture and describe how it can potentially be used to enhance shape representation. Specifically, we describe a novel, efficient, automatic algorithm for seamlessly synthesizing, from a sample 2D pattern, a high resolution fitted surface texture in which the dominant orientation of the pattern locally follows a specified vector field over the surface at a per-pixel level and in which seams, projective distortion, and repetition artifacts in the texture pattern are nearly completely avoided. We demonstrate the robustness of our method with a variety of texture swatches applied to standard graphics data sets and we explain how our method can be used to facilitate research in the perception of shape from texture."
IEEE Trans. Vis. Comput. Graph.	Realistic Rendering and Animation of Knitwear.	Yanyun Chen,Stephen Lin,Hua Zhong,Ying-Qing Xu,Baining Guo,Heung-Yeung Shum	2003	We present a framework for knitwear modeling and rendering that accounts for characteristics that are particular to knitted fabrics. We first describe a model for animation that considers knitwear features and their effects on knitwear shape and interaction. With the computed free-form knitwear configurations, we present an efficient procedure for realistic synthesis based on the observation that a single cross section of yarn can serve as the basic primitive for modeling entire articles of knitwear. This primitive, called the lumislice, describes radiance from a yarn cross section that accounts for fine-level interactions among yarn fibers. By representing yarn as a sequence of identical but rotated cross sections, the lumislice can effectively propagate local microstructure over arbitrary stitch patterns and knitwear shapes. The lumislice accommodates varying levels of detail, allows for soft shadow generation, and capitalizes on hardware-assisted transparency blending. These modeling and rendering techniques together form a complete approach for generating realistic knitwear.
IEEE Trans. Vis. Comput. Graph.	Adjoints and Importance in Rendering: An Overview.	Per H. Christensen	2003	Adjoints and Importance in Rendering: An Overview.
IEEE Trans. Vis. Comput. Graph.	External Memory Management and Simplification of Huge Meshes.	Paolo Cignoni,Claudio Montani,Claudio Rocchini,Roberto Scopigno	2003	Very large triangle meshes, i.e., meshes composed of millions of faces, are becoming common in many applications. Obviously, processing, rendering, transmission, and archiving of these meshes are not simple tasks. Mesh simplification and LOD management are a rather mature technology that, in many cases, can efficiently manage complex data. But, only a few available systems can manage meshes characterized by a huge size: RAM size is often a severe bottleneck. In this paper, we present a data structure called Octree-based External Memory Mesh (OEMM). It supports external memory management of complex meshes, loading dynamically in main memory only the selected sections and preserving data consistency during local updates. The functionalities implemented on this data structure (simplification, detail preservation, mesh editing, visualization, and inspection) can be applied to huge triangles meshes on low-cost PC platforms. The time overhead due to the external memory management is affordable. Results of the test of our system on complex meshes are presented.
IEEE Trans. Vis. Comput. Graph.	A Survey of Visibility for Walkthrough Applications.	Daniel Cohen-Or,Yiorgos Chrysanthou,Cláudio T. Silva,Frédo Durand	2003	A Survey of Visibility for Walkthrough Applications.
IEEE Trans. Vis. Comput. Graph.	Cubist Style Rendering from Photographs.	John P. Collomosse,Peter M. Hall	2003	The contribution of this paper is a novel nonphotorealistic rendering (NPR) technique, influenced by the style of Cubist art. Specifically, we are motivated by artists such as Picasso and Braque, who produced art work by composing elements of a scene taken from multiple points of view; paradoxically, such compositions convey a sense of motion without assuming temporal dependence between views. Our method accepts a set of two-dimensional images as input and produces a Cubist style painting with minimal user interaction. We use salient features identified within the image set, such as eyes, noses, and mouths, as compositional elements; we believe the use of such features to be a unique contribution to NPR. Before composing features into a final image, we geometrically distort them to produce the more angular forms common in Cubist art. Finally, we render the composition to give a painterly effect, using an automatic algorithm. This paper describes our method, illustrating the application of our algorithm with a gallery of images. We conclude with a critical appraisal and suggest the use of ¿high-level¿ features is of interest to NPR.
IEEE Trans. Vis. Comput. Graph.	"Editor-in-Chief's Farewell."	Hans Hagen	2003	"Editor-in-Chief's Farewell."
IEEE Trans. Vis. Comput. Graph.	A Model for Volume Lighting and Modeling.	Joe Kniss,Simon Premoze,Charles D. Hansen,Peter Shirley,Allen McPherson	2003	Direct volume rendering is a commonly used technique in visualization applications. Many of these applications require sophisticated shading models to capture subtle lighting effects and characteristics of volumetric data and materials. For many volumes, homogeneous regions pose problems for typical gradient-based surface shading. Many common objects and natural phenomena exhibit visual quality that cannot be captured using simple lighting models or cannot be solved at interactive rates using more sophisticated methods. We present a simple yet effective interactive shading model which captures volumetric light attenuation effects that incorporates volumetric shadows, an approximation to phase functions, an approximation to forward scattering, and chromatic attenuation that provides the subtle appearance of translucency. We also present a technique for volume displacement or perturbation that allows realistic interactive modeling of high frequency detail for both real and synthetic volumetric data.
IEEE Trans. Vis. Comput. Graph.	Polynomial Surfaces Interpolating Arbitrary Triangulations.	Stefanie Hahmann,Georges-Pierre Bonneau	2003	Triangular Bézier patches are an important tool for defining smooth surfaces over arbitrary triangular meshes. The previously introduced 4-split method interpolates the vertices of a 2-manifold triangle mesh by a set of tangent plane continuous triangular Bézier patches of degree five. The resulting surface has an explicit closed form representation and is defined locally. In this paper, we introduce a new method for visually smooth interpolation of arbitrary triangle meshes based on a regular 4-split of the domain triangles. Ensuring tangent plane continuity of the surface is not enough for producing an overall fair shape. Interpolation of irregular control-polygons, be that in 1D or in 2D, often yields unwanted undulations. Note that this undulation problem is not particular to parametric interpolation, but also occurs with interpolatory subdivision surfaces. Our new method avoids unwanted undulations by relaxing the constraint of the first derivatives at the input mesh vertices: The tangent directions of the boundary curves at the mesh vertices are now completely free. Irregular triangulations can be handled much better in the sense that unwanted undulations due to flat triangles in the mesh are now avoided.
IEEE Trans. Vis. Comput. Graph.	Three-Dimensional Flow Characterization Using Vector Pattern Matching.	Einar Heiberg,Tino Ebbers,Lars Wigström,Matts Karlsson	2003	Three-Dimensional Flow Characterization Using Vector Pattern Matching.
IEEE Trans. Vis. Comput. Graph.	Fast and Intuitive Metamorphosis of 3D Polyhedral Models Using SMCC Mesh Merging Scheme.	Tong-Yee Lee,Po-Hua Huang	2003	A very fast and intuitive approach to generate the metamorphosis of two genus 0 3D polyhedral models is presented. There are two levels of correspondence specified by animators to control morphs. The higher level requires the animators to specify scatter features to decompose the input models into several corresponding patches. The lower level optionally allows the animators to specify extra features on each corresponding patch for finer correspondence control. Once these two levels of correspondence are established, the proposed schemes automatically and efficiently establish a complete one-to-one correspondence between two models. We propose a novel technique called SMCC (Structures of Minimal Contour Coverage) to efficiently and robustly merge corresponding embeddings. The SMCC scheme can compute merging in linear time. The performance of the proposed methods is comparable to or better than state-of-the-art 3D polyhedral metamorphosis. We demonstrate several examples of aesthetically pleasing morphs, which can be created very quickly and intuitively.
IEEE Trans. Vis. Comput. Graph.	Volumetric Segmentation Using Weibull E-SD Fields.	Jiuxiang Hu,Anshuman Razdan,Gregory M. Nielson,Gerald E. Farin,D. Page Baluch,David G. Capco	2003	Volumetric Segmentation Using Weibull E-SD Fields.
IEEE Trans. Vis. Comput. Graph.	GraphSplatting: Visualizing Graphs as Continuous Fields.	Robert van Liere,Wim C. de Leeuw	2003	This paper introduces GraphSplatting, a technique which transforms a graph into a two-dimensional scalar field. The scalar field can be rendered as a color coded map, a height field, or a set of contours. Splat fields allow for the visualization of arbitrarily large graphs without cluttering. They provide density information which can be used to determine the structure of the graph. The construction, visualization, and interaction with splat fields is discussed. Two applications illustrate the usage of GraphSplatting.
IEEE Trans. Vis. Comput. Graph.	Approximating Digital 3D Shapes by Rational Gaussian Surfaces.	Marcel Jackowski,Martin Satter,A. Ardeshir Goshtasby	2003	A method for approximating spherical topology digital shapes by rational Gaussian (RaG) surfaces is presented. Points in a shape are parametrized by approximating the shape with a triangular mesh, determining parameter coordinates at mesh vertices, and finding parameter coordinates at shape points from interpolation of parameter coordinates at mesh vertices. Knowing the locations and parameter coordinates of the shape points, the control points of a RaG surface are determined to approximate the shape with a required accuracy. The process starts from a small set of control points and gradually increases the control points until the error between the surface and the digital shape reduces to a required tolerance. Both triangulation and surface approximation proceed from coarse to fine. Therefore, the method is particularly suitable for multiresolution creation and transmission of digital shapes over the Internet. Application of the proposed method in editing of 3D shapes is demonstrated.
IEEE Trans. Vis. Comput. Graph.	Topology Preserving Top-Down Compression of 2D Vector Fields Using Bintree and Triangular Quadtrees.	Suresh K. Lodha,Nikolai M. Faaland,Jose C. Renteria	2003	"We present a hierarchical top-down refinement algorithm for compressing 2D vector fields that preserves topology. Our approach is to reconstruct the data set using adaptive refinement that considers topology. The algorithms start with little data and subdivide regions that are most likely to reconstruct the original topology of the given data set. We use two different refinement techniques. The first technique uses bintree subdivision and linear interpolation. The second algorithm is driven by triangular quadtree subdivision with Coons patch quadratic interpolation. We employ local error metrics to measure the quality of compression and as a global metric we compute Earth Mover's Distance (EMD) to measure the deviation from the original topology. Experiments with both analytic and simulated data sets are presented. Results indicate that one can obtain significant compression with low errors without losing topological information. Advantages and disadvantages of different topology preserving compression algorithms are also discussed in the paper."
IEEE Trans. Vis. Comput. Graph.	Improving the Robustness and Accuracy of the Marching Cubes Algorithm for Isosurfacing.	Adriano Lopes,Ken Brodlie	2003	This paper proposes a modification of the Marching Cubes algorithm for isosurfacing, with the intent of improving the representation of the surface in the interior of each grid cell. Our objective is to create a representation which correctly models the topology of the trilinear interpolant within the cell and which is robust under perturbations of the data and threshold value. To achieve this, we identify a small number of key points in the cell interior that are critical to the surface definition. This allows us to efficiently represent the different topologies that can occur, including the possibility of ¿tunnels.¿ The representation is robust in the sense that the surface is visually continuous as the data and threshold change in value. Each interior point lies on the isosurface. Finally, a major feature of our new approach is the systematic method of triangulating the polygon in the cell interior.
IEEE Trans. Vis. Comput. Graph.	Tricubic Interpolation of Discrete Surfaces for Binary Volumes.	Arie Kadosh,Daniel Cohen-Or,Roni Yagel	2003	Binary-defined 3D objects are common in volume graphics and medical imaging as a result of voxelization algorithms, segmentation methods, and binary operations such as clipping. Traditionally, renderings of binary objects suffer from severe image quality problems, especially when one tries to zoom-in and render the binary data from up close. We present a new rendering technique for discrete binary surfaces. The technique is based on distance-based normal estimation, an accelerated ray casting, and a tricubic interpolator. We demonstrate the quality achieved by our method and report on its interactive rendering speed.
IEEE Trans. Vis. Comput. Graph.	Interactive Volume Rendering of Large Sparse Data Sets Using Adaptive Mesh Refinement Hierarchies.	Ralf Kähler,Mark Simon,Hans-Christian Hege	2003	Interactive Volume Rendering of Large Sparse Data Sets Using Adaptive Mesh Refinement Hierarchies.
IEEE Trans. Vis. Comput. Graph.	Modeling and Rendering of Points with Local Geometry.	Aravind Kalaiah,Amitabh Varshney	2003	We present a novel rendering primitive that combines the modeling brevity of points with the rasterization efficiency of polygons. The surface is represented by a sampled collection of Differential Points (DP), each with embedded curvature information that captures the local differential geometry in the vicinity of that point. This is a more general point representation that, for the cost of a few additional bytes, packs much more information per point than the traditional point-based models. This information is used to efficiently render the surface as a collection of local geometries. To use the hardware acceleration, the DPs are quantized into \big. 256\bigr.different types and each sampled point is approximated by the closest quantized DP and is rendered as a normal-mapped rectangle. The advantages to this representation are: 1) The surface can be represented more sparsely compared to other point primitives, 2) it achieves a robust hardware accelerated per-pixel shading¿even with no connectivity information, and 3) it offers a novel point-based simplification technique that factors in the complexity of the local geometry. The number of primitives being equal, DPs produce a much better quality of rendering than a pure splat-based approach. Visual appearances being similar, DPs are about two times faster and require about 75 percent less disk space in comparison to splatting primitives.
IEEE Trans. Vis. Comput. Graph.	Illustrative Interactive Stipple Rendering.	Aidong Lu,Christopher J. Morris,Joe Taylor,David S. Ebert,Charles D. Hansen,Penny Rheingans,Mark Hartner	2003	Simulating hand-drawn illustration can succinctly express information in a manner that is communicative and informative. We present a framework for an interactive direct stipple rendering of volume and surface-based objects. By combining the principles of artistic and scientific illustration, we explore several feature enhancement techniques to create effective, interactive visualizations of scientific and medical data sets. We also introduce a rendering mechanism that generates appropriate point lists at all resolutions during an automatic preprocess and modifies rendering styles through different combinations of these feature enhancements. The new system is an effective way to interactively preview large, complex volume and surface data sets in a concise, meaningful, and illustrative manner. Stippling is effective for many applications and provides a quick and efficient method to investigate both volume and surface models.
IEEE Trans. Vis. Comput. Graph.	Using Motion to Illustrate Static 3D Shape--Kinetic Visualization.	Eric B. Lum,Aleksander Stompel,Kwan-Liu Ma	2003	In this paper, we present a novel visualization technique¿kinetic visualization¿that uses motion along a surface to aid in the perception of 3D shape and structure of static objects. The method uses particle systems, with rules such that particles flow over the surface of an object to not only bring out, but also attract attention to information on a shape that might not be readily visible with a conventional rendering method which uses lighting and view changes. Replacing still images with animations in this fashion, we demonstrate with both surface and volumetric models in the accompanying videos that, in many cases, the resulting visualizations effectively enhance the perception of three-dimensional shape and structure. We also describe how, for both types of data, a texture-based representation of this motion can be used for interactive visualization using PC graphics hardware. Finally, the results of a user study that we have conducted are presented, which show evidence that the supplemental motion cues can be helpful.
IEEE Trans. Vis. Comput. Graph.	An Architecture for Java-Based Real-Time Distributed Visualization.	Jeffrey Mahovsky,Luigi Benedicenti	2003	In this paper, we present a Java-based software architecture for real-time visualization that utilizes a cluster of conventional PCs to generate high-quality interactive graphics. Normally, a large multiprocessor computer would be needed for interactive visualization tasks requiring more processing power than a single PC can provide. By using clusters of PCs, enormous cost savings can be realized, and proprietary ¿high-end¿ hardware is no longer necessary for these tasks. Our architecture minimizes the amount of synchronization needed between PCs, resulting in excellent scalability. It provides a modular framework that can accommodate a wide variety of rendering algorithms and data formats, provided that the rendering algorithms can generate pixels individually and the data is duplicated on each PC. Demonstration modules that implement ray tracing, fractal rendering, and volume rendering algorithms were developed to evaluate the architecture. Results are encouraging¿using 15 PCs connected to a standard 100 Megabit/s Ethernet network, the system can interactively render simple to moderately complex data sets at modest resolution. Excellent scalability is achieved; however, our tests were limited to a cluster of 15 PCs. Results also demonstrate that Java is a viable platform for real-time distributed visualization.
IEEE Trans. Vis. Comput. Graph.	Transitive Mesh Space of a Progressive Mesh.	Junho Kim,Seungyong Lee	2003	This paper investigates the set of all selectively refined meshes that can be obtained from a progressive mesh. We call the set the transitive mesh space of a progressive mesh and present a theoretical analysis of the space. We define selective edge collapse and vertex split transformations, which we use to traverse all selectively refined meshes in the transitive mesh space. We propose a complete selective refinement scheme for a progressive mesh based on the transformations and compare the scheme with previous selective refinement schemes in both theoretical and experimental ways. In our comparison, we show that the complete scheme always generates selectively refined meshes with smaller numbers of vertices and faces than previous schemes for a given refinement criterion. The concept of dual pieces of the vertices in the vertex hierarchy plays a central role in the analysis of the transitive mesh space and the design of selective edge collapse and vertex split transformations.
IEEE Trans. Vis. Comput. Graph.	Frame-to-Frame Coherent Animation with Two-Pass Radiosity.	Ignacio Martín,Xavier Pueyo,Dani Tost	2003	This paper proposes an efficient method for the production of high quality radiosity solutions which uses an a priori knowledge of the dynamic properties of the scene to exploit temporal coherence. The method is based on a two-pass strategy that provides user-control on the final frame quality. In the first pass, it computes a coarse global solution of the radiosities along a time interval and then, in the second pass, it performs a frame-to-frame incremental gathering step using hardware graphic accelerators. Computing cost is thus reduced because the method takes advantage of frame-to-frame coherence by identifying the changes produced by dynamic objects and by decoupling them from computations that remain unchanged. The input data is a dynamic model of the environment through a period of time corresponding to the same camera recording. The method proceeds by incrementally updating two data structures: a space-time hierarchical radiosity solution for a given interval of time and a hierarchical tree of textures representing the space-time final illumination of the visible surfaces. These data structures are computed for a given viewpoint, either static or dynamic. The main contribution of this work is the efficient construction of the texture tree by identifying the changes produced by dynamic objects and by only recomputing these changes.
IEEE Trans. Vis. Comput. Graph.	"Guest Editor's Introduction: Special Section on Visualization."	Robert J. Moorhead II	2003	"Guest Editor's Introduction: Special Section on Visualization."
IEEE Trans. Vis. Comput. Graph.	On Marching Cubes.	Gregory M. Nielson	2003	On Marching Cubes.
IEEE Trans. Vis. Comput. Graph.	Simplification and Repair of Polygonal Models Using Volumetric Techniques.	Fakir S. Nooruddin,Greg Turk	2003	Two important tools for manipulating polygonal models are simplification and repair and we present voxel-based methods for performing both of these tasks. We describe a method for converting polygonal models to a volumetric representation in a way that handles models with holes, double walls, and intersecting parts. This allows us to perform polygon model repair simply by converting a model to and from the volumetric domain. We also describe a new topology-altering simplification method that is based on 3D morphological operators. Visually unimportant features such as tubes and holes may be eliminated from a model by the open and close morphological operators. Our simplification approach accepts polygonal models as input, scan converts these to create a volumetric description, performs topology modification, and then converts the results back to polygons. We then apply a topology-preserving polygon simplification technique to produce a final model. Our simplification method produces results that are everywhere manifold.
IEEE Trans. Vis. Comput. Graph.	Distributed Virtual Reality Environments Based on Rewriting Systems.	Hansrudi Noser,Christian Stern,Peter Stucki	2003	Ideally, virtual worlds should be dynamic, mutable, and complex in order to be attractive for immersed users. As such worlds can be designed easily by rewriting techniques, we propose a distributed Virtual Reality (VR) system that is based on an interactive animation system using a rewriting technique for geometric and behavioral modeling. The emphasis is on concepts and extensions for the integration of user immersion, user interaction, and networking into a rewriting-based animation system. Finally, the modeling of a ball game with two immersed users, as well as a virtual park, serve as case studies to illustrate the proposed concepts and extensions.
IEEE Trans. Vis. Comput. Graph.	From Visual Data Exploration to Visual Data Mining: A Survey.	Maria Cristina Ferreira de Oliveira,Haim Levkowitz	2003	From Visual Data Exploration to Visual Data Mining: A Survey.
IEEE Trans. Vis. Comput. Graph.	Comparing Simplification and Image-Based Techniques for 3D Client-Server Rendering Systems.	Wouter Pasman,Frederik W. Jansen	2003	A mathematical model is presented for comparing geometric and image-based simplification methods. Geometric simplification reduces the number of polygons in the virtual object and image-based simplification replaces the object with an image. Our model integrates and extrapolates existing accuracy estimates, enabling the comparison of different simplification methods in order to choose the most efficient method in a given situation. The model compares data transfer and rendering load of the methods. Byte size and expected lifetime of simplifications are calculated as a function of the desired visual quality and the position and movement of the viewer. An example result is that, in typical viewing and rendering conditions and for objects with a radius in the order of one meter, imposter techniques can be used at viewing distances above 15 meters. Below that, simplified polygon objects are required and, below one meter distance, the full-resolution virtual object has to be rendered. An electronic version of the model is available on the web.
IEEE Trans. Vis. Comput. Graph.	"Correction to 'Adding Support for High-Level Skeletal Animation'."	Francisco J. Serón,Rafael Rodriguez,Eva Cerezo,Alfredo Pina	2003	"Correction to 'Adding Support for High-Level Skeletal Animation'."
IEEE Trans. Vis. Comput. Graph.	Multiscale Visualization Using Data Cubes.	Chris Stolte,Diane Tang,Pat Hanrahan	2003	Most analysts start with an overview of the data before gradually refining their view to be more focused and detailed. Multiscale pan-and-zoom systems are effective because they directly support this approach. However, generating abstract overviews of large data sets is difficult and most systems take advantage of only one type of abstraction: visual abstraction. Furthermore, these existing systems limit the analyst to a single zooming path on their data and thus to a single set of abstract views. This paper presents: 1) a formalism for describing multiscale visualizations of data cubes with both data and visual abstraction and 2) a method for independently zooming along one or more dimensions by traversing a zoom graph with nodes at different levels of detail. As an example of how to design multiscale visualizations using our system, we describe four design patterns using our formalism. These design patterns show the effectiveness of multiscale visualization of general relational databases.
IEEE Trans. Vis. Comput. Graph.	Dynamic Line Integral Convolution for Visualizing Streamline Evolution.	Andreas Sundquist	2003	Dynamic Line Integral Convolution for Visualizing Streamline Evolution.
IEEE Trans. Vis. Comput. Graph.	Hardware-Based View-Independent Cell Projection.	Manfred Weiler,Martin Kraus,Markus Merz,Thomas Ertl	2003	We present two implementations of a view-independent cell projection algorithm for off-the-shelf programmable graphics hardware. Both implementations perform all computations for the projection and scan conversion of a set of tetrahedra on the graphics hardware and are therefore compatible with many of the hardware-accelerated optimizations for polygonal graphics, e.g., OpenGL vertex arrays and display lists. Apart from our actual implementations, we discuss potential improvements on future, more flexible graphics hardware and applications to interactive volume visualization of unstructured meshes.
IEEE Trans. Vis. Comput. Graph.	Interactive Clipping Techniques for Texture-Based Volume Visualization and Volume Shading.	Daniel Weiskopf,Klaus Engel,Thomas Ertl	2003	Interactive Clipping Techniques for Texture-Based Volume Visualization and Volume Shading.
IEEE Trans. Vis. Comput. Graph.	Global Visualization and Alignments of Whole Bacterial Genomes.	Pak Chung Wong,Kwong-kwok Wong,Harlan Foote,Jim Thomas	2003	Global Visualization and Alignments of Whole Bacterial Genomes.
IEEE Trans. Vis. Comput. Graph.	Natural Motion Animation through Constraining and Deconstraining at Will.	Katsu Yamane,Yoshihiko Nakamura	2003	Natural Motion Animation through Constraining and Deconstraining at Will.
IEEE Trans. Vis. Comput. Graph.	Multilevel Representation and Transmission of Real Objects with Progressive Octree Particles.	Yücel Yemez,Francis Schmitt	2003	We present a multilevel representation scheme adapted to storage, progressive transmission, and rendering of dense data sampled on the surface of real objects. Geometry and object attributes, such as color and normal, are encoded in terms of surface particles associated to a hierarchical space partitioning based on an octree. Appropriate ordering of surface particles results in a compact multilevel representation without increasing the size of the uniresolution model corresponding to the highest level of detail. This compact representation can progressively be decoded by the viewer and transformed by a fast direct triangulation technique into a sequence of triangle meshes with increasing levels of detail. The representation requires approximately 5 bits per particle (2.5 bits per triangle) to encode the basic geometrical structure. The vertex positions can then be refined by means of additional precision bits, resulting in 5 to 9 bits per triangle for representing a 12-bit quantized geometry. The proposed representation scheme is demonstrated with the surface data of various real objects.
IEEE Trans. Vis. Comput. Graph.	Shadows and Soft Shadows with Participating Media Using Splatting.	Caixia Zhang,Roger Crawfis	2003	This paper describes an efficient algorithm to model the light attenuation due to a participating media with low albedo. Here, we consider the light attenuation along a ray, as well as the light attenuation emanating from a surface. The light attenuation is modeled using a splatting volume renderer for both the viewer and the light source. During the rendering, a 2D shadow buffer accumulates the light attenuation. We first summarize the basic shadow algorithm using splatting [30]. Then, an extension of the basic shadow algorithm for projective textured light sources is described. The main part of this paper is an analytic soft shadow algorithm based on convolution techniques. We describe and discuss the soft shadow algorithm, and generate soft shadows, including umbra and penumbra, for extended light sources.
IEEE Trans. Vis. Comput. Graph.	Visualizing Diffusion Tensor MR Images Using Streamtubes and Streamsurfaces.	Song Zhang,Çagatay Demiralp,David H. Laidlaw	2003	We present a new method for visualizing 3D volumetric diffusion tensor MR images. We distinguish between linear anisotropy and planar anisotropy and represent values in the two regimes using streamtubes and streamsurfaces, respectively. Streamtubes represent structures with primarily linear diffusion, typically fiber tracts; streamtube direction correlates with tract orientation. The cross-sectional shape and color of each streamtube represent additional information from the diffusion tensor at each point. Streamsurfaces represent structures in which diffusion is primarily planar. Our algorithm chooses a very small representative subset of the streamtubes and streamsurfaces for display. We describe the set of metrics used for the culling process, which reduces visual clutter and improves interactivity. We also generate anatomical landmarks to identify the locations of such structures as the eyes, skull surface, and ventricles. The final models are complex surface geometries that can be imported into many interactive graphics software environments. We describe a virtual environment to interact with these models. Expert feedback from doctors studying changes in white-matter structures after gamma-knife capsulotomy and preoperative planning for brain tumor surgery shows that streamtubes correlate well with major neural structures, the 2D section and geometric landmarks are important in understanding the visualization, and the stereo and interactivity from the virtual environment aid in understanding the complex geometric models.
CVPR	2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2004), with CD-ROM, 27 June - 2 July 2004, Washington, DC, USA		2004	2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2004), with CD-ROM, 27 June - 2 July 2004, Washington, DC, USA
VAST	3D Face Reconstruction from Skull Aimed to Archaeological Applications. The Site of Murecine: a Case Study.	Andrea F. Abate,Michele Nappi,Stefano Ricciardi,Gabriele Sabatino	2004	3D Face Reconstruction from Skull Aimed to Archaeological Applications. The Site of Murecine: a Case Study.
VAST	Digital representation and multimodal presentation of archeological graffiti at Pompei.	Marcello Balzani,Marco Callieri,M. Fabbri,Andrea Fasano,Claudio Montani,Paolo Pingi,N. Santopuoli,Roberto Scopigno,Federico Uccelli,Alberto Varone	2004	Digital representation and multimodal presentation of archeological graffiti at Pompei.
VAST	Digital reconstruction of the Arrigo VII funerary complex.	Clara Baracchini,Antonio Brogi,Marco Callieri,Lucia Capitani,Paolo Cignoni,Andrea Fasano,Claudio Montani,C. Nenci,R. P. Novello,Paolo Pingi,Federico Ponchio,Roberto Scopigno	2004	Digital reconstruction of the Arrigo VII funerary complex.
VAST	Image-Based Registration of 3D-Range Data Using Feature Surface Elements.	Gerhard H. Bendels,Patrick Degener,Roland Wahl,Marcel Körtgen,Reinhard Klein	2004	Image-Based Registration of 3D-Range Data Using Feature Surface Elements.
VAST	"A Novel Approach to Context-Sensitive Guided e-Tours in Cultural Sites: ""Light"" Augmented Reality on PDAs."	Vassilios Vlahakis,Athanasios M. Demiris,E. Bounos,Nikolaos Ioannidis	2004	"A Novel Approach to Context-Sensitive Guided e-Tours in Cultural Sites: ""Light"" Augmented Reality on PDAs."
VAST	The Shotton River and Mesolithic Dwellings: Recreating the Past from Geo-Seismic Data Sources.	"Eugene Ch'ng,Robert J. Stone,Theodoros N. Arvanitis"	2004	The Shotton River and Mesolithic Dwellings: Recreating the Past from Geo-Seismic Data Sources.
VAST	A case study of a virtual audience in a reconstruction of an ancient Roman odeon in Aphrodisias.	Pablo de Heras Ciechomski,Branislav Ulicny,Rachel Cetre,Daniel Thalmann	2004	A case study of a virtual audience in a reconstruction of an ancient Roman odeon in Aphrodisias.
VAST	iClay: Digitizing Cuneiform.	Jonathan D. Cohen,Donald Duncan,Dean Snyder,Jerrold Cooper,Subodh Kumar,Daniel Hahn,Yuan Chen,Budirijanto Purnomo,John Graettinger	2004	iClay: Digitizing Cuneiform.
VAST	A Cultural Heritage Repository as Source for Learning Materials.	Manjula Patel,Krzysztof Walczak,Fabrizio Giorgini,Martin White	2004	A Cultural Heritage Repository as Source for Learning Materials.
VAST	Telling the Local Story: An Interactive Cultural Presentation System for Community and Regional Settings.	Daniel Pletinckx,Lars De Jaegher,Truus Helsen,Iris Langen,Neil A. Silberman,Marie-Claire Van der Donckt,Jan Stobbe	2004	Telling the Local Story: An Interactive Cultural Presentation System for Community and Regional Settings.
VAST	New Perspectives on Ancient Landscapes: A Case Study of the Foulness Valley.	Julien Pansiot,Paul M. Chapman,Warren J. Viant,Peter Halkon	2004	New Perspectives on Ancient Landscapes: A Case Study of the Foulness Valley.
VAST	A Point-Based Approach for Capture, Display and Illustration of Very Complex Archeological Artefacts.	Florent Duguet,George Drettakis,Daniel Girardeau-Montaut,Jean-Luc Martinez,Francis Schmitt	2004	A Point-Based Approach for Capture, Display and Illustration of Very Complex Archeological Artefacts.
VAST	The Ancient Olympic Games: Being Part of the Experience.	Athanasios Gaitatzes,Dimitrios Christopoulos,Georgios Papaioannou	2004	The Ancient Olympic Games: Being Part of the Experience.
VAST	Generative Parametric Design of Gothic Window Tracery.	Sven Havemann,Dieter W. Fellner	2004	Generative Parametric Design of Gothic Window Tracery.
VAST	As Time Flies By: Mixed Image and Model-Based Rendering of an Historical Landscape from Helicopter Images.	Maarten Vergauwen,Daniel Pletinckx,Geert Willems,Frank Verbiest,Luc J. Van Gool,Truus Helsen	2004	As Time Flies By: Mixed Image and Model-Based Rendering of an Historical Landscape from Helicopter Images.
VAST	Interactive Cloth Editing and Simulation in Virtual Reality Applications for Theater Professionals.	Dimitrios Koutsonanos,Konstantinos Moustakas,Dimitrios Tzovaras,Michael G. Strintzis	2004	Interactive Cloth Editing and Simulation in Virtual Reality Applications for Theater Professionals.
VAST	The potential of internet-based techniques for heritage interpretation.	W. Kreisel,K. Gee,F. Dickmann	2004	The potential of internet-based techniques for heritage interpretation.
VAST	Virtual San Storytelling for Children: Content vs. Experience.	Ilda Ladeira,Edwin H. Blake	2004	Virtual San Storytelling for Children: Content vs. Experience.
VAST	3D orientation of archaeological fragments coming from a Gothic spire.	C. Laugerotte,Nadine Warzée	2004	3D orientation of archaeological fragments coming from a Gothic spire.
VAST	An Interactive Visualisation Interface for Virtual Museums.	Fotis Liarokapis,Stella Sylaiou,Anirban Basu,Nicholaos Mourkoussis,Martin White,Paul F. Lister	2004	An Interactive Visualisation Interface for Virtual Museums.
VAST	An Authoring Tool for Interactive Digital Storytelling.	María Teresa Linaza,Hector Eskudero,Carlos Lamsfus,Gorka Marcos	2004	An Authoring Tool for Interactive Digital Storytelling.
VAST	The Museum of Pure Form: touching real statues in an immersive virtual museum.	Céline Loscos,Franco Tecchia,Antonio Frisoli,Marcello Carrozzino,Hila Ritter Widenfeld,David Swapp,Massimo Bergamasco	2004	The Museum of Pure Form: touching real statues in an immersive virtual museum.
VAST	Simulation of Ancient TechnologyWorks Using Haptic Interaction and Gesture Recognition.	Georgios Nikolakis,Dimitrios Tzovaras,Sotiris Malassiotis,Michael G. Strintzis	2004	Simulation of Ancient TechnologyWorks Using Haptic Interaction and Gesture Recognition.
VAST	Developing the Tourism Aspects of a Cultural Route.	Ruth Owen,Dimitrios Buhalis,Daniel Pletinckx	2004	Developing the Tourism Aspects of a Cultural Route.
VAST	Identifying technologies used in Cultural Heritage.	Ruth Owen,Dimitrios Buhalis,Daniel Pletinckx	2004	Identifying technologies used in Cultural Heritage.
VAST	Generation of High-Resolution Mosaic for Photo-Realistic Texture-Mapping of Cultural Heritage 3D Models.	Fabio Remondino,Jana Niederoest	2004	Generation of High-Resolution Mosaic for Photo-Realistic Texture-Mapping of Cultural Heritage 3D Models.
VAST	Adaptive Crowd Behaviour to Aid Real-Time Rendering of a Cultural Heritage Environment.	G. Ryder,P. A. Flack,A. M. Day	2004	Adaptive Crowd Behaviour to Aid Real-Time Rendering of a Cultural Heritage Environment.
VAST	An environment for the analysis and reconstruction of archaeological objects.	C. Laugerotte,Nadine Warzée	2004	An environment for the analysis and reconstruction of archaeological objects.
VAST	VAST 2004: The 5th International Symposium on Virtual Reality, Archaeology and Intelligent Cultural Heritage, incorporating 2nd Eurographics Workshop on Graphics and Cultural Heritage, Conscience-Auditorium, Brussels and Ename Center, Oudenaarde, Belgium, December 7-10, 2004	Dieter W. Fellner,Stephen N. Spencer	2004	VAST 2004: The 5th International Symposium on Virtual Reality, Archaeology and Intelligent Cultural Heritage, incorporating 2nd Eurographics Workshop on Graphics and Cultural Heritage, Conscience-Auditorium, Brussels and Ename Center, Oudenaarde, Belgium, December 7-10, 2004
IEEE Visualization	"Exploration of the Brain's White Matter Pathways with Dynamic Queries."	David Akers,Anthony Sherbondy,Rachel Mackenzie,Robert Dougherty,Brian A. Wandell	2004	Diffusion Tensor Imaging (DTI) is a magnetic resonance imaging method that can be used to measure local information about the structure of white matter within the human brain. Combining DTI data with the computational methods of MR tractography, neuroscientists can estimate the locations and sizes of nerve bundles (white matter pathways) that course through the human brain. Neuroscientists have used visualization techniques to better understand tractography data, but they often struggle with the abundance and complexity of the pathways. In this paper, we describe a novel set of interaction techniques that make it easier to explore and interpret such pathways. Specifically, our application allows neuroscientists to place and interactively manipulate box-shaped regions (or volumes of interest) to selectively display pathways that pass through specific anatomical areas. A simple and flexible query language allows for arbitrary combinations of these queries using Boolean logic operators. Queries can be further restricted by numerical path properties such as length, mean fractional anisotropy, and mean curvature. By precomputing the pathways and their statistical properties, we obtain the speed necessary for interactive question-andanswer sessions with brain researchers. We survey some questions that researchers have been asking about tractography data and show how our system can be used to answer these questions efficiently.
IEEE Visualization	Visualizing Botanical Trees over Four Seasons.	Derek Bradley	2004	Visualizing Botanical Trees over Four Seasons.
IEEE Visualization	Centroidal Voronoi Tessellation Based Algorithms for Vector Fields Visualization and Segmentation.	Qiang Du,Xiaoqiang Wang	2004	A new method for the simplification and the visualization of vector fields is presented based on the notion of Centroidal Voronoi tessellations (CVTýs). A CVT is a special Voronoi tessellation for which the generators of the Voronoi regions in the tessellation are also the centers of mass (or means) with respect to a prescribed density. A distance function in both the spatial and vector spaces is introduced to measure the similarity of the spatially distributed vector fields. Based on such a distance, vector fields are naturally clustered and their simplified representations are obtained. Our method combines simple geometric intuitions with the rigorously established optimality properties of the CVTs. It is simple to describe, easy to understand and implement. Numerical examples are also provided to illustrate the effectiveness and competitiveness of the CVT-based vector simplification and visualization methodology.
IEEE Visualization	Rendering Planar Cuts Through Quadratic and Cubic Finite Elements.	Michael Brasher,Robert Haimes	2004	Coloring higher order scientific data is problematic using standard linear methods as found in OpenGL. The visual results are inaccurate when there is a large scalar gradient over an element or when the scalar field is nonlinear. In addition to shading nonlinear data, fast and accurate rendering of planar cuts through parametric elements can be implemented using programmable shaders on current graphics hardware. The intersection of a planar cut with geometrically curved volume elements can be rendered using a combination of selective refinement and programmable shaders. This hybrid algorithm also handles curved 2D planar triangles.
IEEE Visualization	Visualization in Grid Computing Environments.	Ken Brodlie,David A. Duce,Julian R. Gallop,Musbah Shahop Sagar,Jeremy Walton,Jason Wood	2004	Grid computing provides a challenge for visualization system designers. In this research, we evolve the dataflow concept to allow parts of the visualization process to be executed remotely in a secure and seamless manner. We see dataflow at three levels: an abstract specification of the intent of the visualization; a binding of these abstract modules to a specific software system; and then a binding of software to processing and other resources. We develop an XML application capable of describing visualization at the three levels. To complement this, we have implemented an extension to a popular visualization system, IRIS Explorer, which allows modules in a dataflow pipeline to run on a set of Grid resources. For computational steering applications, we have developed a library that allows a visualization system front-end to connect to a simulation running remotely on a Grid resource. We demonstrate the work in two applications: the dispersion of a pollutant under different wind conditions; and the solution of a challenging numerical problem in elastohydrodynamic lubrication
IEEE Visualization	Depth Enhanced Panoramas.	Gleb Bahmutov,Voicu Popescu,Elisha Sacks	2004	Depth Enhanced Panoramas.
IEEE Visualization	TexMol: Interactive Visual Exploration of Large Flexible Multi-Component Molecular Complexes.	Chandrajit L. Bajaj,Peter Djeu,Vinay Siddavanahalli,Anthony Thane	2004	While molecular visualization software has advanced over the years, today, most tools still operate on individual molecular structures with limited facility to manipulate large multi-component complexes. We approach this problem by extending 3D image-based rendering via programmable graphics units, resulting in an order of magnitude speedup over traditional triangle-based rendering. By incorporating a biochemically sensitive level-of-detail hierarchy into our molecular representation, we communicate appropriate volume occupancy and shape while dramatically reducing the visual clutter that normally inhibits higher-level spatial comprehension. Our hierarchical, image based rendering also allows dynamically computed physical properties data (e.g. electrostatics potential) to be mapped onto the molecular surface, tying molecular structure to molecular function. Finally, we present another approach to interactive molecular exploration using volumetric and structural rendering in tandem to discover molecular properties that neither rendering mode alone could reveal. These visualization techniques are realized in a high-performance, interactive molecular exploration tool we call TexMol, short for Texture Molecular viewer.
IEEE Visualization	Hierarchy Based 3D Visualization of Large Software Structures.	Michael Balzer,Oliver Deussen	2004	Modern object-oriented programs are hierarchical systems with many thousands of interrelated subsystems. Visualization helps developers to better comprehend these large and complex systems. This work presents a three-dimensional visualization technique that represents the static structure of object-oriented software using distributions of three-dimensional objects on a two-dimensional plane. The visual complexity is reduced by adjusting the transparency of object surfaces to the distance of the viewpoint. An approach called Hierarchical Net is proposed for a clear representation of the relationships between the subsystems.
IEEE Visualization	2D Maps for Visual Analysis and Retrieval in Large Multi-Feature 3D Model Databases.	Benjamin Bustos,Daniel A. Keim,Christian Panse,Tobias Schreck	2004	Multimedia objects are often described by high-dimensional feature vectors which can be used for retrieval and clustering tasks. We have built an interactive retrieval system for 3D model databases that implements a variety of different feature transforms. Recently, we have enhanced the functionality of our system by integrating a SOM-based visualization module. In this poster demo, we show how 2D maps can be used to improve the effectiveness of retrieval, clustering, and over-viewing tasks in a 3D multimedia system.
IEEE Visualization	Live Range Visibility Constraints for Adaptive Terrain Visualization.	Xiaohong Bao,Renato Pajarola,Michael Shafae	2004	Live Range Visibility Constraints for Adaptive Terrain Visualization.
IEEE Visualization	Simplifying Flexible Isosurfaces Using Local Geometric Measures.	Hamish Carr,Jack Snoeyink,Michiel van de Panne	2004	The contour tree, an abstraction of a scalar field that encodes the nesting relationships of isosurfaces, can be used to accelerate isosurface extraction, to identify important isovalues for volume-rendering transfer functions, and to guide exploratory visualization through a flexible isosurface interface. Many real-world data sets produce unmanageably large contour trees which require meaningful simplification. We define local geometric measures for individual contours, such as surface area and contained volume, and provide an algorithm to compute these measures in a contour tree. We then use these geometric measures to simplify the contour trees, suppressing minor topological features of the data. We combine this with a flexible isosurface interface to allow users to explore individual contours of a dataset interactively.
IEEE Visualization	Volume Refinement Fairing Isosurfaces.	Martin Bertram	2004	We propose an interpolating refinement method for two- and three-dimensional scalar fields defined on hexahedral grids. Iterative fairing of the underlying contours (isosurfaces) provides the function values of new grid points. Our method can be considered as a non-linear variational subdivision scheme for volumes. It can be applied locally for adaptive mesh refinement in regions of high geometric complexity. We use our scheme to increase the quality of low-resolution data sets and to reduce interpolation artifacts in texture-based volume rendering.
IEEE Visualization	On the Visualization of Time-Varying Structured Grids Using a 3D Warp Texture.	Yuan Chen,Jonathan D. Cohen,Subodh Kumar	2004	We present a novel scheme to interactively visualize time-varying scalar fields defined on a structured grid. The underlying approach is to maximize the use of current graphics hardware by using 3D texture mapping. This approach commonly suffers from an expensive voxelization of each time-step as well as from large size of the voxel array approximating each step. Hence, in our scheme, instead of explicitly voxelizing each scalar field, we directly store each time-step as a three dimensional texture in its native form. We create the function that warps a voxel grid into the given structured grid. At rendering time, we reconstruct the function at each pixel using hardware-based trilinear interpolation. The resulting coordinates allow us to compute the scalar value at this pixel using a second texture lookup. For fixed grids, the function remains constant across time-steps and only the scalar field table needs to be re-loaded as a texture. Our new approach achieves excellent performance with relatively low texture memory requirements and low approximation error.
IEEE Visualization	Hardware-Accelerated Adaptive EWA Volume Splatting.	Wei Chen,Liu Ren,Matthias Zwicker,Hanspeter Pfister	2004	We present a hardware-accelerated adaptive EWA (elliptical weighted average) volume splatting algorithm. EWA splatting combines a Gaussian reconstruction kernel with a low-pass image filter for high image quality without aliasing artifacts or excessive blurring. We introduce a novel adaptive filtering scheme to reduce the computational cost of EWA splatting. We show how this algorithm can be efficiently implemented on modern graphics processing units (GPUs). Our implementation includes interactive classification and fast lighting. To accelerate the rendering we store splat geometry and 3D volume data locally in GPU memory. We present results for several rectilinear volume datasets that demonstrate the high image quality and interactive rendering speed of our method.
IEEE Visualization	Guaranteed Quality Triangulation of Molecular Skin Surfaces.	Ho-Lun Cheng,Xinwei Shi	2004	We present an efficient algorithm to mesh the macromolecules surface model represented by the skin surface defined by Edelsbrunner. Our algorithm overcomes several challenges residing in current surface meshing methods. First, we guarantee the mesh quality with a provable lower bound of 21. on its minimum angle. Second, we ensure the triangulation is homeomorphic to the original surface. Third, we improve the efficiency of constructing the Restricted Delaunay Triangulation(RDT) of smooth surfaces. We achieve this by constructing the RDT using the advancing front method without computing the Delaunay tetrahedrization of the sample points on the surfaces. The difficulty of handling the front collision problem is tackled by employing the Morse theory. In particular, we construct the Morse-Smale complex to simplify the topological changes of the front. Our implementation results suggest that the algorithm decrease the time of generating high quality homeomorphic skin mesh from hours to a few minutes.
IEEE Visualization	Local and Global Comparison of Continuous Functions.	Herbert Edelsbrunner,John Harer,Vijay Natarajan,Valerio Pascucci	2004	We introduce local and global comparison measures for a collection of k ¿ d real-valued smooth functions on a common d-dimensional Riemannian manifold. For k = d = 2 we relate the measures to the set of critical points of one function restricted to the level sets of the other. The definition of the measures extends to piecewise linear functions for which they are easy to compute. The computation of the measures forms the centerpiece of a software tool which we use to study scientific datasets.
IEEE Visualization	Visualizing Gyrokinetic Simulations.	David Crawford,Kwan-Liu Ma,Min-Yu Huang,Scott Klasky,Stéphane Ethier	2004	The continuing advancement of plasma science is central to realizing fusion as an inexpensive and safe energy source. Gryokinetic simulations of plasmas are fundamental to the understanding of turbulent transport in fusion plasma. This paper discusses the visualization challenges presented by gyrokinetic simulations using magnetic field line following coordinates, and presents an effective solution exploiting programmable graphics hardware to enable interactive volume visualization of 3D plasma flow on a toroidal coordinate system. The new visualization capability can help scientists better understand three-dimensional structures of the modeled phenomena. Both the limitations and future promise of the hardware-accelerated approach are also discussed.
IEEE Visualization	Interactive Terascale Particle Visualization.	David Ellsworth,Bryan Green,Patrick J. Moran	2004	This paper describes the methods used to produce an interactive visualization of a 2 TB computational fluid dynamics (CFD) data set using particle tracing (streaklines). We use the method introduced by Bruckschen et al. [3] that precomputes a large number of particles, stores them on disk using a space-filling curve ordering that minimizes seeks, then retrieves and displays the particles according to the userýs command. We describe how the particle computation can be performed using a PC cluster, how the algorithm can be adapted to work with a multi-block curvilinear mesh, how scalars can be extracted and used to color the particles, and how the out-of-core visualization can be scaled to 293 billion particles while still achieving interactive performance on PC hardware. Compared to the earlier work, our data set size and total number of particles are an order of magnitude larger. We also describe a new compression technique that losslessly reduces the amount of particle storage by 41% and speeds the particle retrieval by about 20%.
IEEE Visualization	Visualization of Salt-Induced Stress Perturbations.	Patricia Crossno,David H. Rogers,Rebecca M. Brannon,David Coblentz	2004	An important challenge encountered during post-processing of finite element analyses is the visualizing of three-dimensional fields of real-valued second-order tensors. Namely, as finite element meshes become more complex and detailed, evaluation and presentation of the principal stresses becomes correspondingly problematic. In this paper, we describe techniques used to visualize simulations of perturbed in-situ stress fields associated with hypothetical salt bodies in the Gulf of Mexico. We present an adaptation of the Mohr diagram, a graphical paper and pencil method used by the material mechanics community for estimating coordinate transformations for stress tensors, as a new tensor glyph for dynamically exploring tensor variables within three-dimensional finite element models. This interactive glyph can be used as either a probe or a filter through brushing and linking.
IEEE Visualization	Non-Linear Model Fitting to Parameterize Diseased Blood Vessels.	Alexandra La Cruz,Matús Straka,Arnold Köchl,Milos Srámek,Eduard Gröller,Dominik Fleischmann	2004	Accurate estimation of vessel parameters is a prerequisite for automated visualization and analysis of healthy and diseased blood vessels. The objective of this research is to estimate the dimensions of lower extremity arteries, imaged by computed tomography (CT). These parameters are required to get a good quality visualization of healthy as well as diseased arteries using a visualization technique such as curved planar reformation (CPR). The vessel is modeled using an elliptical or cylindrical structure with specific dimensions, orientation and blood vessel mean density. The model separates two homogeneous regions: Its inner side represents a region of density for vessels, and its outer side a region for background. Taking into account the point spread function (PSF) of a CT scanner, a function is modeled with a Gaussian kernel, in order to smooth the vessel boundary in the model. A new strategy for vessel parameter estimation is presented. It stems from vessel model and model parameter optimization by a nonlinear optimization procedure, i.e., the Levenberg-Marquardt technique. The method provides center location, diameter and orientation of the vessel as well as blood and background mean density values. The method is tested on synthetic data and real patient data with encouraging results.
IEEE Visualization	Linear and Cubic Box Splines for the Body Centered Cubic Lattice.	Alireza Entezari,Ramsay Dyer,Torsten Möller	2004	In this paper we derive piecewise linear and piecewise cubic box spline reconstruction filters for data sampled on the body centered cubic (BCC) lattice. We analytically derive a time domain representation of these reconstruction filters and using the Fourier slice-projection theorem we derive their frequency responses. The quality of these filters, when used in reconstructing BCC sampled volumetric data, is discussed and is demonstrated with a raycaster. Moreover, to demonstrate the superiority of the BCC sampling, the resulting reconstructions are compared with those produced from similar filters applied to data sampled on the Cartesian lattice.
IEEE Visualization	Context-Adaptive Mobile Visualization and Information Management.	Jochen Ehret,Achim Ebert,Lars Schuchardt,Heidrun Steinmetz,Hans Hagen	2004	This poster abstract presents a scalable information visualization system for mobile devices and desktop systems. It is designed to support the operation and the workflow of wastewater systems. The regarded information data includes general information about buildings and units, process data, occupational safety regulations, work directions and first aid instructions in case of an accident. Technically, the presented framework combines visualization with agent technology in order to automatically scale various visualization types to fit on different platforms like PDAs (Personal Digital Assistants) or Tablet PCs. The implementation is based on but not limited to SQL, JSP, HTML and VRML.
IEEE Visualization	Linking Representation with Meaning.	David J. Duke	2004	The purpose of visualization is not just to depict data, but to gain or present insight into the domain represented in data. However in visualization systems, this link between features in the data and the meaning of those features is often missing or implicit. It is assumed that the user, through looking at the output, will close the loop between representation and insight. An alternative is to view visualization tools as interfaces between data and insight, and to enrich this interface with capabilities linked to usersý conceptual models of the data. Preliminary work has been carried out to develop such an interface as a modular component that can be installed in a pipelined architecture. This poster expands the motivation for this work, and describes the initial implementation carried out within the Visualization Toolkit (VTK).
IEEE Visualization	Building an Ontology of Visualization.	David J. Duke,Ken W. Brodlie,David A. Duce	2004	Recent activity within the UK National e-Science Programme has identified a need to establish an ontology for visualization. Motivation for this includes defining web and grid services for visualization (the ýsemantic gridý), supporting collaborative work, curation, and underpinning visualization research and education. At a preliminary meeting, members of the UK visualization community identified a skeleton for the ontology. We have started to build on this by identifying how existing work might be related and utilized. We believe that the greatest challenge is reaching a consensus within the visualization community itself. This poster is intended as one step in this process, setting out the perceived needs for the ontology, and sketching initial directions. It is hoped that this will lead to debate, feedback and involvement across the community.
IEEE Visualization	Panel 2: In the Eye of the Beholder: The Role of Perception in Scientific Visualization.	Kelly P. Gaither,David S. Ebert,Bill Geisler,David H. Laidlaw	2004	Panel 2: In the Eye of the Beholder: The Role of Perception in Scientific Visualization.
IEEE Visualization	Visibility Culling for Time-Varying Volume Rendering Using Temporal Occlusion Coherence.	Jinzhu Gao,Han-Wei Shen,Jian Huang,James Arthur Kohl	2004	"Typically there is a high coherence in data values between neighboring time steps in an iterative scientific software simulation; this characteristic similarly contributes to a corresponding coherence in the visibility of volume blocks when these consecutive time steps are rendered. Yet traditional visibility culling algorithms were mainly designed for static data, without consideration of such potential temporal coherency. In this paper, we explore the use of Temporal Occlusion Coherence (TOC) to accelerate visibility culling for time-varying volume rendering. In our algorithm, the opacity of volume blocks is encoded by means of Plenoptic Opacity Functions (POFs). A coherence-based block fusion technique is employed to coalesce time-coherent data blocks over a span of time steps into a single, representative block. Then POFs need only be computed for these representative blocks. To quickly determine the subvolumes that do not require updates in their visibility status for each subsequent time step, a hierarchical ""TOC tree"" data structure is constructed to store the spans of coherent time steps. To achieve maximal culling potential, while remaining conservative, we have extended our previous POF into an Optimized POF (OPOF) encoding scheme for this specific scenario. To test our general TOC and OPOF approach, we have designed a parallel time-varying volume rendering algorithm accelerated by visibility culling. Results from experimental runs on a 32-processor cluster confirm both the effectiveness and scalability of our approach."
IEEE Visualization	Tracking of Vector Field Singularities in Unstructured 3D Time-Dependent Datasets.	Christoph Garth,Xavier Tricoche,Gerik Scheuermann	2004	In this paper, we present an approach for monitoring the positions of vector field singularities and related structural changes in time-dependent datasets. The concept of singularity index is discussed and extended from the well-understood planar case to the more intricate three-dimensional setting. Assuming a tetrahedral grid with linear interpolation in space and time, vector field singularities obey rules imposed by fundamental invariants (Poincaré index), which we use as a basis for an efficient tracking algorithm. We apply the presented algorithm to CFD datasets to illustrate its purpose. We examine structures that exhibit topological variations with time and describe some of the insight gained with our method. Examples are given that show a correlation in the evolution of physical quantities that play a role in vortex breakdown.
IEEE Visualization	Keynote Address: The Human Visual System: How Is Its Design Related to the Physics of the Natural Environment?	Wilson S. Geisler	2004	Keynote Address: The Human Visual System: How Is Its Design Related to the Physics of the Natural Environment?
IEEE Visualization	Panel 4: What Should We Teach in a Scientific Visualization Class?	Jon D. Genetti,Michael J. Bailey,David H. Laidlaw,Robert J. Moorhead,Ross T. Whitaker	2004	Panel 4: What Should We Teach in a Scientific Visualization Class?
IEEE Visualization	Augmented Reality with Tangible Auto-Fabricated Models for Molecular Biology Applications.	Alexandre Gillet,Michel F. Sanner,Daniel Stoffler,David S. Goodsell,Arthur J. Olson	2004	"The evolving technology of computer auto-fabrication (""3-D printing"") now makes it possible to produce physical models for complex biological molecules and assemblies. We report on an application that demonstrates the use of auto-fabricated tangible models and augmented reality for research and education in molecular biology, and for enhancing the scientific environment for collaboration and exploration. We have adapted an augmented reality system to allows virtual 3-D representations (generated by the Python Molecular Viewer) to be overlaid onto a tangible molecular model. Users can easily change the overlaid information, switching between different representations of the molecule, displays of molecular properties such as electrostatics, or dynamic information. The physical model provides a powerful, intuitive interface for manipulating the computer models, streamlining the interface between human intent, the physical model, and the computational activity."
IEEE Visualization	Capillary Histology Imagery Visualization and Exploration.	Michael Gleicher,Tom Brunet,K. Evan Nowak,Liz Osten,Matt McElwee,Kevin Tanty,Adam Gepner,Garet Lahvis	2004	Capillary Histology Imagery Visualization and Exploration.
IEEE Visualization	Flow Field Clustering via Algebraic Multigrid.	Michael Griebel,Tobias Preußer,Martin Rumpf,Marc Alexander Schweitzer,Alexandru Telea	2004	We present a novel multiscale approach for flow visualization. We define a local alignment tensor that encodes a measure for alignment to the direction of a given flow field. This tensor induces an anisotropic differential operator on the flow domain, which is discretized with a standard finite element technique. The entries of the corresponding stiffness matrix represent the anisotropically weighted couplings of adjacent nodes of the domain mesh. We use an algebraic multigrid algorithm to generate a hierarchy of fine to coarse descriptions for the above coupling data. This hierarchy comprises a set of coarse grid nodes, a multiscale of basis functions and their corresponding supports. We use these supports to obtain a multilevel decomposition of the flow structure. Standard streamline icons are used to visualize this decomposition at any user-selected level of detail. The method provides a single framework for vector field decomposition independent on the domain dimension or mesh type. Applications are shown in 2D, for flow fields on curved surfaces, and for 3D volumetric flow fields.
IEEE Visualization	Visualizing the Energetics of the Dissociation of a Metastable Molecule.	David Guzman,Reuben Reyes,Karla Vega,Kelly P. Gaither,Robert Wyatt	2004	Visualizing the Energetics of the Dissociation of a Metastable Molecule.
IEEE Visualization	Visual Inspection Methods for Quality Control in Automotive Engineering.	Hans Hagen,Andreas Disch,Jochen Ehret,Ralf Klein,Sascha Köhn,Dirk Zeckzer,Michael Münchhofen	2004	The automotive industry demands visual support for the verification of the quality of their products from the design phase to the manufacturing phase. This implies the need of tools for measurement planning, programming measuring devices, managing measurement data, and the visual exploration of the measurement results. To improve the quality control throughout the whole process chain an integration of such tools in a platform independent framework is crucial. We present eMMA (enhanced Measure Management Application), a client/server system integrating measurement planning, data management, and straightforward as well as sophisticated visual exploration tools in a single framework.
IEEE Visualization	DaMI - Data Management for Multimedial Information Systems.	Hans Hagen,Gerhard Steinebach,Michael Münchhofen,Inga Scheler,Maja Ruby,Michael Wadlé	2004	This paper describes a Data Management System for Multimedial Information Visualization called DaMI. It is possible to create 2D or 3D model based on data out of standard databases and additional metainformation. DaMI is a generic system guaranteeing an optimal reusability and compatibility.
IEEE Visualization	Capstone Address: Self-Illustrating Phenomena.	Pat Hanrahan	2004	Capstone Address: Self-Illustrating Phenomena.
IEEE Visualization	VisBiz: A Simplified Visualization of Business Operation.	Ming C. Hao,Daniel A. Keim,Umeshwar Dayal	2004	VisBiz: A Simplified Visualization of Business Operation.
IEEE Visualization	PQuad: Visualization of Predicted Peptides and Proteins.	Susan L. Havre,Mudita Singhal,Deborah A. Payne,Bobbie-Jo M. Webb-Robertson	2004	New high-throughput proteomic techniques generate data faster than biologists can analyze it. Hidden within this massive and complex data are answers to basic questions about how cells function. The data afford an opportunity to take a global or systems approach studying whole proteomes comprising all the proteins in an organism. However, the tremendous size and complexity of the high-throughput data make it difficult to process and interpret. Existing tools for studying a few proteins at a time are not suitable for global analysis. Visualization provides powerful analysis capabilities for enormous, complex data at multiple resolutions. We developed a novel interactive visualization tool, PQuad, for the visual analysis of proteins and peptides identified from high-throughput data on biological samples. PQuad depicts the peptides in the context of their source protein and DNA, thereby integrating proteomic and genomic information. A wrapped line metaphor is applied across key resolutions of the data, from a compressed view of an entire chromosome to the actual nucleotide sequence. PQuad provides a difference visualization for comparing peptides from samples prepared under different experimental conditions. We describe the requirements for such a visual analysis tool, the design decisions, and the novel aspects of PQuad.
IEEE Visualization	Physically Based Methods for Tensor Field Visualization.	Ingrid Hotz,Louis Feng,Hans Hagen,Bernd Hamann,Kenneth I. Joy,Boris Jeremic	2004	"The physical interpretation of mathematical features of tensor fields is highly application-specific. Existing visualization methods for tensor fields only cover a fraction of the broad application areas. We present a visualization method tailored specifically to the class of tensor field exhibiting properties similar to stress and strain tensors, which are commonly encountered in geomechanics. Our technique is a global method that represents the physical meaning of these tensor fields with their central features: regions of compression or expansion. The method is based on two steps: first, we define a positive definite metric, with the same topological structure as the tensor field; second, we visualize the resulting metric. The eigenvector fields are represented using a texture-based approach resembling line integral convolution (LIC) methods. The eigenvalues of the metric are encoded in free parameters of the texture definition. Our method supports an intuitive distinction between positive and negative eigenvalues. We have applied our method to synthetic and some standard data sets, and ""real"" data from Earth science and mechanical engineering application."
IEEE Visualization	Visualization of Vortices in Simulated Airflow around Bat Wings During Flight.	Eduardo Hueso,Igor Pivkin,Sharon Swartz,David H. Laidlaw,George E. Karniadakis,Kenny Breuer	2004	Visualization of Vortices in Simulated Airflow around Bat Wings During Flight.
IEEE Visualization	Visualizing Competitive Behaviors in Multi-User Virtual Environments.	Nate Hoobler,Greg Humphreys,Maneesh Agrawala	2004	We present a system for enhancing observation of user interactions in virtual environments. In particular, we focus on analyzing behavior patterns in the popular team-based first-person perspective game Return to Castle Wolfenstein: Enemy Territory. This game belongs to a genre characterized by two moderate-sized teams (usually 6 to 12 players each) competing over a set of objectives. Our system allows spectators to visualize global features such as large-scale behaviors and team strategies, as opposed to the limited, local view that traditional spectating modes provide. We also add overlay visualizations of semantic information related to the action that might be important to a spectator in order to reduce the information overload that plagues traditional overview visualizations. These overlays can visualize information about abstract concepts such as player distribution over time and areas of intense combat activity, and also highlight important features like player paths, fire coverage, etc. This added information allows spectators to identify important game events more easily and reveals large-scale player behaviors that might otherwise be overlooked.
IEEE Visualization	Adaptive 4-8 Texture Hierarchies.	Lok M. Hwa,Mark A. Duchaineau,Kenneth I. Joy	2004	We address the texture level-of-detail problem for extremely large surfaces such as terrain during realtime, view-dependent rendering. A novel texture hierarchy is introduced based on 4-8 refinement of raster tiles, in which the texture grids in effect rotate 45 degrees for each level of refinement. This hierarchy provides twice as many levels of detail as conventional quadtree-style refinement schemes such as mipmaps, and thus provides per-pixel view-dependent filtering that is twice as close to the ideal cutoff frequency for an average pixel. Because of this more gradual change in low-pass filtering, and due to the more precise emulation of the ideal cutoff frequency, we find in practice that the transitions between texture levels of detail are not perceptible. This allows rendering systems to avoid the complexity and performance costs of per-pixel blending between texture levels of detail. The 4-8 texturing scheme is integrated into a variant of the Real-time Optimally Adapting Meshes (ROAM) algorithm for viewdependent multiresolution mesh generation. Improvements to ROAM included here are: the diamond data structure as a stream-lined replacement for the triangle bintree elements, the use of low-pass- filtered geometry patches in place of individual triangles, integration of 4-8 textures, and a simple out-of-core data access mechanism for texture and geometry tiles.
IEEE Visualization	ImageSurfer: A Tool for Visualizing Correlations between Two Volume Scalar Fields.	Dennis Jen,Peter Parente,Jonathan Robbins,Christopher Weigle,Russell M. Taylor II,Alain Burette,Richard Weinberg	2004	ImageSurfer is a tool designed to explore correlations between two 3D scalar fields. Our scientific goal was to determine where a protein is located, and how much its concentration varies along the membrane of a neuronal dendrite. The 3D scalar field data sets fall into two categories: dendritic plasma membranes (defining the structure) and immunofluorescent staining (defining protein concentration along the structure). ImageSurfer enables scientists to analyze relationships between multiple data sets obtained with confocal microscopy by providing 3D surface view, height field, and graphing tools. Each tool reduces the complexity of the problem by extracting a restricted subset of data: finding a region of interest in 3D; getting a sense of relative concentrations in 2D, and getting exact concentration values in 1D. The current design is presented, along with the rationale for each design decision. The tool is already proving useful for data exploration, analysis, and presentation.
IEEE Visualization	Optimal Global Conformal Surface Parameterization.	Miao Jin,Yalin Wang,Shing-Tung Yau,Xianfeng Gu	2004	All orientable metric surfaces are Riemann surfaces and admit global conformal parameterizations. Riemann surface structure is a fundamental structure and governs many natural physical phenomena, such as heat diffusion and electro-magnetic fields on the surface. A good parameterization is crucial for simulation and visualization. This paper provides an explicit method for finding optimal global conformal parameterizations of arbitrary surfaces. It relies on certain holomorphic differential forms and conformal mappings from differential geometry and Riemann surface theories. Algorithms are developed to modify topology, locate zero points, and determine cohomology types of differential forms. The implementation is based on a finite dimensional optimization method. The optimal parameterization is intrinsic to the geometry, preserves angular structure, and can play an important role in various applications including texture mapping, remeshing, morphing and simulation. The method is demonstrated by visualizing the Riemann surface structure of real surfaces represented as triangle meshes.
IEEE Visualization	Panel 3: The Future Visualization Platform.	Greg Johnson,David S. Ebert,Chuck Hansen,David Kirk,Bill Mark,Hanspeter Pfister	2004	Panel 3: The Future Visualization Platform.
IEEE Visualization	Visualizing Turbulent Flow.	Gregory P. Johnson,Kelly P. Gaither,Victor Calo	2004	Visualizing Turbulent Flow.
IEEE Visualization	Light Weight Space Leaping using Ray Coherence.	Sarang Lakare,Arie E. Kaufman	2004	We present a space leaping technique for accelerating volume rendering with very low space and run-time complexity. Our technique exploits the ray coherence during ray casting by using the distance a ray traverses in empty space to leap its neighboring rays. Our technique works with parallel as well as perspective volume rendering, does not require any pre-processing or 3D data structures, and is independent of the transfer function. Being an image-space technique, it is independent of the complexity of the data being rendered. It can be used to accelerate both time-coherent and non coherent animation sequences.
IEEE Visualization	Investigating Swirl and Tumble Flow with a Comparison of Visualization Techniques.	Robert S. Laramee,Daniel Weiskopf,Jürgen Schneider,Helwig Hauser	2004	We investigate two important, common fluid flow patterns from computational fluid dynamics (CFD) simulations, namely, swirl and tumble motion typical of automotive engines. We study and visualize swirl and tumble flow using three different flow visualization techniques: direct, geometric, and texture-based. When illustrating these methods side-by-side, we describe the relative strengths and weaknesses of each approach within a specific spatial dimension and across multiple spatial dimensions typical of an engineerýs analysis. Our study is focused on steady-state flow. Based on this investigation we offer perspectives on where and when these techniques are best applied in order to visualize the behavior of swirl and tumble motion.
IEEE Visualization	Rough Interface Reconstruction Using the Level Set Method.	Yootai Kim,Raghu Machiraju,David Thompson	2004	We present a new level set method for reconstructing interfaces from point aggregations. Although level-set-based methods are advantageous because they can handle complicated topologies and noisy data, most tend to smooth the inherent roughness of the original data. Our objective is to enhance the quality of a reconstructed surface by preserving certain roughness-related characteristics of the original dataset. Our formulation employs the total variation of the surface as a roughness measure. The algorithm consists of two steps: a roughness-capturing flow and a roughness-preserving flow. The roughness capturing step attempts to construct a surface for which the original roughness is captured ¿ distance flow is well suited for roughness capturing. Surface reconstruction is enhanced by using a total variation preserving (TVP) scheme for the roughness-preserving flow. The shock filter formulation of Osher and Rudin is exploited to achieve this goal. In practice, we have found that better results are obtained by balancing the TVP term with a smoothing term based on curvature. The algorithm is applied to both fractal surface growth simulations and scanned data sets to demonstrate the efficacy of our approach.
IEEE Visualization	Light Collages: Lighting Design for Effective Visualization.	Chang Ha Lee,Xuejun Hao,Amitabh Varshney	2004	We introduce Light Collages ¿ a lighting design system for effective visualization based on principles of human perception. Artists and illustrators enhance perception of features with lighting that is locally consistent and globally inconsistent. Inspired by these techniques, we design the placement of light sources to convey a greater sense of realism and better perception of shape with globally inconsistent lighting. Our algorithm segments the objects into local surface patches and uses a number of perceptual heuristics, such as highlights, shadows, and silhouettes, to enhance the perception of shape. We show our results on scientific and sculptured datasets.
IEEE Visualization	Projecting Tetrahedra without Rendering Artifacts.	Martin Kraus,Wei Qiao,David S. Ebert	2004	Hardware-accelerated direct volume rendering of unstructured volumetric meshes is often based on tetrahedral cell projection, in particular, the Projected Tetrahedra (PT) algorithm and its variants. Unfortunately, even implementations of the most advanced variants of the PT algorithm are very prone to rendering artifacts. In this work, we identify linear interpolation in screen coordinates as a cause for significant rendering artifacts and implement the correct perspective interpolation for the PT algorithm with programmable graphics hardware. We also demonstrate how to use features of modern graphics hardware to improve the accuracy of the coloring of individual tetrahedra and the compositing of the resulting colors, in particular, by employing a logarithmic scale for the pre-integrated color lookup table, using textures with high color resolution, rendering to floating-point color buffers, and alpha dithering. Combined with a correct visibility ordering, these techniques result in the first implementation of the PT algorithm without objectionable rendering artifacts. Apart from the important improvement in rendering quality, our approach also provides a test bed for different implementations of the PT algorithm that allows us to study the particular rendering artifacts introduced by these variants.
IEEE Visualization	Methods for Efficient, High Quality Volume Resampling in the Frequency Domain.	Aili Li,Klaus Mueller,Thomas Ernst	2004	Resampling is a frequent task in visualization and medical imaging. It occurs whenever images or volumes are magnified, rotated, translated, or warped. Resampling is also an integral procedure in the registration of multi-modal datasets, such as CT, PET, and MRI, in the correction of motion artifacts in MRI, and in the alignment of temporal volume sequences in fMRI. It is well known that the quality of the resampling result depends heavily on the quality of the interpolation filter used. However, high-quality filters are rarely employed in practice due to their large spatial extents. In this paper, we explore a new resampling technique that operates in the frequency-domain where high- quality filtering is feasible. Further, unlike previous methods of this kind, our technique is not limited to integer-ratio scaling factors, but can resample image and volume datasets at any rate. This would usually require the application of slow Discrete Fourier Transforms (DFT) to return the data to the spatial domain. We studied two methods that successfully avoid these delays: the chirp-z transform and the FFTW package. We also outline techniques to avoid the ringing artifacts that may occur with frequency-domain filtering. Thus, our method can achieve high-quality interpolation at speeds that are usually associated with spatial filters of far lower quality.
IEEE Visualization	Interactive Point-Based Isosurface Extraction.	Yarden Livnat,Xavier Tricoche	2004	We propose a novel point-based approach to view dependent iso-surface extraction. We introduce a fast visibility query system for the view dependent traversal, which exhibits moderate memory requirements. This technique allows for an interactive interrogation of the full visible woman dataset (1GB) at four to fifteen frames per second on a desktop computer. The point-based approach is built on an extraction scheme that classifies different sections of the iso-surface into four categories, depending on the size of the geometry when projected onto the screen. In particular, we use points to represent small and sub-pixel triangles, as well as larger sections of the isosurface whose projection has sub-pixel size. To assign consistent and robust normals to individual points representing such regions, we propose to compute them during post processing of the extracted isosurface and provide the corresponding hardware implementation.
IEEE Visualization	Lighting Transfer Functions Using Gradient Aligned Sampling.	Eric B. Lum,Kwan-Liu Ma	2004	An important task in volume rendering is the visualization of boundaries between materials. This is typically accomplished using transfer functions that increase opacity based on a voxelýs value and gradient. Lighting also plays a crucial role in illustrating surfaces. In this paper we present a multi-dimensional transfer function method for enhancing surfaces, not through the variation of opacity, but through the modification of surface shading. The technique uses a lighting transfer function that takes into account the distribution of values along a material boundary and features a novel interface for visualizing and specifying these transfer functions. With our method, the user is given a means of visualizing boundaries without modifying opacity, allowing opacity to be used for illustrating the thickness of homogeneous materials through the absorption of light.
IEEE Visualization	Constrained Inverse Volume Rendering for Planetary Nebulae.	Marcus A. Magnor,Gordon L. Kindlmann,Charles D. Hansen	2004	Determining the three-dimensional structure of distant astronomical objects is a challenging task, given that terrestrial observations provide only one viewpoint. For this task, bipolar planetary nebulae are interesting objects of study because of their pronounced axial symmetry due to fundamental physical processes. Making use of this symmetry constraint, we present a technique to automatically recover the axisymmetric structure of bipolar planetary nebulae from two-dimensional images. With GPU-based volume rendering driving a non-linear optimization, we estimate the nebulaýs local emission density as a function of its radial and axial coordinates, and we recover the orientation of the nebula relative to Earth. The optimization refines the nebula model and its orientation by minimizing the differences between the rendered image and the original astronomical image. The resulting model enables realistic 3D visualizations of planetary nebulae, e.g. for educational purposes in planetarium shows. In addition, the recovered spatial distribution of the emissive gas allows validating computer simulation results of the astrophysical formation processes of planetary nebulae.
IEEE Visualization	Automatic Fast Detection of Tumor Suspect Areas on CT Scan.	Matei Mancas,Bernard Gosselin,Benoit M. Macq	2004	Automatic Fast Detection of Tumor Suspect Areas on CT Scan.
IEEE Visualization	JointViewer - An Interactive System for Exploring Orthopedic Data.	G. Elisabeta Marai,Çagatay Demiralp,Stuart Andrews,David H. Laidlaw	2004	JointViewer - An Interactive System for Exploring Orthopedic Data.
IEEE Visualization	TetSplat Real-Time Rendering and Volume Clipping of Large Unstructured Tetrahedral Meshes.	Ken Museth,Santiago V. Lombeyda	2004	We present a novel approach to interactive visualization and exploration of large unstructured tetrahedral meshes. These massive 3D meshes are used in mission-critical CFD and structural mechanics simulations, and typically sample multiple field values on several millions of unstructured grid points. Our method relies on the preprocessing of the tetrahedral mesh to partition it into non-convex boundaries and internal fragments that are subsequently encoded into compressed multi-resolution data representations. These compact hierarchical data structures are then adaptively rendered and probed in real-time on a commodity PC. Our point-based rendering algorithm, which is inspired by QSplat, employs a simple but highly efficient splatting technique that guarantees interactive frame-rates regardless of the size of the input mesh and the available rendering hardware. It furthermore allows for real-time probing of the volumetric data-set through constructive solid geometry operations as well as interactive editing of color transfer functions for an arbitrary number of field values. Thus, the presented visualization technique allows end-users for the first time to interactively render and explore very large unstructured tetrahedral meshes on relatively inexpensive hardware.
IEEE Visualization	STEPS - An Application for Simulation of Transsphenoidal Endonasal Pituitary Surgery.	André Neubauer,Stefan Wolfsberger,Marie-Thérèse Forster,Lukas Mroz,Rainer Wegenkittl,Katja Bühler	2004	Endonasal transsphenoidal pituitary surgery is a minimally invasive endoscopic procedure, applied to remove various kinds of pituitary tumors. To reduce the risk associated with this treatment, the surgeon must be skilled and well-prepared. Virtual endoscopy can be beneficial as a tool for training, preoperative planning and intraoperative support. This paper introduces STEPS, a virtual endoscopy system designed to aid surgeons in getting acquainted with the endoscopic view, the handling of instruments, the transsphenoidal approach and challenges associated with the procedure. STEPS also assists experienced surgeons in planning a real endoscopic intervention by getting familiar with the individual patient anatomy, identifying landmarks, planning the approach and deciding upon the ideal target position of the actual surgical activity. Besides interactive visualization using two different first-hit ray casting techniques, the application provides navigation and perception aids and the possibility to simulate the procedure, including haptic feedback and simulation of surgical instruments.
IEEE Visualization	Compression, Segmentation, and Modeling of Large-Scale Filamentary Volumetric Data.	Bruce H. McCormick,Purna Doddapaneni,David Mayerich,Zeki Melek,John Keyser	2004	We describe a method for processing large amounts of volumetric data collected from a Knife Edge Scanning Microscope (KESM). The neuronal data that we acquire consists of thin, branching structures extending over very large regions that prior volumetric representations have difficulty dealing with efficiently. Since the full volume data set can be extremely large, on-the-fly processing of the data is necessary.
IEEE Visualization	Scout: A Hardware-Accelerated System for Quantitatively Driven Visualization and Analysis.	Patrick S. McCormick,Jeff T. Inman,James P. Ahrens,Charles D. Hansen,Greg Roth	2004	Quantitative techniques for visualization are critical to the successful analysis of both acquired and simulated scientific data. Many visualization techniques rely on indirect mappings, such as transfer functions, to produce the final imagery. In many situations, it is preferable and more powerful to express these mappings as mathematical expressions, or queries, that can then be directly applied to the data. In this paper, we present a hardware-accelerated system that provides such capabilities and exploits current graphics hardware for portions of the computational tasks that would otherwise be executed on the CPU. In our approach, the direct programming of the graphics processor using a concise data parallel language, gives scientists the capability to efficiently explore and visualize data sets.
IEEE Visualization	Radial Hermite Operators for Scattered Point Cloud Data with Normal Vectors and Applications to Implicitizing Polygon Mesh Surfaces for Generalized CSG Operations and Smoothing.	Gregory M. Nielson	2004	We describe a new technique for fitting scattered point cloud data. Given a scattered point cloud of 3D data points and associated normal vectors, our new method produces an implicit volume model whose zero level isosurface interpolates the given points and associated normal vectors. In this paper, we concentrate on certain application of these new volume modeling techniques. We take existing polygon mesh surfaces and use the present methods to construct implicit volume models for these surfaces. Implicit models allow for the application of Boolean operations on these surfaces through the techniques of constructive solid geometry. Also, standard wavelet and filter operators can be applied to the implicit volume model leading to effective smoothing and filtering algorithms which are simple to implement.
IEEE Visualization	Dual Marching Cubes.	Gregory M. Nielson	2004	We present the definition and computational algorithms for a new class of surfaces which are dual to the isosurface produced by the widely used marching cubes (MC) algorithm. These new isosurfaces have the same separating properties as the MC surfaces but they are comprised of quad patches that tend to eliminate the common negative aspect of poorly shaped triangles of the MC isosurfaces. Based upon the concept of this new dual operator, we describe a simple, but rather effective iterative scheme for producing smooth separating surfaces for binary, enumerated volumes which are often produced by segmentation algorithms. Both the dual surface algorithm and the iterative smoothing scheme are easily implemented.
IEEE Visualization	Detection and Visualization of Anomalous Structures in Molecular Dynamics Simulation Data.	Sameep Mehta,Kaden Hazzard,Raghu Machiraju,Srinivasan Parthasarathy,John Wilkins	2004	In this article we explore techniques to detect and visualize features in data from molecular dynamics (MD) simulations. Although the techniques proposed are general, we focus on silicon (Si) atomic systems. The first set of methods use 3D location of atoms. Defects are detected and categorized using local operators and statistical modeling. Our second set of exploratory techniques employ electron density data. This data is visualized to glean the defects. We describe techniques to automatically detect the salient iso-values for iso-surface extraction and designing transfer functions.We compare and contrast the results obtained from both sources of data. Essentially, we find that the methods of defect (feature) detection are at least as robust as those based on the exploration of electron density for Si systems.
IEEE Visualization	Modeling Decomposing Objects under Combustion.	Zeki Melek,John Keyser	2004	We present a simple yet effective method for modeling of object decomposition under combustion. A separate simulation models the flame production and generates heat from a combustion process, which is used to trigger pyrolysis of the solid object. The decomposition is modeled using level set methods, and can handle complex topological changes. Even with a very simple flame model on a coarse grid, we can achieve a plausible decomposition of the burning object.
IEEE Visualization	Interactive Thickness Visualization of Articular Cartilage.	Matej Mlejnek,Anna Vilanova,Eduard Gröller	2004	This paper describes a method to visualize the thickness of curved thin objects. Given the MRI volume data of articular cartilage, medical doctors investigate pathological changes of the thickness. Since the tissue is very thin, it is impossible to reliably map the thickness information by direct volume rendering. Our idea is based on unfolding of such structures preserving their thickness. This allows to perform anisotropic geometrical operations (e.g., scaling the thickness). However, flattening of a curved structure implies a distortion of its surface. The distortion problem is alleviated through a focus-and-context minimization approach. Distortion is smallest close to a focal point which can be interactively selected by the user.
IEEE Visualization	Haptic Display of Interaction between Textured Models.	Miguel A. Otaduy,Nitin Jain,Avneesh Sud,Ming C. Lin	2004	Surface texture is among the most salient haptic characteristics of objects; it can induce vibratory contact forces that lead to perception of roughness. In this paper, we present a new algorithm to display haptic texture information resulting from the interaction between two textured objects. We compute contact forces and torques using low-resolution geometric representations along with texture images that encode surface details. We also introduce a novel force model based on directional penetration depth and describe an efficient implementation on programmable graphics hardware that enables interactive haptic texture rendering of complex models. Our force model takes into account important factors identified by psychophysics studies and is able to haptically display interaction due to fine surface textures that previous algorithms do not capture.
IEEE Visualization	Interactive Exploration of Large Remote Micro-CT Scans.	Steffen Prohaska,Andrei Hutanu,Ralf Kähler,Hans-Christian Hege	2004	Datasets of tens of gigabytes are becoming common in computational and experimental science. This development is driven by advances in imaging technology, producing detectors with growing resolutions, as well as availability of cheap processing power and memory capacity in commodity-based computing clusters. In this article we describe the design of a visualization system that allows scientists to interactively explore large remote data sets in an efficient and flexible way. The system is broadly applicable and currently used by medical scientists conducting an osteoporosis research project. Human vertebral bodies are scanned using a high resolution micro-CT scanner producing scans of roughly 8 GB size each. All participating research groups require access to the centrally stored data. Due to the rich internal bone structure, scientists need to interactively explore the full dataset at coarse levels, as well as visualize subvolumes of interest at the highest resolution. Our solution is based on HDF5 and GridFTP. When accessing data remotely, the HDF5 data processing pipeline is modified to support efficient retrieval of subvolumes. We reduce the overall latency and optimize throughput by executing high-level operations on the remote side. The GridFTP protocol is used to pass the HDF5 requests to a customized server. The approach takes full advantage of local graphics hardware for rendering. Interactive visualization is accomplished using a background thread to access the datasets stored in a multi-resolution format. A hierarchical volume renderer provides seamless integration of high resolution details with low resolution overviews.
IEEE Visualization	Dispersion Simulation and Visualization For Urban Security.	Feng Qiu,Ye Zhao,Zhe Fan,Xiaoming Wei,Haik Lorenz,Jianning Wang,Suzanne Yoakum-Stover,Arie E. Kaufman,Klaus Mueller	2004	We present a system for simulating and visualizing the propagation of dispersive contaminants with an application to urban security. In particular, we simulate airborne contaminant propagation in open environments characterized by sky-scrapers and deep urban canyons. Our approach is based on the Multiple Relaxation Time Lattice Boltzmann Model (MRTLBM), which can efficiently handle complex boundary conditions such as buildings. In addition, we model thermal effects on the flow field using the hybrid thermal MRTLBM. Our approach can also accommodate readings from various sensors distributed in the environment and adapt the simulation accordingly. We accelerate the computation and efficiently render many buildings with small textures on the GPU.We render streamlines and the contaminant smoke with self-shadowing composited with the textured buildings.
IEEE Visualization	Visualizing the Evolution of Horned Lizards Using 3D Morphing Techniques.	Reuben Reyes,Wendy L. Hodges,Kelly P. Gaither	2004	Visualizing the Evolution of Horned Lizards Using 3D Morphing Techniques.
IEEE Visualization	Panel 1: Can We Determine the Top Unresolved Problems of Visualization?	Theresa-Marie Rhyne,William L. Hibbard,Chris Johnson,Chaomei Chen,Steve Eick	2004	Panel 1: Can We Determine the Top Unresolved Problems of Visualization?
IEEE Visualization	Visualizing Cortical Waves and Timing from Data.	Kay A. Robbins,Mark Robinson,David M. Senseman	2004	Waves are a fundamental mechanism for conveying information in many physical problems. Direct visualization techniques are often used to display wave fronts. However, the information derived from such visualizations may not be as central to an investigation as an understanding of how the location, structure and time course of the wave change as key experimental parameters are varied. In experimental data, these questions are confounded by noise and incomplete data. Recognition of waves in networks of neurons is additionally complicated by the presence of long-range physical connections and recurrent excitation. This paper applies visual techniques to analyze the structural details of waves in response data from the turtle visual cortex. We emphasize low-cost visualizations that allow comparisons across neural data sets and variables to reconstruct the choreography for a complex response.
IEEE Visualization	Interactive Design of Multi-Perspective Images for Visualizing Urban Landscapes.	Augusto Román,Gaurav Garg,Marc Levoy	2004	Multi-perspective images are a useful way to visualize extended, roughly planar scenes such as landscapes or city blocks. However, constructing effective multi-perspective images is something of an art. In this paper, we describe an interactive system for creating multi-perspective images composed of serially blended cross-slits images. Beginning with a sideways-looking video of the scene as might be captured from a moving vehicle, we allow the user to interactively specify a set of cross-slits cameras, possibly with gaps between them. In each camera, one of the slits is defined to be the camera path, which is typically horizontal, and the user is left to choose the second slit, which is typically vertical. The system then generates intermediate views between these cameras using a novel interpolation scheme, thereby producing a multi-perspective image with no seams. The user can also choose the picture surface in space onto which viewing rays are projected, thereby establishing a parameterization for the image. We show how the choice of this surface can be used to create interesting visual effects. We demonstrate our system by constructing multi-perspective images that summarize city blocks, including corners, blocks with deep plazas and other challenging urban situations.
IEEE Visualization	Intuitive and Interactive Modification of Large Finite Element Models.	Dirc Rose,Katrin Bidmon,Thomas Ertl	2004	Virtual prototyping is increasingly replacing real mock-ups and experiments in industrial product development. Part of this process is the simulation of structural and functional properties, which is in many cases based on Finite Element Analysis (FEA). One prominent example from the automotive industry is the safety improvement resulting from crash worthiness simulations. A simulation model for this purpose usually consists of up to one million finite elements and is assembled from many parts which are individually meshed out of their CAD representation. In order to accelerate the development cycle, simulation engineers want to be able to modify their FE models without going back to the CAD department. Furthermore, valid CAD models might even not be available in preliminary design stages. However, in contrast to CAD, there is a lack of tools that offer the possibility of modification and processing of finite element components while maintaining the properties relevant to the simulation. In this application paper we present interactive algorithms for intuitive and fast editing of FE models and appropriate visualization techniques to support engineers in understanding these models. This includes new kinds of manipulators, feedback mechanisms and facilities for virtual reality and immersion at the workplace, e.g. autostereoscopic displays and haptic devices.
IEEE Visualization	Efficient Point-Based Isosurface Exploration Using the Span-Triangle.	Bartosz von Rymon-Lipinski,Nils Hanssen,Thomas Jansen,Lutz Ritter,Erwin Keeve	2004	We introduce a novel span-triangle data structure, based on the span-space representation for isosurfaces. It stores all necessary cell information for dynamic manipulation of the isovalue in an efficient way. We have found that using our data structure in combination with point-based techniques, implemented on graphics hardware, effects in real-time rendering and exploration. Our extraction algorithm utilizes an incremental and progressive update scheme, enabling smooth interaction without significant latency. Moreover, the corresponding visualization pipeline is capable of processing large data sets by utilizing all three levels of memory: disk, system and graphics. We address practical usability in actual medical applications, achieving a new level of interactivity.
IEEE Visualization	Vorticity Based Flow Analysis and Visualization for Pelton Turbine Design Optimization.	Filip Sadlo,Ronald Peikert,Etienne Parkinson	2004	Vorticity is the quantity used to describe the creation, transformation and extinction of vortices. It is present not only in vortices but also in shear flow. Especially in ducted flows, most of the overall vorticity is usually contained in the boundary layer. When a vortex develops from the boundary layer, this can be described by transport of vorticity. For a better understanding of a flow it is therefore of interest to examine vorticity in all of its different roles. The goal of this application study was not primarily the visualization of vortices but of vorticity distribution and its role in vortex phenomena. The underlying industrial case is a design optimization for a Pelton turbine. An important industrial objective is to improve the quality of the water jets driving the runner. Jet quality is affected mostly by vortices originating in the distributor ring. For a better understanding of this interrelation, it is crucial to not only visualize these vortices but also to analyze the mechanisms of their creation. We used various techniques for the visualization of vorticity, including field lines and modified isosurfaces. For field line based visualization, we extended the image-guided streamline placement algorithm of Turk and Banks to data-guided field line placement on three-dimensional unstructured grids.
IEEE Visualization	Display of Vector Fields Using a Reaction-Diffusion Model.	Allen R. Sanderson,Chris R. Johnson,Robert M. Kirby	2004	Effective visualization of vector fields relies on the ability to control the size and density of the underlying mapping to visual cues used to represent the field. In this paper we introduce the use of a reaction-diffusion model, already well known for its ability to form irregular spatio-temporal patters, to control the size, density, and placement of the vector field representation. We demonstrate that it is possible to encode vector field information (orientation and magnitude) into the parameters governing a reaction-diffusion model to form a spot pattern with the correct orientation, size, and density, creating an effective visualization. To encode direction we texture the spots using a light to dark fading texture. We also show that it is possible to use the reaction-diffusion model to visualize an additional scalar value, such as the uncertainty in the orientation of the vector field. An additional benefit of the reaction-diffusion visualization technique arises from its automatic density distribution. This benefit suggests using the technique to augment other vector visualization techniques. We demonstrate this utility by augmenting a LIC visualization with a reaction-diffusion visualization. Finally, the reaction-diffusion visualization method provides a technique that can be used for streamline and glyph placement.
IEEE Visualization	Generating Realistic Images from Hydrothermal Plume Data.	Kristina Santilli,Karen G. Bemis,Deborah Silver,Jamshed Dastur,Peter A. Rona	2004	Most data used in the study of seafloor hydrothermal plumes consists of sonar (acoustic) scans and sensor readings. Visual data captures only a portion of the sonar data range due to the prohibitive cost and physical infeasibility of taking sufficient lighting and video equipment to such extreme depths. However, visual images are available from research dives and from the recent IMAX movie, Volcanoes of the Deep Sea. In this application paper, we apply existing lighting models with forward scattering and light attenuation to the 3D sonar data in order to mimic the visual images available. These generated images are compared to existing visual images. This can help the geoscientists understand the relationship between these different data modalities and elucidate some of the mechanisms used to capture the data.
IEEE Visualization	Visualization of Nanoparticle Formation in Turbulent Flows.	P. Coleman Saunders,Sean C. Garrick,Victoria Interrante	2004	In this paper we offer methods for visualization of the formation of nanoparticles in turbulent flows. We present the use of pointillism as a technique to convey the distribution of nanoparticle sizes as texture in an area. We also demonstrate a method of producing and packing spot glyphs representative of the distribution of nanoparticle sizes at every point in the flow to produce an intuitive and extensible framework for the visualization.
IEEE Visualization	Immersive Design of DNA Molecules with a Tangible Interface.	Steven Schkolne,Hiroshi Ishii,Peter Schröder	2004	This paper presents an experimental immersive interface for designing DNA components for application in nanotechnology. While much research has been done on immersive visualization, this is one of the first systems to apply advanced interface techniques to a scientific design problem. This system uses tangible 3D input devices (tongs, a raygun, and a multipurpose handle tool) to create and edit a purely digital representation of DNS. The tangible controllers are associated with functions (not data) while a virtual display is used to render the model. This interface was built in collaboration with a research group investigating the design of DNA tiles. A user study shows that scientists find the immersive interface more satisfying than a 2D interface due to the enhanced understanding gained by directly interacting with molecules in 3D space.
IEEE Visualization	Compatible Triangulations of Spatial Decompositions.	William J. Schroeder,Berk Geveci,Mathieu Malaterre	2004	We describe a general algorithm to produce compatible 3D triangulations from spatial decompositions. Such triangulations match edges and faces across spatial cell boundaries, solving several problems in graphics and visualization including the crack problem found in adaptive isosurface generation, triangulation of arbitrary grids (including unstructured grids), clipping, and the interval tetrahedrization problem. The algorithm produces compatible triangulations on a cell-by-cell basis, using a modified Delaunay triangulation with a simple point ordering rule to resolve degenerate cases and produce unique triangulations across cell boundaries. The algorithm is naturally parallel since it requires no neighborhood cell information, only a unique, global point numbering. We show application of this algorithm to adaptive contour generation; tetrahedrization of unstructured meshes; clipping and interval volume mesh generation.
IEEE Visualization	Real-Time Volume Rendering of Four Channel Data Sets.	Jürgen P. Schulze,Alexander Rice	2004	We present a novel method to encode four data channels in a volumetric data set, and render it at interactive frame rates with maximum intensity projection (MIP) using textured polygons. The first three channels are stored in the volume textureýs red, green, and blue components. The fourth channel is stored in the alpha channel. To achieve real-time rendering speed we are using a pixel shader.
IEEE Visualization	Anisotropic Volume Rendering for Extremely Dense, Thin Line Data.	Gregory L. Schussman,Kwan-Liu Ma	2004	Many large scale physics-based simulations which take place on PC clusters or supercomputers produce huge amounts of data including vector fields. While these vector data such as electromagnetic fields, fluid flow fields, or particle paths can be represented by lines, the sheer number of the lines overwhelms the memory and computation capability of a high-end PC used for visualization. Further, very dense or intertwined lines, rendered with traditional visualization techniques, can produce unintelligible results with unclear depth relationships between the lines and no sense of global structure. Our approach is to apply a lighting model to the lines and sample them into an anisotropic voxel representation based on spherical harmonics as a preprocessing step. Then we evaluate and render these voxels for a given view using traditional volume rendering. For extremely large line based datasets, conversion to anisotropic voxels reduces the overall storage and rendering for O(n) lines to O(1) with a large constant that is still small enough to allow meaningful visualization of the entire dataset at nearly interactive rates on a single commodity PC.
IEEE Visualization	Vol-a-Tile - A Tool for Interactive Exploration of Large Volumetric Data on Scalable Tiled Displays.	Nicholas Schwarz,Shalini Venkataraman,Luc Renambot,Naveen K. Krishnaprasad,Venkatram Vishwanath,Jason Leigh,Andrew E. Johnson,Graham Kent,Atul Nayak	2004	We present the current state of Vol-a-Tile, an interactive tool for exploring large volumetric data on scalable tiled displays. Vol-a-Tile presents a variety of features employed by scientists at the Scripps Institution of Oceanography on data collected from the Anatomy of a Ridge-Axis Discontinuity seismic experiment. Hardware texture mapping and level-of-detail techniques provide interactivity. A high-performance network protocol is used to connect remote data sources over high-bandwidth photonic networks.
IEEE Visualization	Atlas-Aware Laplacian Smoothing.	Peter G. Sibley,Gabriel Taubin	2004	Atlas-Aware Laplacian Smoothing.
IEEE Visualization	Visualization of Topological Defects in Nematic Liquid Crystals Using Streamtubes, Streamsurfaces and Ellipsoids.	Vadim A. Slavin,David H. Laidlaw,Robert Pelcovits,Song Zhang,George Loriot,Andrew Callan-Jones	2004	Researchers in computational condensed matter physics deal with complex data sets consisting of time varying 3D tensor, vector, and scalar quantities. Particularly, in the research of topological defects in nematic liquid crystals (LC) displaying the results of the computer simulation of molecular dynamics presents a challenge. Combining existing immersive and interactive visualization methods we developed new methods that attempt to provide a clear, efficient, and intuitive way to visualize and explore LC data. In addition, the visualization of the data has presented us with a novel method of obtaining the locations of the topological defects present in a liquid crystal system.
IEEE Visualization	A Graphics Hardware-Based Vortex Detection and Visualization System.	Simon Stegmaier,Thomas Ertl	2004	Feature detection in flow fields is a well researched area, but practical application is often difficult due to the numerical complexity of the algorithms preventing interactive use and due to noise in experimental or high-resolution simulation data sets. We present an integrated system that provides interactive denoising, vortex detection, and visualization of vector data on Cartesian grids. All three major phases are implemented in such a way that the system runs completely on a modern GPU once the vector field is downloaded into graphics memory. The application aspect of our paper is two-fold. First, we show how recently presented, prototypical GPU-based algorithms for filtering, numerical computation, and volume rendering can be combined into one productive system by handling all idiosyncrasies of a chosen graphics card. Second, we demonstrate that the significant speedup achieved compared to an optimized software implementation now allows interactive exploration of characteristic structures in turbulent flow fields.
IEEE Visualization	The VesselGlyph: Focus & Context Visualization in CT-Angiography.	Matús Straka,Michal Cervenanský,Alexandra La Cruz,Arnold Köchl,Milos Srámek,Eduard Gröller,Dominik Fleischmann	2004	Accurate and reliable visualization of blood vessels is still a challenging problem, notably in the presence of morphologic changes resulting from atherosclerotic diseases. In this paper we take advantage of partially segmented data with approximately identified vessel centerlines to comprehensively visualize the diseased peripheral arterial tree. We introduce the VesselGlyph as an abstract notation for novel focus & context visualization techniques of tubular structures such as contrast-medium enhanced arteries in CT-Angiography (CTA). The proposed techniques combine direct volume rendering (DVR) and curved planar reformation (CPR) within a single image. The VesselGlyph consists of several regions where different rendering methods are used. The region type, the used visualization method and the region parameters depend on the distance from the vessel centerline and on viewing parameters as well. By selecting proper rendering techniques for different regions, vessels are depicted in a naturally looking and undistorted anatomic context. This may facilitate the diagnosis and treatment planning of patients with peripheral arterial occlusive disease. In this paper we furthermore present a way how to implement the proposed techniques in software and by means of modern 3D graphics accelerators.
IEEE Visualization	Real-Time Motion Estimation and Visualization on Graphics Cards.	Robert Strzodka,Christoph S. Garbe	2004	We present a tool for real-time visualization of motion features in 2D image sequences. The motion is estimated through an eigen-vector analysis of the spatio-temporal structure tensor at every pixel location. This approach is computationally demanding but allows reliable velocity estimates as well as quality indicators for the obtained results. We use a 2D color map and a region of interest selector for the visualization of the velocities. On the selected velocities we apply a hierarchical smoothing scheme which allows the choice of the desired scale of the motion field. We demonstrate several examples of test sequences in which some persons are moving with different velocities than others. These persons are visually marked in the real-time display of the image sequence. The tool is also applied to angiography sequences to emphasize the blood flow and its distribution. An efficient processing of the data streams is achieved by mapping the operations onto the stream architecture of standard graphics cards. The card receives the images and performs both the motion estimation and visualization, taking advantage of the parallelism in the graphics processor and the superior memory bandwidth. The integration of data processing and visualization also saves on unnecessary data transfers and thus allows the real-time analysis of 320x240 images. We expect that on the newest generation of graphics hardware our tool could run in real time for the standard VGA format.
IEEE Visualization	Topology Visualization of the Optical Power Flow through a Novel C-Shaped Nano-Aperture.	Liying Sun,Rajesh Batra,Xiaolei Shi,Lambertus Hesselink	2004	We recently discovered that C-shaped sub-wavelength (nano) metallic apertures when irradiated at specific resonance frequencies have extraordinary power transmission five to six orders of magnitude beyond what is observed for conventional round or square apertures. These apertures produce optical spot sizes as small as 25-50 nm using visible light in the near-field of the aperture with a brightness 10-100 times higher than that of the illuminating beam. A proper understanding into this remarkable phenomenon can aid in the development and understanding of a multitude of applications of these apertures including dense data storage, particle manipulation, and nano-scale photonic devices. Current scalar visualization approaches typically are insufficient to significantly aid in the understanding of these complex near-field optical problems. For example, two common approaches involving either visualization of scalar electromagnetic wave amplitudes in 2-D or rudimentary arrow plots of the vector fields produced in Finite-Difference-Time- Domain simulations are clearly inadequate. Both techniques provide only partial insight into the problem, as only specific planes can be visualized and therefore the global structure of the fields cannot be readily inferred. Understanding of the three-dimensional electromagnetic vector fields and energy flows related to the illumination of nano-sized apertures is critically important in near-field applications, as simple scalar analysis is not suitable at these small dimensions [8]. An ideal visualization tool that has not been used before in studying the optical behavior of near-field apertures is three-dimensional vector field topology. The global view of the vector field structure is deduced by locating singularities (critical points) within the field and augmenting these points with nearby streamlines. We have used for the first time, to the best of our knowledge, three-dimensional topology to analyze the topological differences between a resonant C-shaped nano-aperture and various non-resonant conventional apertures. The topological differences between these apertures are related to the superiority in power throughput of the C-aperture versus conventional round and square sub-wavelength apertures. We demonstrate how topological visualization techniques provide significant insight into the energy enhancement mechanism of the C aperture, and also shed light on critical issues related to the interaction between multiple apertures located in close proximity to each other, which gives rise to cross-talk, for example as a function of distance. Topological techniques allow us to develop design rules for the geometry of these apertures and their desired spot sizes and brightness. The performance of various sub-wavelength apertures can also be compared quantitatively based on their topology. Since topological methods are generically applicable to tensor and vector fields, our approach can be readily extended to provide insight into the broader category of Finite-Difference-Time-Domain nano-photonics and nano-science problems.
IEEE Visualization	Introducing Topological Attributes for Objective-Based Visualization.	Yuriko Takeshima,Shigeo Takahashi,Issei Fujishiro,Gregory M. Nielson	2004	Introducing Topological Attributes for Objective-Based Visualization.
IEEE Visualization	Stream Line and Path Line Oriented Topology for 2D Time-Dependent Vector Fields.	Holger Theisel,Tino Weinkauf,Hans-Christian Hege,Hans-Peter Seidel	2004	Topological methods aim at the segmentation of a vector field into areas of different flow behavior. For 2D time-dependent vector fields, two such segmentations are possible: either concerning the behavior of stream lines, or of path lines. While stream line oriented topology is well established, we introduce path line oriented topology as a new visualization approach in this paper. As a contribution to stream line oriented topology we introduce new methods to detect global bifurcations like saddle connections and cyclic fold bifurcations. To get the path line oriented topology we segment the vector field into areas of attracting, repelling and saddle-like behavior of the path lines. We compare both kinds of topologies and apply them to a number of data sets.
IEEE Visualization	Visualization of Intricate Flow Structures for Vortex Breakdown Analysis.	Xavier Tricoche,Christoph Garth,Gordon L. Kindlmann,Eduard Deines,Gerik Scheuermann,Markus Rütten,Charles D. Hansen	2004	Vortex breakdowns and flow recirculation are essential phenomena in aeronautics where they appear as a limiting factor in the design of modern aircrafts. Because of the inherent intricacy of these features, standard flow visualization techniques typically yield cluttered depictions. The paper addresses the challenges raised by the visual exploration and validation of two CFD simulations involving vortex breakdown. To permit accurate and insightful visualization we propose a new approach that unfolds the geometry of the breakdown region by letting a plane travel through the structure along a curve. We track the continuous evolution of the associated projected vector field using the theoretical framework of parametric topology. To improve the understanding of the spatial relationship between the resulting curves and lines we use direct volume rendering and multi-dimensional transfer functions for the display of flow-derived scalar quantities. This enriches the visualization and provides an intuitive context for the extracted topological information. Our results offer clear, synthetic depictions that permit new insight into the structural properties of vortex breakdowns.
IEEE Visualization	LoD Volume Rendering of FEA Data.	Shyh-Kuang Ueng,Yan-Jen Su,Chi-Tang Chang	2004	In this article, a new multiple resolution volume rendering method for Finite Element Analysis (FEA) data is presented. Our method is composed of three stages: At the first stage, the Gauss points of the FEA cells are calculated. The function values, gradients, diffusions, and influence scopes of the Gauss points are computed. By representing the Gauss points as graph vertices and connecting adjacent Gauss points with edges, an adjacency graph is created. The adjacency graph is used to represent the FEA data in the subsequent computation. At the second stage, a hierarchical structure is established upon the adjacency graph. Any two neighboring vertices with similar function values are merged into a new vertex. The similarity is measured by using a user-defined threshold. Consequently a new adjacency graph is constructed. Then the threshold is increased, and the graph reduction is triggered again to generate another adjacency graph. By repeating the processing, multiple adjacency graphs are computed, and a Level of Detail (LoD) representation of the FEA data is established. At the third stage, the LoD structure is rendered by using a splatting method. At first, a level of adjacency graph is selected by users. The graph vertices are sorted based on their visibility orders and projected onto the image plane in back-to-front order. Billboards are used to render the vertices in the projection. The function values, gradients, and influence scopes of the vertices are utilized to decide the colors, opacities, orientations, and shapes of the billboards. The billboards are then modulated with texture maps to generate the footprints of the vertices. Finally, these footprints are composited to produce the volume rendering image.
IEEE Visualization	Interactive Poster: Illustrating Different Convection Velocities of Turbulent Flow.	Timothy Urness,Victoria Interrante,Ellen Longmire,Ivan Marusic,Bharathram Ganapathisubramani	2004	The visualization of any vector field is dependent on the relative velocity of the observer. In experimentally generated vector fields, the average value of the streamwise component of the global vector field is typically calculated and subtracted from each vector. We demonstrate that the resulting image, critical points, and vector field features are greatly influenced by the magnitude of the value subtracted from the streamwise velocity.
IEEE Visualization	Importance-Driven Volume Rendering.	Ivan Viola,Armin Kanitsar,Eduard Gröller	2004	This paper introduces importance-driven volume rendering as a novel technique for automatic focus and context display of volumetric data. Our technique is a generalization of cut-away views, which ¿ depending on the viewpoint ¿ remove or suppress less important parts of a scene to reveal more important underlying information. We automatize and apply this idea to volumetric data. Each part of the volumetric data is assigned an object importance which encodes visibility priority. This property determines which structures should be readily discernible and which structures are less important. In those image regions, where an object occludes more important structures it is displayed more sparsely than in those areas where no occlusion occurs. Thus the objects of interest are clearly visible. For each object several representations, i.e., levels of sparseness, are specified. The display of an individual object may incorporate different levels of sparseness. The goal is to emphasize important structures and to maximize the information content in the final image. This paper also discusses several possible schemes for level of sparseness specification and different ways how object importance can be composited to determine the final appearance of a particular object.
IEEE Visualization	Generating Sub-Resolution Detail in Images and Volumes Using Constrained Texture Synthesis.	Lujin Wang,Klaus Mueller	2004	"A common deficiency of discretized datasets is that detail beyond the resolution of the dataset has been irrecoverably lost. This lack of detail becomes immediately apparent once one attempts to zoom into the dataset and only recovers blur. Here, we describe a method that generates the missing detail from any available and plausible high-resolution data, using texture synthesis. Since the detail generation process is guided by the underlying image or volume data and is designed to fill in plausible detail in accordance with the coarse structure and properties of the zoomed-in neighborhood, we refer to our method as constrained texture synthesis. Regular zooms become ""semantic zooms"", where each level of detail stems from a data source attuned to that resolution. We demonstrate our approach by a medical application ¿ the visualization of a human liver ¿ but its principles readily apply to any scenario, as long as data at all resolutions are available. We will first present a 2D viewing application, called the ""virtual microscope"", and then extend our technique to 3D volumetric viewing."
IEEE Visualization	On the Role of Color in the Perception of Motion in Animated Visualizations.	Daniel Weiskopf	2004	Although luminance contrast plays a predominant role in motion perception, significant additional effects are introduced by chromatic contrasts. In this paper, relevant results from psychophysical and physiological research are described to clarify the role of color in motion detection. Interpreting these psychophysical experiments, we propose guidelines for the design of animated visualizations, and a calibration procedure that improves the reliability of visual motion representation. The guidelines are applied to examples from texture-based flow visualization, as well as graph and tree visualization.
IEEE Visualization	Vector Wavelet Thresholding for Vector Field Denoising.	Michel A. Westenberg,Thomas Ertl	2004	Noise reduction is an important preprocessing step for many visualization techniques that make use of feature extraction. We propose a method for denoising 2-D vector fields that are corrupted by additive noise. The method is based on the vector wavelet transform and wavelet coefficient thresholding. We compare our wavelet-based denoising method with Gaussian filtering, and test the effect of these methods on the signal-to-noise ratio (SNR) of the vector fields before and after denoising. We also study the effect on relevant details for visualization, such as vortex measures. The results show that for low SNR, Gaussian filtering with large kernels has a somewhat higher performance than the wavelet-based method in terms of SNR. For larger SNR, the wavelet-based method outperforms Gaussian filtering. This is mostly due to the fact that Gaussian filtering tends to remove small details, which are preserved by the wavelet-based method.
IEEE Visualization	Force-Feedback-Enhanced Navigation for Interactive Visualization of Coronary Vessels.	Thomas Wischgoll,Elke Moritz,Jörg Meyer	2004	Coronary heart disease (CHD) is the number one killer in the United States. Although it is well known that CHD mainly occurs due to blocked arteries, there are contradictory results from studies designed to identify basic causes for this common disease is. To find out more about the true reason for CHD, virtual models can be employed to better understand the way the heart functions. With such a model, scientists and surgeons are able to analyze the effects of different treatment options, and ultimately find more suited ways to prevent coronary heart diseases. To investigate a given model, appropriate navigation methods are required, including suitable input devices. For the visualization, graphics cards originally designed for gaming applications are used; so, it is a just natural transition to adapt gaming input devices to a visualization system for controlling of the navigation. These devices are usually well designed with respect to ergonomics and durability, yielding more degrees of freedom in steering than two-dimensional input devices, such as desktop mice. This poster describes a visualization system that provides the user with advanced control devices for navigation enabling interactive exploration of the model. Force-feedback and sound effects provide additional cues.
IEEE Visualization	Interactive Poster: Grid-Enabled Collaborative Scientific Visualization Environment.	"Eric Christopher Wyatt,Patrick O'Leary"	2004	Interactive Poster: Grid-Enabled Collaborative Scientific Visualization Environment.
IEEE Visualization	Surface Reconstruction of Noisy and Defective Data Sets.	Hui Xie,Kevin T. McDonnell,Hong Qin	2004	We present a novel surface reconstruction algorithm that can recover high-quality surfaces from noisy and defective data sets without any normal or orientation information. A set of new techniques are introduced to afford extra noise tolerability, robust orientation alignment, reliable outlier removal, and satisfactory feature recovery. In our algorithm, sample points are first organized by an octree. The points are then clustered into a set of monolithically singly-oriented groups. The inside/outside orientation of each group is determined through a robust voting algorithm. We locally fit an implicit quadric surface in each octree cell. The locally fitted implicit surfaces are then blended to produce a signed distance field using the modified Shepardýs method. We develop sophisticated iterative fitting algorithms to afford improved noise tolerance both in topology recognition and geometry accuracy. Furthermore, this iterative fitting algorithm, coupled with a local model selection scheme, provides a reliable sharp feature recovery mechanism even in the presence of bad input.
IEEE Visualization	Code Checking and Visualization of an Architecture Design.	Rong Xu,Wawan Solihin,Zhiyong Huang	2004	Code Checking and Visualization of an Architecture Design.
IEEE Visualization	Rendering Implicit Flow Volumes.	Daqing Xue,Caixia Zhang,Roger Crawfis	2004	Traditional flow volumes construct an explicit geometrical or parametrical representation from the vector field. The geometry is updated interactively and then rendered using an unstructured volume rendering technique. Unless a detailed refinement of the flow volume is specified for the interior, information inside the underlying flow volume is lost in the linear interpolation. These disadvantages can be avoided and/or alleviated using an implicit flow model. An implicit flow is a scalar field constructed such that any point in the field is associated with a termination surface using an advection operator on the flow. We present two techniques, a slice-based three-dimensional texture mapping and an interval volume segmentation coupled with a tetrahedron projection-based renderer, to render implicit stream flows. In the first method, the implicit flow representation is loaded as a 3D texture and manipulated using a dynamic texture operation that allows the flow to be investigated interactively. In our second method, a geometric flow volume is extracted from the implicit flow using a high dimensional iso-contouring or interval volume routine. This provides a very detailed flow volume or set of flow volumes that can easily change topology, while retaining accurate characteristics within the flow volume. The advantages and disadvantages of these two techniques are compared with traditional explicit flow volumes.
IEEE Visualization	Quick-VDR: Interactive View-Dependent Rendering of Massive Models.	Sung-Eui Yoon,Brian Salomon,Russell Gayle,Dinesh Manocha	2004	We present a novel approach for interactive view-dependent rendering of massive models. Our algorithm combines view-dependent simplification, occlusion culling, and out-of-core rendering. We represent the model as a clustered hierarchy of progressive meshes (CHPM). We use the cluster hierarchy for coarse-grained selective refinement and progressive meshes for fine-grained local refinement. We present an out-of-core algorithm for computation of a CHPM that includes cluster decomposition, hierarchy generation, and simplification. We make use of novel cluster dependencies in the preprocess to generate crack-free, drastic simplifications at runtime. The clusters are used for occlusion culling and out-of-core rendering. We add a frame of latency to the rendering pipeline to fetch newly visible clusters from the disk and to avoid stalls. The CHPM reduces the refinement cost for view-dependent rendering by more than an order of magnitude as compared to a vertex hierarchy. We have implemented our algorithm on a desktop PC. We can render massive CAD, isosurface, and scanned models, consisting of tens or a few hundreds of millions of triangles at 10-35 frames per second with little loss in image quality.
IEEE Visualization	Fast Rendering of Foveated Volume in the Wavelet Domain.	Hang Yu,Ee-Chien Chang,Zhiyong Huang,Zhijian Zheng	2004	Fast Rendering of Foveated Volume in the Wavelet Domain.
IEEE Visualization	Dual Contouring with Topology-Preserving Simplification Using Enhanced Cell Representation.	Nan Zhang,Wei Hong,Arie E. Kaufman	2004	We present a fast, topology-preserving approach for isosurface simplification. The underlying concept behind our approach is to preserve the disconnected surface components in cells during isosurface simplification. We represent isosurface components in a novel representation, called enhanced cell, where each surface component in a cell is represented by a vertex and its connectivity information. A topology-preserving vertex clustering algorithm is applied to build a vertex octree. An enhanced dual contouring algorithm is applied to extract error-bounded multiresolution isosurfaces from the vertex octree while preserving the finest resolution isosurface topology. Cells containing multiple vertices are properly handled during contouring. Our approach demonstrates better results than existing octree-based simplification techniques.
IEEE Visualization	DTI Fiber Clustering in the Whole Brain.	Song Zhang,David H. Laidlaw	2004	DTI Fiber Clustering in the Whole Brain.
IEEE Visualization	Visualization of the Interaction of Multiple Sclerosis Lesions with Adjacent White Matter Fibers Using Streamtubes and Streamsurfaces.	Song Zhang,David H. Laidlaw,Jack Simon,Mark Brown,David Miller	2004	Visualization of the Interaction of Multiple Sclerosis Lesions with Adjacent White Matter Fibers Using Streamtubes and Streamsurfaces.
IEEE Visualization	Topological Lines in 3D Tensor Fields.	Xiaoqiang Zheng,Alex Pang	2004	Visualization of 3D tensor fields continues to be a major challenge in terms of providing intuitive and uncluttered images that allow the users to better understand their data. The primary focus of this paper is on finding a formulation that lends itself to a stable numerical algorithm for extracting stable and persistent topological features from 2nd order real symmetric 3D tensors. While features in 2D tensors can be identified as either wedge or trisector points, in 3D, the corresponding stable features are lines, not just points. These topological feature lines provide a compact representation of the 3D tensor field and are essential in helping scientists and engineers understand their complex nature. Existing techniques work by finding degenerate points and are not numerically stable, and worse, produce both false positive and false negative feature points. This paper seeks to address this problem with a robust algorithm that can extract these features in a numerically stable, accurate, and complete manner.
IEEE Visualization	Pixel-Exact Rendering of Spacetime Finite Element Solutions.	Yuan Zhou,Michael Garland,Robert Haber	2004	Computational simulation of time-varying physical processes is of fundamental importance for many scientific and engineering applications. Most frequently, time-varying simulations are performed over multiple spatial grids at discrete points in time. In this paper, we investigate a new approach to time-varying simulation: spacetime discontinuous Galerkin finite element methods. The result of this simulation method is a simplicial tessellation of spacetime with per-element polynomial solutions for physical quantities such as strain, stress, and velocity. To provide accurate visualizations of the resulting solutions, we have developed a method for per-pixel evaluation of solution data on the GPU.We demonstrate the importance of per-pixel rendering versus simple linear interpolation for producing high quality visualizations. We also show that our system can accommodate reasonably large datasets ¿ spacetime meshes containing up to 20 million tetrahedra are not uncommon in this domain.
IEEE Visualization	15th IEEE Visualization 2004 Conference (VIS 2004), 10-15 October 2004, Austin, TX, USA		2004	15th IEEE Visualization 2004 Conference (VIS 2004), 10-15 October 2004, Austin, TX, USA
Information Visualization	Dynamical system visualization and analysis via performance maps.	James J. Alpigini	2004	"Visualization techniques are common in the study of chaotic motion. These techniques range from simple time graphs and phase portraits to robust Julia sets, which are familiar to many as 'fractal images.' The utility of the Julia sets rests not in their considerable visual impact, but rather, in the color-coded information that they display about the dynamics of an iterated function. In this paper, a paradigm termed the performance map is presented, which is derived from the familiar Julia set. Performance maps are generated automatically for control or other dynamical system models over ranges of system parameters. The resulting visualizations require a minimum of a priori knowledge of the system under evaluation. By the use of color-coding, these images convey a wealth of information to the informed user about dynamic behaviors of a system that may be hidden from all but the expert analyst."
Information Visualization	Book Review: About Face 2.0: The essentials of interaction design.	Protima Banerjee	2004	Book Review: About Face 2.0: The essentials of interaction design.
Information Visualization	Multiple foci visualisation of large hierarchies with FlexTree.	Hongzhi Song,Edwin P. Curran,Roy Sterritt	2004	One of the main tasks in information visualisation research is creating visual tools to facilitate human understanding of large and complex information spaces. Hierarchies, being a good mechanism for organising such information, are ubiquitous. Although much research effort has been spent on finding useful representations for hierarchies, visualising large hierarchies is still a difficult topic. One of the difficulties is how to handle the ever increasing scale of hierarchies. Another is how to enable the user to focus on multiple selections of interest while maintaining context. This paper describes a hierarchy visualisation technique called FlexTree to address these problems. It contains some important features that have not been exploited so far. A profile or contour unique to the hierarchy being visualised can be viewed in a bar chart layout. A normalised view of a common attribute of all nodes can be selected by the user. Multiple foci are consistently accessible within a global context through interaction. Furthermore it can handle a large hierarchy that contains 10,000 nodes in a PC environment. This technique has been applied to visualise computer file system structures and decision trees from data mining results. The results from informal user evaluations against these two applications are also presented. User feedback suggests that FlexTree is suitable for visualising large decision trees.
Information Visualization	"Special Issue of selected and extended InfoVis '03 papers - Guest Editor' Introduction."	Stephen C. North,Tamara Munzner	2004	"Special Issue of selected and extended InfoVis '03 papers - Guest Editor' Introduction."
Information Visualization	Graph visualization for the analysis of the structure and dynamics of extreme-scale supercomputers.	Kenneth L. Summers,Thomas P. Caudell,Kathryn Berkbigler,Brian Bush,Kei Davis,Steve Smith	2004	We are exploring the development and application of information visualization techniques for the analysis of new massively parallel supercomputer architectures. Modern supercomputers typically comprise very large clusters of commodity SMPs interconnected by possibly dense and often non-standard networks. The scale, complexity, and inherent non-locality of the structure and dynamics of this hardware, and the operating systems and applications distributed over them, challenge traditional analysis methods. As part of the á la carte (A Los Alamos Computer Architecture Toolkit for Extreme-Scale Architecture Simulation) team at Los Alamos National Laboratory, who are simulating these new architectures, we are exploring advanced visualization techniques and creating tools to enhance analysis of these simulations with intuitive three-dimensional representations and interfaces. This work complements existing and emerging algorithmic analysis tools. In this paper, we give background on the problem domain, a description of a prototypical computer architecture of interest (on the order of 10,000 processors connected by a quaternary fat-tree communications network), and a presentation of three classes of visualizations that clearly display the switching fabric and the flow of information in the interconnecting network.
Information Visualization	Design choices when architecting visualizations.	Diane Tang,Chris Stolte,Robert Bosch	2004	In this paper, we focus on some of the key design decisions we faced during the process of architecting a visualization system and present some possible choices, with their associated advantages and disadvantages. We frame this discussion within the context of Rivet, our general visualization environment designed for rapidly prototyping interactive, exploratory visualization tools for analysis. As we designed increasingly sophisticated visualizations, we needed to refine Rivet in order to be able to create these richer displays for larger and more complex data sets. The design decisions we discuss in this paper include the internal data model, data access, semantic meta-data information the visualization can use to create effective visual encodings, the need for data transformations in a visualization tool, modular objects for flexibility, and the tradeoff between simplicity and expressiveness when providing methods for creating visualizations.
Information Visualization	The representation of neural data using visualization.	Martin A. Walter,Liz J. Stuart,Roman Borisyuk	2004	"Currently, the focus of research within Information Visualization is steering towards genomic data visualization due to the level of activity that the Human Genome Project has generated. However, the Human Brain project, renowned within Neuroinformatics, is equally challenging and exciting. Its main aim is to increase current understanding of brain function such as memory, learning, attention, emotions and consciousness. It is understood that this task will require the 'integration of information from the level of the gene to the level of behaviour'. The work presented in this paper focuses on the visualization of neural data. More specifically, the data being analysed is multi-dimensional spike train data. Traditional methods, such as the 'raster plot' and the 'cross-correlogram', are still useful but they do not scale up for larger assemblies of neurons. In this paper, a new innovative method called the Tunnel is defined. Its design is based on the principles of Information Visualization; overview the data, zoom and filter data, data details on demand. The features of this visualization environment are described. This includes data filtering, navigation and a 'flat map' overview facility. Additionally, a 'coincidence overlay map' is presented. This map washes the Tunnel with colour, which encodes the coincidence of spikes."
Information Visualization	Compound brushing explained.	Hong Chen	2004	This paper proposes a conceptual model called compound brushing for modeling the brushing techniques used in dynamic data visualization. In this approach, brushing techniques are modeled as higraphs with five types of basic entities: data, selection, device, renderer, and transformation. Using this model, a flexible visual programming tool is designed not only to configure and control various common types of brushing techniques currently used in dynamic data visualization but also to investigate new brushing techniques.
Information Visualization	VISTA: validating and refining clusters via visualization.	Keke Chen,Ling Liu	2004	Clustering is an important technique for understanding of large multi-dimensional datasets. Most of clustering research to date has been focused on developing automatic clustering algorithms and cluster validation methods. The automatic algorithms are known to work well in dealing with clusters of regular shapes, for example, compact spherical shapes, but may incur higher error rates when dealing with arbitrarily shaped clusters. Although some efforts have been devoted to addressing the problem of skewed datasets, the problem of handling clusters with irregular shapes is still in its infancy, especially in terms of dimensionality of the datasets and the precision of the clustering results considered. Not surprisingly, the statistical indices works ineffective in validating clusters of irregular shapes, too. In this paper, we address the problem of clustering and validating arbitrarily shaped clusters with a visual framework (VISTA). The main idea of the VISTA approach is to capitalize on the power of visualization and interactive feedbacks to encourage domain experts to participate in the clustering revision and clustering validation process. The VISTA system has two unique features. First, it implements a linear and reliable visualization model to interactively visualize multi-dimensional datasets in a 2D star-coordinate space. Second, it provides a rich set of user-friendly interactive rendering operations, allowing users to validate and refine the cluster structure based on their visual experience as well as their domain knowledge.
Information Visualization	Analyzing perceptual organization in information graphics.	Martin Wattenberg,Danyel Fisher	2004	"We propose a new method for assessing the perceptual organization of information graphics, based on the premise that the visual structure of an image should match the structure of the data it is intended to convey. The core of our method is a new formal model of one type of perceptual structure, based on classical machine vision techniques for analyzing an image at multiple resolutions. The model takes as input an arbitrary grayscale image and returns a lattice structure describing the visual organization of the image. We show how this model captures several aspects of traditional design aesthetics, and we describe a software tool that implements the model to help designers analyze and refine visual displays. Our emphasis here is on demonstrating the model's potential as a design aid rather than as a description of human perception, but given its initial promise we propose a variety of ways in which the model could be extended and validated."
Information Visualization	Feel the information with VisPad: a large area vibrotactile device.	Doanna Weissgerber,Bruce Bridgeman,Alex Pang	2004	"A new haptics design for visualizing data is constructed out of commodity massage pads and custom controllers and interfaces to a computer. It is an output device for information that can be transmitted to a user who sits on the pad. Two unique properties of the design are: (a) its large feedback area and (b) its passive nature, where unlike most current haptics devices, the user's hands are free to work on other things. To test how useful such a device is for visualizing data, we added the VisPad interface to our protein structure-alignment program (ProtAlign) and performed usability studies. The studies demonstrated that information could be perceived significantly faster utilizing our multi-modal presentation compared to vision-based graphical visualization alone."
Information Visualization	Visualising changes in fund manager holdings in two and a half-dimensions.	Tim Dwyer,David R. Gallagher	2004	"We explore a multiple view, or overview and detail, method for visualising a high-dimensional portfolio holdings data set with attributes that change over time. The method employs techniques from multidimensional scaling and graph visualisation to find a two-dimensional mapping for high-dimensional data. In both the overview and detail views, time is mapped to the third dimension providing a two and a half-dimensional view of changes in the data. We demonstrate the utility of the paradigm with a prototype system for visualisation of movements within a large set of UK fund managers' stock portfolios."
Information Visualization	Automatic layout of UML class diagrams in orthogonal style.	Markus Eiglsperger,Carsten Gutwenger,Michael Kaufmann,Joachim Kupke,Michael Jünger,Sebastian Leipert,Karsten Klein,Petra Mutzel,Martin Siebenhaller	2004	Unified modelling language (UML) diagrams have become increasingly important in engineering and re-engineering processes for software systems. Of particular interest are UML class diagrams whose purpose is to display generalizations, associations, aggregations, and compositions in one picture. The combination of directed and undirected relations poses a special challenge to a graph layout tool. Current approaches for the automatic layout of class diagrams are based on the layered graph drawing paradigm. These algorithms produce good results for class diagrams with large and deep structural information, that is, diagrams with a large and deep inheritance hierarchy. However, they do not perform satisfactorily in absence of this information. We suggest to use the topology-shape--metrics paradigm for automatic layout of class diagrams, which has been used very successfully for drawing undirected graphs in orthogonal style. Moreover, we introduce the algorithms UML-Kandinsky and GoVisual fitting into this paradigm. Both algorithms work for class diagrams with rich structural information as well as for class diagrams with few or no structural information. Therefore, they improve the existing algorithms significantly.
Information Visualization	Animated visualization of causal relations through growing 2D geometry.	Niklas Elmqvist,Philippas Tsigas	2004	Causality visualization is an important tool for many scientific domains that involve complex interactions between multiple entities (examples include parallel and distributed systems in computer science). However, traditional visualization techniques such as Hasse diagrams are not well-suited to large system executions, and users often have difficulties answering even basic questions using them, or have to spend inordinate amounts of time to do so. In this paper, we present the Growing Squares and Growing Polygons methods, two sibling visualization techniques that were designed to solve this problem by providing efficient 2D causality visualization through the use of color, texture, and animation. Both techniques have abandoned the traditional linear timeline and instead map the time parameter to the size of geometrical primitives representing the processes; in the Growing Squares case, each process is a color-coded square that receives color influences from other process squares as messages reach it; in the Growing Polygons case, each process is instead an n-sided polygon consisting of triangular sectors showing color-coded influences from the other processes. We have performed user studies of both techniques, comparing them with Hasse diagrams, and they have been shown to be significantly more efficient than old techniques, both in terms of objective performance as well as the subjective opinion of the test subjects (the Growing Squares technique is, however, only significantly more efficient for small systems).
Information Visualization	Representations of Time and Space.	Ian Greatbatch	2004	Representations of Time and Space.
Information Visualization	Dynamic query tools for time series data sets: Timebox widgets for interactive exploration.	Harry Hochheiser,Ben Shneiderman	2004	"Timeboxes are rectangular widgets that can be used in direct-manipulation graphical user interfaces (GUIs) to specify query constraints on time series data sets. Timeboxes are used to specify simultaneously two sets of constraints: given a set of N time series profiles, a timebox covering time periods x1...x2 (x1<x2) and values y1...y2(y1≤y2) will retrieve only those n ∈N that have values y1≤y≤y2 during all times x1≤x≤x2. TimeSearcher is an information visualization tool that combines timebox queries with overview displays, query-by-example facilities, and support for queries over multiple time-varying attributes. Query manipulation tools including pattern inversion and 'leaders & laggards' graphical bookmarks provide additional support for interactive exploration of data sets. Extensions to the basic timebox model that provide additional expressivity include variable time timeboxes, which can be used to express queries with variability in the time interval, and angular queries, which search for ranges of differentials, rather than absolute values. Analysis of the algorithmic requirements for providing dynamic query performance for timebox queries showed that a sequential search outperformed searches based on geometric indices. Design studies helped identify the strengths and weaknesses of the query tools. Extended case studies involving the analysis of two different types of data from molecular biology experiments provided valuable feedback and validated the utility of both the timebox model and the TimeSearcher tool. Timesearcher is available at http://www.cs.umd.edu/hcil/timesearcher"
Information Visualization	GAMMATELLA: visualizing program-execution data for deployed software.	James A. Jones,Alessandro Orso,Mary Jean Harrold	2004	Software systems are often released with missing functionality, errors, or incompatibilities that may result in failures in the field, inferior performances, or, more generally, user dissatisfaction. In previous work, some of the authors presented the GAMMA approach, whose goal is to improve software quality by augmenting software-engineering tasks with dynamic information collected from deployed software. The GAMMA approach enables analyses that (1) rely on actual field data instead of synthetic in-house data and (2) leverage the vast and heterogeneous resources of an entire user community instead of limited, and often homogeneous, in-house resources. When monitoring a large number of deployed instances of a software product, however, a significant amount of data is collected. Such raw data are useless in the absence of suitable data-mining and visualization techniques that support exploration and understanding of the data. In this paper, we present a new technique for collecting, storing, and visualizing program-execution data gathered from deployed instances of a software product. We also present a prototype toolset, GAMMATELLA, that implements the technique. Finally, we show how the visualization capabilities of GAMMATELLA facilitate effective investigation of several kinds of execution-related information in an interactive fashion, and discuss our initial experience with a semi-public display of GAMMATELLA.
Information Visualization	Visualization of high-dimensional data with relational perspective map.	James Xinzhi Li	2004	This paper introduces a method called relational perspective map (RPM) to visualize distance information in high-dimensional spaces. Like conventional multidimensional scaling, the RPM algorithm aims to produce proximity preserving 2-dimensional (2-D) maps. The main idea of the RPM algorithm is to simulate a multiparticle system on a closed surface: whereas the repulsive forces between the particles reflect the distance information, the closed surface holds the whole system in balance and prevents the resulting map from degeneracy. A special feature of RPM algorithm is its ability to partition a complex dataset into pieces and map them onto a 2-D space without overlapping. Compared to other multidimensional scaling methods, RPM is able to reveal more local details of complex datasets. This paper demonstrates the properties of RPM maps with four examples and provides extensive comparison to other multidimensional scaling methods, such as Sammon Mapping and Curvilinear Principle Analysis.
Information Visualization	A pivot-based routine for improved parent-finding in hybrid MDS.	Alistair Morrison,Matthew Chalmers	2004	The problem of exploring or visualising data of high dimensionality is central to many tools for information visualisation. Through representing a data set in terms of inter-object proximities, multidimensional scaling may be employed to generate a configuration of objects in low-dimensional space in such a way as to preserve high-dimensional relationships. An algorithm is presented here for a heuristic hybrid model for the generation of such configurations. Building on a model introduced in 2002, the algorithm functions by means of sampling, spring model and interpolation phases. The most computationally complex stage of the original algorithm involved the execution of a series of nearest-neighbour searches. In this paper, we describe how the complexity of this phase has been reduced by treating all high-dimensional relationships as a set of discretised distances to a constant number of randomly selected items: pivots. In improving this computational bottleneck, the algorithmic complexity is reduced from <i>O(N√N)</i> to <i>O</i>(<i>N</i><sup>5/4</sup>). As well as documenting this improvement, the paper describes evaluation with a data set of 108,000 13-dimensional items and a set of 23,141 17-dimensional items. Results illustrate that the reduction in complexity is reflected in significantly improved run times and that no negative impact is made upon the quality of layout produced.
Information Visualization	Mapping nominal values to numbers for effective visualization.	Geraldine E. Rosario,Elke A. Rundensteiner,David C. Brown,Matthew O. Ward,Shiping Huang	2004	Data sets with a large numbers of nominal variables, including some with large number of distinct values, are becoming increasingly common and need to be explored. Unfortunately, most existing visual exploration tools are designed to handle numeric variables only. When importing data sets with nominal values into such visualization tools, most solutions to date are rather simplistic. Often, techniques that map nominal values to numbers do not assign order or spacing among the values in a manner that conveys semantic relationships. Moreover, displays designed for nominal variables usually cannot handle high cardinality variables well. This paper addresses the problem of how to display nominal variables in general-purpose visual exploration tools designed for numeric variables. Specifically, we investigate (1) how to assign order and spacing among the nominal values, and (2) how to reduce the number of distinct values to display. We propose a new technique, called the Distance-Quantification-Classing (DQC) approach, to preprocess nominal variables before being imported into a visual exploration tool. In the Distance Step, we identify a set of independent dimensions that can be used to calculate the distance between nominal values. In the Quantification Step, we use the independent dimensions and the distance information to assign order and spacing among the nominal values. In the Classing Step, we use results from the previous steps to determine which values within the domain of a variable are similar to each other and thus can be grouped together. Each step in the DQC approach can be accomplished by a variety of techniques. We extended the XmdvTool package to incorporate this approach. We evaluated our approach on several data sets using a variety of measures.
Information Visualization	Visualizing roles of variables in program animation.	Jorma Sajaniemi,Marja Kuittinen	2004	Visualizations may be used to assist in learning elementary computer programming - a skill that is hard to acquire by many students. This article presents a program animation system, PlanAni, that is based on the concept of the roles of variables. Roles represent schematic uses of variables that occur in programs over and over again, and a set of nine roles covers practically all variables in novice-level programs. PlanAni has been evaluated in a classroom experiment comparing traditional teaching with role-based teaching and animation. The results suggest that the introduction of roles provides students with a new conceptual framework that enables them to mentally process program information in a way similar to that of good code comprehenders. The use of role-based animation seems to assist in the adoption of role knowledge and expert-like programming strategies. A semi-structured interview with the teacher indicates that students like to work with the animator and that the system clarifies many concepts in programming.
IEEE Trans. Vis. Comput. Graph.	Fast Evolution of Image Manifolds and Application to Filtering and Segmentation in 3D Medical Images.	Thomas Deschamps,Ravi Malladi,Igor Ravve	2004	"In many instances, numerical integration of space-scale PDEs is the most time consuming operation of image processing. This is because the scale step is limited by conditional stability of explicit schemes. In this work, we introduce the unconditionally stable semi-implicit linearized difference scheme that is fashioned after additive operator split (AOS) [1], [2] for Beltrami and the subjective surface computation. The Beltrami flow [3], [4], [5] is one of the most effective denoising algorithms in image processing. For gray-level images, we show that the flow equation can be arranged in an advection-diffusion form, revealing the edge-enhancing properties of this flow. This also suggests the application of AOS method for faster convergence. The subjective surface [6] deals with constructing a perceptually meaningful interpretation from partial image data by mimicking the human visual system. However, initialization of the surface is critical for the final result and its main drawbacks are very slow convergence and the huge number of iterations required. In this paper, we first show that the governing equation for the subjective surface flow can be rearranged in an AOS implementation, providing a near real-time solution to the shape completion problem in 2D and 3D. Then, we devise a new initialization paradigm where we first ""condition¿ the viewpoint surface using the Fast-Marching algorithm. We compare the original method with our new algorithm on several examples of real 3D medical images, thus revealing the improvement achieved."
IEEE Trans. Vis. Comput. Graph.	"Editor's Note."	David S. Ebert	2004	"Editor's Note."
IEEE Trans. Vis. Comput. Graph.	Image-Based Collision Detection for Deformable Cloth Models.	George Baciu,Wingo Sai-Keung Wong	2004	Modeling the natural interaction of cloth and garments with objects in a 3D environment is currently one of the most computationally demanding tasks. These highly deformable materials are subject to a very large number of contact points in the proximity of other moving objects. Furthermore, cloth objects often fold, roll, and drape within themselves, generating a large number of self-collision areas. The interactive requirements of 3D games and physically driven virtual environments make the cloth collisions and self-collision computations more challenging. By exploiting mathematically well-defined smoothness conditions over smaller patches of deformable surfaces and resorting to image-based collision detection tests, we developed an efficient collision detection method that achieves interactive rates while tracking self-interactions in highly deformable surfaces consisting of a large number of elements. The method makes use of a novel technique for dynamically generating a hierarchy of cloth bounding boxes in order to perform object-level culling and image-based intersection tests using conventional graphics hardware support. An efficient backward voxel-based AABB hierarchy method is proposed to handle deformable surfaces which are highly compressed.
IEEE Trans. Vis. Comput. Graph.	"Editor's Note."	David S. Ebert	2004	reg
IEEE Trans. Vis. Comput. Graph.	Counting Cases in Substitope Algorithms.	David C. Banks,Stephen A. Linton,Paul K. Stockmeyer	2004	Counting Cases in Substitope Algorithms.
IEEE Trans. Vis. Comput. Graph.	Faster Shading by Equal Angle Interpolation of Vectors.	Tony Barrera,Anders Hast,Ewert Bengtsson	2004	"Abstract--In this paper, we show how spherical linear interpolation can be used to produce shading with a quality at least similar to Phong shading at a computational effort in the inner loop that is close to that of the Gouraud method. We show how to use the Chebyshev's recurrence relation in order to compute the shading very efficiently. Furthermore, it can also be used to interpolate vectors in such a way that normalization is not necessary, which will make the interpolation very fast. The somewhat larger setup effort required by this approach can be handled through table look up techniques."
IEEE Trans. Vis. Comput. Graph.	Multiresolution Indexing of Triangulated Irregular Networks.	John J. Bartholdi III,Paul Goldsman	2004	Multiresolution Indexing of Triangulated Irregular Networks.
IEEE Trans. Vis. Comput. Graph.	Real-Time Finite Element Modeling for Surgery Simulation: An Application to Virtual Suturing.	Jeffrey Berkley,George Turkiyyah,Daniel Berg,Mark A. Ganter,Suzanne Weghorst	2004	Abstract--Real-time finite element (FE) analysis can be used to represent complex deformable geometries in virtual environments. The need for accurate surgical simulation has spurred the development of many of the new real-time FE methodologies that enable haptic support and real-time deformation. These techniques are computationally intensive and it has proved to be a challenge to achieve the high modeling resolutions required to accurately represent complex anatomies. The authors present a new real-time methodology based on linear FE analysis that is appropriate for a wide range of surgical simulation applications. A methodology is proposed that is characterized by high model resolution, low preprocessing time, unrestricted multipoint surface contact, and adjustable boundary conditions. These features make the method ideal for modeling suturing, which is an element common to almost every surgical procedure. This paper describes constraints in the context of a Suturing Simulator currently being developed by the authors.
IEEE Trans. Vis. Comput. Graph.	Generalized B-Spline Subdivision-Surface Wavelets for Geometry Compression.	Martin Bertram,Mark A. Duchaineau,Bernd Hamann,Kenneth I. Joy	2004	Abstract--We present a new construction of lifted biorthogonal wavelets on surfaces of arbitrary two-manifold topology for compression and multiresolution representation. Our method combines three approaches: subdivision surfaces of arbitrary topology, B-spline wavelets, and the lifting scheme for biorthogonal wavelet construction. The simple building blocks of our wavelet transform are local lifting operations performed on polygonal meshes with subdivision hierarchy. Starting with a coarse, irregular polyhedral base mesh, our transform creates a subdivision hierarchy of meshes converging to a smooth limit surface. At every subdivision level, geometric detail can be expanded from wavelet coefficients and added to the surface. We present wavelet constructions for bilinear, bicubic, and biquintic B-Spline subdivision. While the bilinear and bicubic constructions perform well in numerical experiments, the biquintic construction turns out to be unstable. For lossless compression, our transform can be computed in integer arithmetic, mapping integer coordinates of control points to integer wavelet coefficients. Our approach provides a highly efficient and progressive representation for complex geometries of arbitrary topology.
IEEE Trans. Vis. Comput. Graph.	Image-Space Visibility Ordering for Cell Projection Volume Rendering of Unstructured Data.	Richard Cook,Nelson L. Max,Cláudio T. Silva,Peter L. Williams	2004	"Projection methods for volume rendering unstructured data work by projecting, in visibility order, the polyhedral cells of the mesh onto the image plane, and incrementally compositing each cell's color and opacity into the final image. Normally, such methods require an algorithm to determine a visibility order of the cells. The Meshed Polyhedra Visibility Order (MPVO) algorithm can provide such an order for convex meshes by considering the implications of local ordering relations between cells sharing a common face. However, in nonconvex meshes, one must also consider ordering relations along viewing rays which cross empty space between cells. In order to include these relations, the algorithm described in this paper, the scanning exact meshed polyhedra visibility ordering (SXMPVO) algorithm, scan-converts the exterior faces of the mesh and saves the ray-face intersections in an A-Buffer data structure which is then used for retrieving the extra ordering relations. The image which SXMPVO produces is the same as would be produced by ordering the cells exactly, even though SXMPVO does not compute an exact visibility ordering. This is because the image resolution used for computing the visibility ordering relations is the same as that which is used for the actual volume rendering and we choose our A-Buffer rays at the same sample points that are used to establish a polygon's pixel coverage during hardware scan conversion. Thus, the algorithm is image-space correct. The SXMPVO algorithm has several desirable features; among them are speed, simplicity of implementation, and no extra (i.e., with respect to MPVO) preprocessing."
IEEE Trans. Vis. Comput. Graph.	O-Buffer: A Framework for Sample-Based Graphics.	Huamin Qu,Arie E. Kaufman	2004	O-Buffer: A Framework for Sample-Based Graphics.
IEEE Trans. Vis. Comput. Graph.	Isosurface Construction in Any Dimension Using Convex Hulls.	Praveen Bhaniramka,Rephael Wenger,Roger Crawfis	2004	"Abstract--We present an algorithm for constructing isosurfaces in any dimension. The input to the algorithm is a set of scalar values in a d-dimensional regular grid of (topological) hypercubes. The output is a set of (d-1)-dimensional simplices forming a piecewise linear approximation to the isosurface. The algorithm constructs the isosurface piecewise within each hypercube in the grid using the convex hull of an appropriate set of points. We prove that our algorithm correctly produces a triangulation of a (d-1)-manifold with boundary. In dimensions three and four, lookup tables with 2^8 and 2^{16} entries, respectively, can be used to speed the algorithm's running time. In three dimensions, this gives the popular Marching Cubes algorithm. We discuss applications of four-dimensional isosurface construction to time varying isosurfaces, interval volumes, and morphing."
IEEE Trans. Vis. Comput. Graph.	Visualizing a Sphere Eversion.	George Francis,John M. Sullivan	2004	The mathematical process of everting a sphere (turning it inside-out allowing self-intersections) is a grand challenge for visualization because of the complicated, ever-changing internal structure. We have computed an optimal minimax eversion, requiring the least bending energy. Here, we discuss techniques we used to help visualize this eversion for visitors to virtual environments and viewers of our video The Optiverse.
IEEE Trans. Vis. Comput. Graph.	Binary-Space-Partitioned Images for Resolving Image-Based Visibility.	Chi-Wing Fu,Tien-Tsin Wong,Wai-Shun Tong,Chi-Keung Tang,Andrew J. Hanson	2004	Abstract-We propose a novel 2D representation for 3D visibility sorting, the Binary-Space-Partitioned Image (BSPI), to accelerate real-time image-based rendering. BSPI is an efficient 2D realization of a 3D BSP tree, which is commonly used in computer graphics for time-critical visibility sorting. Since the overall structure of a BSP tree is encoded in a BSPI, traversing a BSPI is comparable to traversing the corresponding BSP tree. BSPI performs visibility sorting efficiently and accurately in the 2D image space by warping the reference image triangle-by-triangle instead of pixel-by-pixel. Multiple BSPIs can be combined to solve ¿disocclusion,¿ when an occluded portion of the scene becomes visible at a novel viewpoint. Our method is highly automatic, including a tensor voting preprocessing step that generates candidate image partition lines for BSPIs, filters the noisy input data by rejecting outliers, and interpolates missing information. Our system has been applied to a variety of real data, including stereo, motion, and range images.
IEEE Trans. Vis. Comput. Graph.	Augmented Scene Modeling and Visualization by Optical and Acoustic Sensor Integration.	Andrea Fusiello,Vittorio Murino	2004	In this paper, underwater scene modeling from multisensor data is addressed. Acoustic and optical devices aboard an underwater vehicle are used to sense the environment in order to produce an output that is readily understandable even by an inexperienced operator. The main idea is to integrate multiple-sensor data by geometrically registering such data to a model. The geometrical structure of this model is a priori known but not ad hoc designed for this purpose. As a result, the vehicle pose is derived and model objects can be superimposed upon actual images, thus generating an augmented-reality representation. Results on a real underwater scene are reported, showing the effectiveness of the proposed approach.
IEEE Trans. Vis. Comput. Graph.	A Topological Hierarchy for Functions on Triangulated Surfaces.	Peer-Timo Bremer,Herbert Edelsbrunner,Bernd Hamann,Valerio Pascucci	2004	A Topological Hierarchy for Functions on Triangulated Surfaces.
IEEE Trans. Vis. Comput. Graph.	Combining Hierarchy and Energy Drawing Directed Graphs.	Liran Carmel,David Harel,Yehuda Koren	2004	Abstract- We present an algorithm for drawing directed graphs which is based on rapidly solving a unique one-dimensional optimization problem for each of the axes. The algorithm results in a clear description of the hierarchy structure of the graph. Nodes are not restricted to lie on fixed horizontal layers, resulting in layouts that convey the symmetries of the graph very naturally. The algorithm can be applied without change to cyclic or acyclic digraphs and even to graphs containing both directed and undirected edges. We also derive a hierarchy index from the input digraph, which quantitatively measures its amount of hierarchy.
IEEE Trans. Vis. Comput. Graph.	Footprint Area Sampled Texturing.	Baoquan Chen,Frank Dachille,Arie E. Kaufman	2004	Abstract--We study texture projection based on a four region subdivision: magnification, minification, and two mixed regions. We propose improved versions of existing techniques by providing exact filtering methods which reduce both aliasing and overblurring, especially in the mixed regions. We further present a novel texture mapping algorithm called FAST (Footprint Area Sampled Texturing), which not only delivers high quality, but also is efficient. By utilizing coherence between neighboring pixels, performing prefiltering, and applying an area sampling scheme, we guarantee a minimum number of samples sufficient for effective antialiasing. Unlike existing methods (e.g., MIP-map, Feline), our method adapts the sampling rate in each chosen MIP-map level separately to avoid undersampling in the lower level l for effective antialiasing and to avoid oversampling in the higher level l+1 for efficiency. Our method has been shown to deliver superior image quality to Feline and other methods while retaining the same efficiency. We also provide implementation trade offs to apply a variable degree of accuracy versus speed.
IEEE Trans. Vis. Comput. Graph.	Adaptive Extraction of Time-Varying Isosurfaces.	Benjamin F. Gregorski,Joshua G. Senecal,Mark A. Duchaineau,Kenneth I. Joy	2004	We present an algorithm for adaptively extracting and rendering isosurfaces from compressed time-varying volume data sets. Tetrahedral meshes defined by longest edge bisection are used to create a multiresolution representation of the volume in the spatial domain that is adapted over time to approximate the time-varying volume. The reextraction of the isosurface at each time step is accelerated with the vertex programming capabilities of modern graphics hardware. A data layout scheme which follows the access pattern indicated by mesh refinement is used to access the volume in a spatially and temporally coherent manner. This data layout scheme allows our algorithm to be used for out-of-core visualization.
IEEE Trans. Vis. Comput. Graph.	Point-Based Probabilistic Surfaces to Show Surface Uncertainty.	Gevorg Grigoryan,Penny Rheingans	2004	Efficient and informative visualization of surfaces with uncertainties is an important topic with many applications in science and engineering. In these applications, the correct course of action may depend not only on the location of a boundary, but on the precision with which that location is known. Examples include environmental pollution borderline detection, oil basin edge characterization, or discrimination between cancerous and healthy tissue in medicine. This paper presents a method for producing visualizations of surfaces with uncertainties using points as display primitives. Our approach is to render the surface as a collection of points and to displace each point from its original location along the surface normal by an amount proportional to the uncertainty at that point. This approach can be used in combination with other techniques such as pseudocoloring to produce efficient and revealing visualizations. The basic approach is sufficiently flexible to allow natural extensions; we show incorporation of expressive modulation of opacity, change of the stroke primitive, and addition of an underlying polygonal model. The method is used to visualize real and simulated tumor formations with uncertainty of tumor boundaries. The point-based technique is compared to pseudocoloring for a position estimation task in a preliminary user study.
IEEE Trans. Vis. Comput. Graph.	Selective Refinement Queries for Volume Visualization of Unstructured Tetrahedral Meshes.	Paolo Cignoni,Leila De Floriani,Paola Magillo,Enrico Puppo,Roberto Scopigno	2004	Abstract- In this paper, we address the problem of the efficient visualization of large irregular volume data sets by exploiting a multiresolution model based on tetrahedral meshes. Multiresolution models, also called Level-Of-Detail (LOD) models, allow encoding the whole data set at a virtually continuous range of different resolutions. We have identified a set of queries for extracting meshes at variable resolution from a multiresolution model, based on field values, domain location, or opacity of the transfer function. Such queries allow trading off between resolution and speed in visualization. We define a new compact data structure for encoding a multiresolution tetrahedral mesh built through edge collapses to support selective refinement efficiently and show that such a structure has a storage cost from 3 to 5.5 times lower than standard data structures used for tetrahedral meshes. The data structures and variable resolution queries have been implemented together with state-of-the art visualization techniques in a system for the interactive visualization of three-dimensional scalar fields defined on tetrahedral meshes. Experimental results show that selective refinement queries can support interactive visualization of large data sets.
IEEE Trans. Vis. Comput. Graph.	Robust Feature Detection and Local Classification for Surfaces Based on Moment Analysis.	Ulrich Clarenz,Martin Rumpf,Alexandru Telea	2004	The stable local classification of discrete surfaces with respect to features such as edges and corners or concave and convex regions, respectively, is as quite difficult as well as indispensable for many surface processing applications. Usually, the feature detection is done via a local curvature analysis. If concerned with large triangular and irregular grids, e.g., generated via a marching cube algorithm, the detectors are tedious to treat and a robust classification is hard to achieve. Here, a local classification method on surfaces is presented which avoids the evaluation of discretized curvature quantities. Moreover, it provides an indicator for smoothness of a given discrete surface and comes together with a built-in multiscale. The proposed classification tool is based on local zero and first moments on the discrete surface. The corresponding integral quantities are stable to compute and they give less noisy results compared to discrete curvature quantities. The stencil width for the integration of the moments turns out to be the scale parameter. Prospective surface processing applications are the segmentation on surfaces, surface comparison, and matching and surface modeling. Here, a method for feature preserving fairing of surfaces is discussed to underline the applicability of the presented approach.
IEEE Trans. Vis. Comput. Graph.	Robust Linear Dimensionality Reduction.	Yehuda Koren,Liran Carmel	2004	Robust Linear Dimensionality Reduction.
IEEE Trans. Vis. Comput. Graph.	"Guest Editor's Introduction: Special Section on Mathematics and Visualization."	Hans-Christian Hege,Konrad Polthier	2004	"Guest Editor's Introduction: Special Section on Mathematics and Visualization."
IEEE Trans. Vis. Comput. Graph.	Visualization of Vector Fields Using Seed LIC and Volume Rendering.	Anders Helgeland,Oyvind Andreassen	2004	Line Integral Convolution (LIC) is a powerful texture-based technique for visualizing vector fields. Due to the high computational expense of generating the 3D textures and the difficulties of effectively displaying the result, LIC has most commonly been used to depict vector fields in 2D or over a surface in 3D. Here, we propose new methods for more effective volume visualization of three-dimensional vector fields using LIC: 1) We present a fast method for computing volume LIC textures that exploits the sparsity of the input texture. 2) We propose the use of a shading technique, called limb darkening, to reveal the depth relations among the field lines. The shading effect is obtained simply by using appropriate transfer functions and, therefore, avoids using expensive shading techniques. 3) We demonstrate how two-field visualization techniques can be used to enhance the visual information describing a vector field. The volume LIC textures are rendered using texture-based rendering techniques, which allows interactive exploration of a vector field.
IEEE Trans. Vis. Comput. Graph.	ISA and IBFVS: Image Space-Based Visualization of Flow on Surfaces.	Robert S. Laramee,Jarke J. van Wijk,Bruno Jobard,Helwig Hauser	2004	We present a side-by-side analysis of two recent image space approaches for the visualization of vector fields on surfaces. The two methods, Image Space Advection (ISA) and Image-Based Flow Visualization for Curved Surfaces (IBFVS) generate dense representations of time-dependent vector fields with high spatio-temporal correlation. While the 3D vector fields are associated with arbitrary surfaces represented by triangular meshes, the generation and advection of texture properties is confined to image space. Fast frame rates are achieved by exploiting frame-to-frame coherency and graphics hardware. In our comparison of ISA and IBFVS, we point out the strengths and weaknesses of each approach and give recommendations as to when and where they are best applied.
IEEE Trans. Vis. Comput. Graph.	Virtual Trackballs Revisited.	Knud Henriksen,Jon Sporring,Kasper Hornbæk	2004	"Abstract--Rotation of three-dimensional objects by a two-dimensional mouse is a typical task in computer-aided design, operation simulations, and desktop virtual reality. The most commonly used rotation technique is a virtual trackball surrounding the object and operated by the mouse pointer. This article reviews and provides a mathematical foundation for virtual trackballs. The first, but still popular, virtual trackball was described by Chen et al. [CHECK END OF SENTENCE]. We show that the virtual trackball by Chen et al. does not rotate the object along the intended great circular arc on the virtual trackball and we give a correction. Another popular virtual trackball is Shoemake's quaternion implementation [CHECK END OF SENTENCE], which we show to be a special case of the virtual trackball by Chen et al.. Shoemake extends the scope of the virtual trackball to the full screen. Unfortunately, Shoemake's virtual trackball is inhomogeneous and discontinuous with consequences for usability. Finally, we review Bell's virtual trackball [CHECK END OF SENTENCE] and discuss studies of the usability of virtual trackballs."
IEEE Trans. Vis. Comput. Graph.	High-Quality Animation of 2D Steady Vector Fields.	Wilfrid Lefer,Bruno Jobard,Claire Leduc	2004	Abstract--Simulators for dynamic systems are now widely used in various application areas and raise the need for effective and accurate flow visualization techniques. Animation allows us to depict direction, orientation, and velocity of a vector field accurately. This paper extends a former proposal for a new approach to produce perfectly cyclic and variable-speed animations for 2D steady vector fields (see [1] and [2]). A complete animation of an arbitrary number of frames is encoded in a single image. The animation can be played using the color table animation technique, which is very effective even on low-end workstations. A cyclic set of textures can be produced as well and then encoded in a common animation format or used for texture mapping on 3D objects. As compared to other approaches, the method presented in this paper produces smoother animations and is more effective, both in memory requirements to store the animation, and in computation time.
IEEE Trans. Vis. Comput. Graph.	Haptics-Based Dynamic Implicit Solid Modeling.	Jing Hua,Hong Qin	2004	This paper systematically presents a novel, interactive solid modeling framework, Haptics-based Dynamic Implicit Solid Modeling, which is founded upon volumetric implicit functions and powerful physics-based modeling. In particular, we augment our modeling framework with a haptic mechanism in order to take advantage of additional realism associated with a 3D haptic interface. Our dynamic implicit solids are semi-algebraic sets of volumetric implicit functions and are governed by the principles of dynamics, hence responding to sculpting forces in a natural and predictable manner. In order to directly manipulate existing volumetric data sets as well as point clouds, we develop a hierarchical fitting algorithm to reconstruct and represent discrete data sets using our continuous implicit functions, which permit users to further design and edit those existing 3D models in real-time using a large variety of haptic and geometric toolkits, and visualize their interactive deformation at arbitrary resolution. The additional geometric and physical constraints afford more sophisticated control of the dynamic implicit solids. The versatility of our dynamic implicit modeling enables the user to easily modify both the geometry and the topology of modeled objects, while the inherent physical properties can offer an intuitive haptic interface for direct manipulation with force feedback.
IEEE Trans. Vis. Comput. Graph.	A Streaming Narrow-Band Algorithm: Interactive Computation and Visualization of Level Sets.	Aaron E. Lefohn,Joe Michael Kniss,Charles D. Hansen,Ross T. Whitaker	2004	Deformable isosurfaces, implemented with level-set methods, have demonstrated a great potential in visualization and computer graphics for applications such as segmentation, surface processing, and physically-based modeling. Their usefulness has been limited, however, by their high computational cost and reliance on signi&pound;cant parameter tuning. This paper presents a solution to these challenges by describing graphics processor (GPU) based algorithms for solving and visualizing level-set solutions at interactive rates. The proposed solution is based on a new, streaming implementation of the narrow-band algorithm. The new algorithm packs the level-set isosurface data into 2D texture memory via a multi-dimensional virtual memory system. As the level-set moves, this texture-based representation is dynamically updated via a novel GPU-to-CPU message passing scheme. By integrating the level-set solver with a real-time volume renderer, a user can visualize and intuitively steer the level-set surface as it evolves. We demonstrate the capabilities of this technology for interactive volume segmentation and visualization.
IEEE Trans. Vis. Comput. Graph.	"Applications of Forman's Discrete Morse Theory to Topology Visualization and Mesh Compression."	Thomas Lewiner,Hélio Lopes,Geovan Tavares	2004	"Morse theory is a powerful tool for investigating the topology of smooth manifolds. It has been widely used by the computational topology, computer graphics, and geometric modeling communities to devise topology-based algorithms and data structures. Forman introduced a discrete version of this theory which is purely combinatorial. This work aims to build, visualize, and apply the basic elements of Forman's discrete Morse theory. It intends to use some of those concepts to visually study the topology of an object. As a basis, an algorithmic construction of optimal Forman's discrete gradient vector fields is provided. This construction is then used to topologically analyze mesh compression schemes, such as Edgebreaker and Grow&Fold. In particular, this paper proves that the complexity class of the strategy optimization of Grow&Fold is MAX-SNP hard."
IEEE Trans. Vis. Comput. Graph.	Camera-Sampling Field and Its Applications.	Ping-Hsien Lin,Tong-Yee Lee	2004	Abstract--In this paper, we propose a novel vector field, called a camera-sampling field, to represent the sampling density distribution of a pinhole camera. We give the derivation and discuss some essential properties of the camera-sampling field, including flux, divergence, curl, gradient, level surface, and sampling patterns. This vector field reveals camera-sampling concisely and facilitates camera sampling analysis. The usage for this vector field in several computer graphics applications is introduced, such as determining the splat kernel for image-based rendering, texture filtering, mipmap level selection, level transition criteria for LOD, and LDI-construction.
IEEE Trans. Vis. Comput. Graph.	Hierarchical Data Visualization Using a Fast Rectangle-Packing Algorithm.	Takayuki Itoh,Yumi Yamaguchi,Yuko Ikehata,Yasumasa Kajinaga	2004	Abstract--This paper presents a technique for the representation of large-scale hierarchical data which aims to provide good overviews of complete structures and the content of the data in one display space. The technique represents the data by using nested rectangles. It first packs icons or thumbnails of the lowest-level data and then generates rectangular borders that enclose the packed data. It repeats the process of generating rectangles that enclose the lower-level rectangles until the highest-level rectangles are packed. This paper presents two rectangle-packing algorithms for placing items of hierarchical data onto display spaces. The algorithms refer to Delaunay triangular meshes connecting the centers of rectangles to find gaps where rectangles can be placed. The first algorithm places rectangles where they do not overlap each other and where the extension of the layout area is minimal. The second algorithm places rectangles by referring to templates describing the ideal positions for nodes of input data. It places rectangles where they do not overlap each other and where the combination of the layout area and the distances between the positions described in the template and the actual positions is minimal. It can smoothly represent time-varying data by referring to templates that describe previous layout results. It is also suitable for semantics-based or design-based data layout by generating templates according to the semantics or design.
IEEE Trans. Vis. Comput. Graph.	Synthesis and Rendering of Bidirectional Texture Functions on Arbitrary Surfaces.	Xinguo Liu,Yaohua Hu,Jingdan Zhang,Xin Tong,Baining Guo,Heung-Yeung Shum	2004	Abstract--The bidirectional texture function (BTF) is a 6D function that describes the appearance of a real-world surface as a function of lighting and viewing directions. The BTF can model the fine-scale shadows, occlusions, and specularities caused by surface mesostructures. In this paper, we present algorithms for efficient synthesis of BTFs on arbitrary surfaces and for hardware-accelerated rendering. For both synthesis and rendering, a main challenge is handling the large amount of data in a BTF sample. To addresses this challenge, we approximate the BTF sample by a small number of 4D point appearance functions (PAFs) multiplied by 2D geometry maps. The geometry maps and PAFs lead to efficient synthesis and fast rendering of BTFs on arbitrary surfaces. For synthesis, a surface BTF can be generated by applying a texton-based sysnthesis algorithm to a small set of 2D geometry maps while leaving the companion 4D PAFs untouched. As for rendering, a surface BTF synthesized using geometry maps is well-suited for leveraging the programmable vertex and pixel shaders on the graphics hardware. We present a real-time BTF rendering algorithm that runs at the speed of about 30 frames/second on a mid-level PC with an ATI Radeon 8500 graphics card. We demonstrate the effectiveness of our synthesis and rendering algorithms using both real and synthetic BTF samples.
IEEE Trans. Vis. Comput. Graph.	Camera-Based Detection and Removal of Shadows from Interactive Multiprojector Displays.	Christopher O. Jaynes,Stephen B. Webb,R. Matt Steele	2004	Abstract--Front-projection displays are a cost-effective and increasingly popular method for large format visualization and immersive rendering of virtual models. New approaches to projector tiling, automatic calibration, and color balancing have made multiprojector display systems feasible without undue infrastructure changes and maintenance. As a result, front-projection displays are being used to generate seamless, visually immersive worlds for virtual reality and visualization applications with reasonable cost and maintenance overhead. However, these systems suffer from a fundamental problem: Users and other objects in the environment can easily and inadvertently block projectors, creating shadows on the displayed image. Shadows occlude potentially important information and detract from the sense of presence an immersive display may have conveyed. We introduce a technique that detects and corrects shadows in a multiprojector display while it is in use. Cameras observe the display and compare observations with an expected image to detect shadowed regions. These regions are transformed to the appropriate projector frames, where corresponding pixel values are increased and/or attenuated. In display regions where more than one projector contributes to the image, shadow regions are eliminated.
IEEE Trans. Vis. Comput. Graph.	"Guest Editor's Introduction: Special Section on InfoVis."	Daniel A. Keim,Tamara Munzner,Stephen C. North	2004	"Guest Editor's Introduction: Special Section on InfoVis."
IEEE Trans. Vis. Comput. Graph.	CartoDraw: A Fast Algorithm for Generating Contiguous Cartograms.	Daniel A. Keim,Stephen C. North,Christian Panse	2004	"Abstract- Cartograms are a well-known technique for showing geography-related statistical information, such as population demographics and epidemiological data. The basic idea is to distort a map by resizing its regions according to a statistical parameter, but in a way that keeps the map recognizable. In this study, we formally define a family of cartogram drawing problems. We show that even simple variants are unsolvable in the general case. Because the feasible variants are NP-complete, heuristics are needed to solve the problem. Previously proposed solutions suffer from problems with the quality of the generated drawings. For a cartogram to be recognizable, it is important to preserve the global shape or outline of the input map, a requirement that has been overlooked in the past. To address this, our objective function for cartogram drawing includes both global and local shape preservation. To measure the degree of shape preservation, we propose a shape similarity function, which is based on a Fourier transformation of the polygons' curvatures. Also, our application is visualization of dynamic data, for which we need an algorithm that recalculates a cartogram in a few seconds. None of the previous algorithms provides adequate performance with an acceptable level of quality for this application. In this paper, we therefore propose an efficient iterative scanline algorithm to reposition edges while preserving local and global shapes. Scanlines may be generated automatically or entered interactively to guide the optimization process more closely. We apply our algorithm to several example data sets and provide a detailed comparison of the two variants of our algorithm and previous approaches."
IEEE Trans. Vis. Comput. Graph.	Topological Segmentation in Three-Dimensional Vector Fields.	Karim Mahrous,Janine Bennett,Gerik Scheuermann,Bernd Hamann,Kenneth I. Joy	2004	Abstract--We present a new method for topological segmentation in steady three-dimensional vector fields. Depending on desired properties, the algorithm replaces the original vector field by a derived segmented data set, which is utilized to produce separating surfaces in the vector field. We define the concept of a segmented data set, develop methods that produce the segmented data by sampling the vector field with streamlines, and describe algorithms that generate the separating surfaces. This method is applied to generate local separatrices in the field, defined by a movable boundary region placed in the field. The resulting partitions can be visualized using standard techniques for a visualization of a vector field at a higher level of abstraction.
IEEE Trans. Vis. Comput. Graph.	Color Nonuniformity in Projection-Based Displays: Analysis and Solutions.	Aditi Majumder,Rick Stevens	2004	Abstract--Large-area displays made up of several projectors show significant variation in color. In this paper, we identify different projector parameters that cause the color variation and study their effects on the luminance and chrominance characteristics of the display. This work leads to the realization that luminance varies significantly within and across projectors, while chrominance variation is relatively small, especially across projectors of same model. To address this situation, we present a method to achieve luminance matching across all pixels of a multiprojector display that results in photometrically uniform displays. We use a camera as a measurement device for this purpose. Our method comprises a one-time calibration step that generates a per channel per projector luminance attenuation map (LAM), which is then used to correct any image projected on the display at interactive rates on commodity graphics hardware. To the best of our knowledge, this is the first effort to match luminance across all the pixels of a multiprojector display.
IEEE Trans. Vis. Comput. Graph.	"Conveying Shape with Texture: Experimental Investigations of Texture's Effects on Shape Categorization Judgments."	Sunghee Kim,Haleh Hagh-Shenas,Victoria Interrante	2004	"Conveying Shape with Texture: Experimental Investigations of Texture's Effects on Shape Categorization Judgments."
IEEE Trans. Vis. Comput. Graph.	Incremental Penetration Depth Estimation between Convex Polytopes Using Dual-Space Expansion.	Young J. Kim,Ming C. Lin,Dinesh Manocha	2004	"Abstract--We present a fast algorithm to estimate the penetration depth between convex polytopes in 3D. The algorithm incrementally seeks a ""locally optimal solution¿ by walking on the surface of the Minkowski sums. The surface of the Minkowski sums is computed implicitly by constructing a local dual mapping on the Gauss map. We also present three heuristic techniques that are used to estimate the initial features used by the walking algorithm. We have implemented the algorithm and compared its performance with earlier approaches. In our experiments, the algorithm is able to estimate the penetration depth in about a milli-second on an 1 GHz Pentium PC. Moreover, its performance is almost independent of model complexity in environments with high coherence between successive instances."
IEEE Trans. Vis. Comput. Graph.	Simplification of Three-Dimensional Density Maps.	Vijay Natarajan,Herbert Edelsbrunner	2004	We consider scientific data sets that describe density functions over three-dimensional geometric domains. Such data sets are often large and coarsened representations are needed for visualization and analysis. Assuming a tetrahedral mesh representation, we construct such representations with a simplification algorithm that combines three goals: the approximation of the function, the preservation of the mesh topology, and the improvement of the mesh quality. The third goal is achieved with a novel extension of the well-known quadric error metric. We perform a number of computational experiments to understand the effect of mesh quality improvement on the density map approximation. In addition, we study the effect of geometric simplification on the topological features of the function by monitoring its critical points.
IEEE Trans. Vis. Comput. Graph.	v-Quaternion Splines for the Smooth Interpolation of Orientations.	Gregory M. Nielson	2004	v-Quaternion Splines for the Smooth Interpolation of Orientations.
IEEE Trans. Vis. Comput. Graph.	Efficient Implementation of Real-Time View-Dependent Multiresolution Meshing.	Renato Pajarola,Christopher DeCoro	2004	Abstract--In this paper, we present an efficient (topology preserving) multiresolution meshing framework for interactive level-of-detail (LOD) generation and rendering of large triangle meshes. More specifically, the presented approach, called FastMesh, provides view-dependent LOD generation and real-time mesh simplification that minimizes visual artifacts. Multiresolution triangle mesh representations are an important tool for reducing triangle mesh complexity in interactive rendering environments. Ideally, for interactive visualization, a triangle mesh is simplified to the maximal tolerated visible error and, thus, mesh simplification is view-dependent. This paper introduces an efficient hierarchical multiresolution triangulation framework based on a half-edge triangle mesh data structure and presents optimized implementations of several view-dependent or visual mesh simplification heuristics within that framework. Despite being optimized for performance, these error heuristics provide conservative error bounds. The presented framework is highly efficient both in space and time cost and needs only a fraction of the time required for rendering to perform the error calculations and dynamic mesh updates.
IEEE Trans. Vis. Comput. Graph.	Confetti: Object-Space Point Blending and Splatting.	Renato Pajarola,Miguel Sainz,Patrick Guidotti	2004	In this paper, we present Confetti, a novel point-based rendering approach based on object-space point interpolation of densely sampled surfaces. We introduce the concept of a transformation-invariant covariance matrix of a set of points which can efficiently be used to determine splat sizes in a multiresolution point hierarchy. We also analyze continuous point interpolation in object-space and we define a new class of parameterized blending kernels as well as a normalization procedure to achieve smooth blending. Furthermore, we present a hardware accelerated rendering algorithm based on texture mapping and \alpha{\hbox{-}}{\rm blending} as well as programmable vertex and pixel-shaders.
IEEE Trans. Vis. Comput. Graph.	An Intestinal Surgery Simulator: Real-Time Collision Processing and Visualization.	Laks Raghupathi,Laurent Grisoni,François Faure,Damien Marchal,Marie-Paule Cani,Christophe Chaillou	2004	This research work is aimed toward the development of a VR-based trainer for colon cancer removal. It enables the surgeons to interactively view and manipulate the concerned virtual organs as during a real surgery. First, we present a method for animating the small intestine and the mesentery (the tissue that connects it to the main vessels) in real-time, thus enabling user interaction through virtual surgical tools during the simulation. We present a stochastic approach for fast collision detection in highly deformable, self-colliding objects. A simple and efficient response to collisions is also introduced in order to reduce the overall animation complexity. Second, we describe a new method based on generalized cylinders for fast rendering of the intestine. An efficient curvature detection method, along with an adaptive sampling algorithm, is presented. This approach, while providing improved tessellation without the classical self-intersection problem, also allows for high-performance rendering thanks to the new 3D skinning feature available in recent GPUs. The rendering algorithm is also designed to ensure a guaranteed frame rate. Finally, we present the quantitative results of the simulations and describe the qualitative feedback obtained from the surgeons.
IEEE Trans. Vis. Comput. Graph.	Reconstruction of Volume Data with Quadratic Super Splines.	Christian Rössl,Frank Zeilfelder,Günther Nürnberger,Hans-Peter Seidel	2004	Reconstruction of Volume Data with Quadratic Super Splines.
IEEE Trans. Vis. Comput. Graph.	A Fast Impulsive Contact Suite for Rigid Body Simulation.	Harald Schmidl,Victor J. Milenkovic	2004	"Abstract--A suite of algorithms is presented for contact resolution in rigid body simulation under the Coulomb friction model: Given a set of rigid bodies with many contacts among them, resolve dynamic contacts (collisions) and static (persistent) contacts. The suite consists of four algorithms: 1) partial sequential collision resolution, 2) final resolution of collisions through the solution of a single convex QP (positive semidefinite quadratic program), 3) resolution of static contacts through the solution of a single convex QP, 4) freezing of ""stationary¿ bodies. This suite can generate realistic-looking results for simple examples yet, for the first time, can also tractably resolve contacts for a simulation as large as 1,000 cubes in an ""hourglass.¿ Freezing speeds up this simulation by more than 25 times. Thanks to excellent commercial QP technology, the contact resolution suite is simple to implement and can be ""plugged into¿ any simulation algorithm to provide fast and realistic-looking animations of rigid bodies."
IEEE Trans. Vis. Comput. Graph.	Fuzzy Vector Median-Based Surface Smoothing.	Yuzhong Shen,Kenneth E. Barner	2004	Abstract--This paper proposes a novel approach for smoothing surfaces represented by triangular meshes. The proposed method is a two-step procedure: surface normal smoothing through fuzzy vector median (FVM) filtering followed by integration of surface normals for vertex position update based on the least square error (LSE) criteria. Median and Order Statistic-based filters are extensively used in signal processing, especially image processing, due to their ability to reject outliers and preserve features such as edges and monotonic regions. More recently, fuzzy ordering theory has been introduced to allow averaging among similarly valued samples. Fuzzy ordering theory leads naturally to the fuzzy median, which yields improved noise smoothing over traditional crisp median filters. This paper extends the fuzzy ordering concept to vector-based data and introduces the fuzzy vector median filter. The application of FVM filters to surface normal smoothing yields improved results over previously introduced normal smoothing algorithms. The improved filtering results, coupled with LSE vertex position update, produces surface smoothing that minimizes the effects of noise while simultaneously preserving detail features. The proposed method is simple to implement and relatively fast. Simulation results are presented showing the performance of the proposed method and its advantages over commonly used surface smoothing algorithms. Additionally, optimization procedures for FVM filters are derived and evaluated.
IEEE Trans. Vis. Comput. Graph.	Interactive Visualization of Three-Dimensional Vector Fields with Flexible Appearance Control.	Han-Wei Shen,Guo-Shi Li,Udeepta Bordoloi	2004	Interactive Visualization of Three-Dimensional Vector Fields with Flexible Appearance Control.
IEEE Trans. Vis. Comput. Graph.	Crest Lines for Surface Segmentation and Flattening.	Georgios Stylianou,Gerald E. Farin	2004	We present a method for extracting feature curves called crest lines from a triangulated surface. Then, we calculate the geodesic Voronoi diagram of crest lines to segment the surface into several regions. Afterward, barycentric surface flattening using theory from graph embeddings is implemented and, using the Geodesic Voronoi diagram, we develop a faster surface flattening algorithm.
IEEE Trans. Vis. Comput. Graph.	Human Factors in Visualization Research.	Melanie Tory,Torsten Möller	2004	"Abstract--Visualization can provide valuable assistance for data analysis and decision making tasks. However, how people perceive and interact with a visualization tool can strongly influence their understanding of the data as well as the system's usefulness. Human factors therefore contribute significantly to the visualization process and should play an important role in the design and evaluation of visualization tools. Several research initiatives have begun to explore human factors in visualization, particularly in perception-based design. Nonetheless, visualization work involving human factors is in its infancy, and many potentially promising areas have yet to be explored. Therefore, this paper aims to 1) review known methodology for doing human factors research, with specific emphasis on visualization, 2) review current human factors research in visualization to provide a basis for future investigation, and 3) identify promising areas for future research."
IEEE Trans. Vis. Comput. Graph.	Wavelet-Based Multiresolution Analysis of Irregular Surface Meshes.	Sébastien Valette,Rémy Prost	2004	"Abstract--This paper extends Lounsbery's multiresolution analysis wavelet-based theory for triangular 3D meshes, which can only be applied to regularly subdivided meshes and thus involves a remeshing of the existing 3D data. Based on a new irregular subdivision scheme, the proposed algorithm can be applied directly to irregular meshes, which can be very interesting when one wants to keep the connectivity and geometry of the processed mesh completely unchanged. This is very convenient in CAD (Computer-Assisted Design), when the mesh has attributes such as texture and color information, or when the 3D mesh is used for simulations, and where a different connectivity could lead to simulation errors. The algorithm faces an inverse problem for which a solution is proposed. For each level of resolution, the simplification is processed in order to keep the mesh as regular as possible. In addition, a geometric criterion is used to keep the geometry of the approximations as close as possible to the original mesh. Several examples on various reference meshes are shown to prove the efficiency of our proposal."
IEEE Trans. Vis. Comput. Graph.	Wavelet-Based Progressive Compression Scheme for Triangle Meshes: Wavemesh.	Sébastien Valette,Rémy Prost	2004	Abstract--This paper proposes a new lossy to lossless progressive compression scheme for triangular meshes, based on a wavelet multiresolution theory for irregular 3D meshes. Although remeshing techniques obtain better compression ratios for geometric compression, this approach can be very effective when one wants to keep the connectivity and geometry of the processed mesh completely unchanged. The simplification is based on the solving of an inverse problem. Optimization of both the connectivity and geometry of the processed mesh improves the approximation quality and the compression ratio of the scheme at each resolution level. We show why this algorithm provides an efficient means of compression for both connectivity and geometry of 3D meshes and it is illustrated by experimental results on various sets of reference meshes, where our algorithm performs better than previously published approaches for both lossless and progressive compression.
IEEE Trans. Vis. Comput. Graph.	Comparative Flow Visualization.	Vivek Verma,Alex Pang	2004	There are many situations where one needs to compare two or more data sets. It may be to compare different models, different resolutions, differences in algorithms, different experimental results, etc. There is therefore a need for comparative visualization tools to help analyze the differences. This paper focuses on comparative visualization tools for analyzing flow or vector data sets. The techniques presented allow one to compare individual streamlines and streamribbons as well as a dense field of streamlines. These comparison methods can also be used to study differences in vortex cores that are represented as polylines.
IEEE Trans. Vis. Comput. Graph.	Interactive Stereoscopic Rendering of Volumetric Environments.	Ming Wan,Nan Zhang,Huamin Qu,Arie E. Kaufman	2004	Abstract--We present an efficient stereoscopic rendering algorithm supporting interactive navigation through large-scale 3D voxel-based environments. In this algorithm, most of the pixel values of the right image are derived from the left image by a fast 3D warping based on a specific stereoscopic projection geometry. An accelerated volumetric ray casting then fills the remaining gaps in the warped right image. Our algorithm has been parallelized on a multiprocessor by employing effective task partitioning schemes and achieved a high cache coherency and load balancing. We also extend our stereoscopic rendering to include view-dependent shading and transparency effects. We have applied our algorithm in two virtual navigation systems, flythrough over terrain and virtual colonoscopy, and reached interactive stereoscopic rendering rates of more than 10 frames per second on a 16-processor SGI Challenge.
IEEE Trans. Vis. Comput. Graph.	Efficient Example-Based Painting and Synthesis of 2D Directional Texture.	Bin Wang,Wenping Wang,Huaiping Yang,Jia-Guang Sun	2004	Abstract--We present a new method for converting a photo or image to a synthesized painting following the painting style of an example painting. Treating painting styles of brush strokes as sample textures, we reduce the problem of learning an example painting to a texture synthesis problem. The proposed method uses a hierarchical patch-based approach to the synthesis of directional textures. The key features of our method are: 1) Painting styles are represented as one or more blocks of sample textures selected by the user from the example painting; 2) image segmentation and brush stroke directions defined by the medial axis are used to better represent and communicate shapes and objects present in the synthesized painting; 3) image masks and a hierarchy of texture patches are used to efficiently synthesize high-quality directional textures. The synthesis process is further accelerated through texture direction quantization and the use of Gaussian pyramids. Our method has the following advantages: First, the synthesized stroke textures can follow a direction field determined by the shapes of regions to be painted. Second, the method is very efficient; the generation time of a synthesized painting ranges from a few seconds to about one minute, rather than hours, as required by other existing methods, on a commodity PC. Furthermore, the technique presented here provides a new and efficient solution to the problem of synthesizing a 2D directional texture. We use a number of test examples to demonstrate the efficiency of the proposed method and the high quality of results produced by the method.
IEEE Trans. Vis. Comput. Graph.	The Lattice-Boltzmann Method for Simulating Gaseous Phenomena.	Xiaoming Wei,Wei Li,Klaus Mueller,Arie E. Kaufman	2004	Abstract--We present a physically-based, yet fast and simple method to simulate gaseous phenomena. In our approach, the incompressible Navier-Stokes (NS) equations governing fluid motion have been modeled in a novel way to achieve a realistic animation. We introduce the Lattice Boltzmann Model (LBM), which simulates the microscopic movement of fluid particles by linear and local rules on a grid of cells so that the macroscopic averaged properties obey the desired NS equations. The LBM is defined on a 2D or 3D discrete lattice, which is used to solve fluid animation based on different boundary conditions. The LBM simulation generates, in real-time, an accurate velocity field and can incorporate an optional temperature field to account for the buoyancy force of hot gas. Because of the linear and regular operations in each local cell of the LBM grid, we implement the computation in commodity texture hardware, further improving the simulation speed. Finally, textured splats are used to add small scale turbulent details, achieving high-quality real-time rendering. Our method can also simulate the physically correct action of stationary or mobile obstacles on gaseous phenomena in real-time, while still maintaining highly plausible visual details.
IEEE Trans. Vis. Comput. Graph.	Lattice-Based Flow Field Modeling.	Xiaoming Wei,Ye Zhao,Zhe Fan,Wei Li,Feng Qiu,Suzanne Yoakum-Stover,Arie E. Kaufman	2004	We present an approach for simulating the natural dynamics that emerge from the interaction between a flow field and immersed objects. We model the flow field using the Lattice Boltzmann Model (LBM) with boundary conditions appropriate for moving objects and accelerate the computation on commodity graphics hardware (GPU) to achieve real-time performance. The boundary conditions mediate the exchange of momentum between the flow field and the moving objects resulting in forces exerted by the flow on the objects as well as the back-coupling on the flow. We demonstrate our approach using soap bubbles and a feather. The soap bubbles illustrate Fresnel reflection, reveal the dynamics of the unseen flow field in which they travel, and display spherical harmonics in their undulations. Our simulation allows the user to directly interact with the flow field to influence the dynamics in real time. The free feather flutters and gyrates in response to lift and drag forces created by its motion relative to the flow. Vortices are created as the free feather falls in an otherwise quiescent flow.
IEEE Trans. Vis. Comput. Graph.	On a Construction of a Hierarchy of Best Linear Spline Approximations Using a Finite Element Approach.	David F. Wiley,Martin Bertram,Bernd Hamann	2004	We present a method for the hierarchical approximation of functions in one, two, or three variables based on the finite element method (Ritz approximation). Starting with a set of data sites with associated function, we first determine a smooth (scattered-data) interpolant. Next, we construct an initial triangulation by triangulating the region bounded by the minimal subset of data sites defining the convex hull of all sites. We insert only original data sites, thus reducing storage requirements. For each triangulation, we solve a minimization problem: computing the best linear spline approximation of the interpolant of all data, based on a functional involving function values and first derivatives. The error of a best linear spline approximation is computed in a Sobolev-like norm, leading to element-specific error values. We use these interval/triangle/tetrahedron-specific values to identify the element to subdivide next. The subdivision of an element with largest error value requires the recomputation of all spline coefficients due to the global nature of the problem. We improve efficiency by 1) subdividing multiple elements simultaneously and 2) by using a sparse-matrix representation and system solver.
IEEE Trans. Vis. Comput. Graph.	Interactive Volume Rendering of Thin Thread Structures within Multivalued Scientific Data Sets.	Andreas Wenger,Daniel F. Keefe,Song Zhang,David H. Laidlaw	2004	We present a threads and halos representation for interactive volume rendering of vector-field structure and describe a number of additional components that combine to create effective visualizations of multivalued 3D scientific data. After filtering linear structures, such as flow lines, into a volume representation, we use a multilayer volume rendering approach to simultaneously display this derived volume along with other data values. We demonstrate the utility of threads and halos in clarifying depth relationships within dense renderings and we present results from two scientific applications: visualization of second-order tensor valued magnetic resonance imaging (MRI) data and simulated 3D fluid flow data. In both application areas, the interactivity of the visualizations proved to be important to the domain scientists. Finally, we describe a PC-based implementation of our framework along with domain specific transfer functions, including an exploratory data culling tool, that enable fast data exploration.
IEEE Trans. Vis. Comput. Graph.	"Guest Editor's Introduction: Special Section on IEEE Visualization."	Jarke J. van Wijk,Robert J. Moorhead II,Greg Turk	2004	"Guest Editor's Introduction: Special Section on IEEE Visualization."
IEEE Trans. Vis. Comput. Graph.	A Model for Smooth Viewing and Navigation of Large 2D Information Spaces.	Jarke J. van Wijk,Wim A. A. Nuij	2004	A Model for Smooth Viewing and Navigation of Large 2D Information Spaces.
IEEE Trans. Vis. Comput. Graph.	Image Registration Using Hierarchical B-Splines.	Zhiyong Xie,Gerald E. Farin	2004	Abstract--Hierarchical B-splines have been widely used for shape modeling since their discovery by Forsey and Bartels. In this paper, we present an application of this concept, in the form of free-form deformation, to image registration by matching two images at increasing levels of detail. Results using MRI brain data are presented that demonstrate high degrees of matching while unnecessary distortions are avoided. We compare our results with the nonlinear ICP (Iterative Closest Point) algorithm (used for landmark-based registration) and optical flow (used for intensity-based registration).
IEEE Trans. Vis. Comput. Graph.	Mesh Simplification with Hierarchical Shape Analysis and Iterative Edge Contraction.	Jingqi Yan,Pengfei Shi,David Zhang	2004	Abstract--This paper presents a novel mesh simplification algorithm. It decouples the simplification process into two phases: shape analysis and edge contraction. In the analysis phase, it imposes a hierarchical structure on a surface mesh by uniform hierarchical partitioning, marks the importance of each vertex in the hierarchical structure, and determines the affected regions of each vertex at the hierarchical levels. In the contraction phase, it also divides the simplification procedure into two steps: half-edge contraction and optimization. In the first step, memoryless quadric metric error and the importance of vertices in the hierarchical structure are combined to determine one operation of half-edge contraction. In the second step, it repositions the vertices in the half-edge simplified mesh by minimizing the multilevel synthesized quadric error on the corresponding affected regions from the immediately local to the more global. The experiments illustrate the competitive results.
IEEE Trans. Vis. Comput. Graph.	A New Physical Model with Multilayer Architecture for Facial Expression Animation Using Dynamic Adaptive Mesh.	Yu Zhang,Edmond C. Prakash,Eric Sung	2004	Abstract--This paper presents a new physically-based 3D facial model based on anatomical knowledge which provides high fidelity for facial expression animation while optimizing the computation. Our facial model has a multilayer biomechanical structure, incorporating a physically-based approximation to facial skin tissue, a set of anatomically-motivated facial muscle actuators, and underlying skull structure. In contrast to existing mass-spring-damper (MSD) facial models, our dynamic skin model uses the nonlinear springs to directly simulate the nonlinear visco-elastic behavior of soft tissue and a new kind of edge repulsion spring is developed to prevent collapse of the skin model. Different types of muscle models have been developed to simulate distribution of the muscle force applied on the skin due to muscle contraction. The presence of the skull advantageously constrain the skin movements, resulting in more accurate facial deformation and also guides the interactive placement of facial muscles. The governing dynamics are computed using a local semi-implicit ODE solver. In the dynamic simulation, an adaptive refinement automatically adapts the local resolution at which potential inaccuracies are detected depending on local deformation. The method, in effect, ensures the required speedup by concentrating computational time only where needed while ensuring realistic behavior within a predefined error threshold. This mechanism allows more pleasing animation results to be produced at a reduced computational cost.
CVPR	2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2005), 20-26 June 2005, San Diego, CA, USA		2005	2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2005), 20-26 June 2005, San Diego, CA, USA
ICCV	Mixtures of Dynamic Textures.	Antoni B. Chan,Nuno Vasconcelos	2005	A dynamic texture is a linear dynamical system used to model a single video as a sample from a spatio-temporal stochastic process. In this work, we introduce the mixture of dynamic textures, which models a collection of videos consisting of different visual processes as samples from a set of dynamic textures. We derive the EM algorithm for learning a mixture of dynamic textures, and relate the learning algorithm and the dynamic texture mixture model to previous works. Finally, we demonstrate the applicability of the proposed model to problems that have traditionally been challenging for computer vision.
ICCV	Good Continuation of General 2D Visual Features: Dual Harmonic Models and Computational Inference.	Ohad Ben-Shahar,Steven W. Zucker	2005	Good continuation is a fundamental principle of perceptual organization that guides the grouping of parts based on how they should succeed one another within coherent wholes. Despite the general language that was used by the Gestalt psychologists in phrasing this principle, computational work has focused almost exclusively on the study of curve-like structures. Here we offer, for the first time, a rigorous generalization of good continuation to arbitrary visual structures that can be abstracted as scalar functions over the image plane. The differential geometry of these structures dictates that their good continuation should be based both on their value and on the geometry of their levelsets, which yield a coupled system of equations solvable for a formal model. We exhibit the resulting computation on shading and intensity functions, demonstrating how it eliminates spurious measurements while preserving both regular structure and singularities. Related implementations could be applied to color channels, motion magnitude, and disparity signals.
ICCV	Face Recognition in the Presence of Multiple Illumination Sources.	Gaurav Aggarwal,Rama Chellappa	2005	Most existing face recognition algorithms work well for controlled images but are quite susceptible to changes in illumination and pose. This has led to the rise of analysis-by-synthesis approaches due to their inherent potential to handle these external factors. Though these approaches work quite well, most of them assume that the face is illuminated by a single light source which is usually not true in realistic conditions. In this paper, we propose an algorithm to recognize faces illuminated by arbitrarily placed, multiple light sources. The algorithm does not need to know the number of light sources and works extremely well even while recognizing faces illuminated by different number of light sources. Results using this algorithm are reported on multiple-illumination datasets generated from PIE [10] and Yale Face Database B [5]. We also highlight the importance of the hard non-linearity in the Lambert¿s law which is often ignored, probably to linearize the estimation process.
ICCV	An Algebraic Approach to Surface Reconstruction from Gradient Fields.	Amit K. Agrawal,Rama Chellappa,Ramesh Raskar	2005	Several important problems in computer vision such as Shape from Shading (SFS) and Photometric Stereo (PS) require reconstructing a surface from an estimated gradient field, which is usually non-integrable, i.e. have non-zero curl. We propose a purely algebraic approach to enforce integrability in discrete domain. We first show that enforcing integrability can be formulated as solving a single linear system Ax = b over the image. In general, this system is under-determined. We show conditions under which the system can be solved and a method to get to those conditions based on graph theory. The proposed approach is non-iterative, has the important property of local errorconfinement and can be applied to several problems. Results on SFS and PS demonstrate the applicability of our method.
ICCV	Actions as Space-Time Shapes.	Moshe Blank,Lena Gorelick,Eli Shechtman,Michal Irani,Ronen Basri	2005	Human action in video sequences can be seen as silhouettes of a moving torso and protruding limbs undergoing articulated motion. We regard human actions as three-dimensional shapes induced by the silhouettes in the space-time volume. We adopt a recent approach [9] for analyzing 2D shapes and generalize it to deal with volumetric space-time action shapes. Our method utilizes properties of the solution to the Poisson equation to extract space-time features such as local space-time saliency, action dynamics, shape structure and orientation. We show that these features are useful for action recognition, detection and clustering. The method is fast, does not require video alignment and is applicable in (but not limited to) many scenarios where the background is known. Moreover, we demonstrate the robustness of our method to partial occlusions, non-rigid deformations, significant changes in scale and viewpoint, high irregularities in the performance of an action and low quality video.
ICCV	Detecting Irregularities in Images and in Video.	Oren Boiman,Michal Irani	2005	"We address the problem of detecting irregularities in visual data, e.g., detecting suspicious behaviors in video sequences, or identifying salient patterns in images. The term ""irregular"" depends on the context in which the ""regular"" or ""valid"" are defined. Yet, it is not realistic to expect explicit definition of all possible valid configurations for a given context. We pose the problem of determining the validity of visual data as a process of constructing a puzzle: We try to compose a new observed image region or a new video segment (""the query"") using chunks of data (""pieces of puzzle"") extracted from previous visual examples (""the database""). Regions in the observed data which can be composed using large contiguous chunks of data from the database are considered very likely, whereas regions in the observed data which cannot be composed from the database (or can be composed, but only using small fragmented pieces) are regarded as unlikely/suspicious. The problem is posed as an inference process in a probabilistic graphical model. We show applications of this approach to identifying saliency in images and video, and for suspicious behavior recognition."
ICCV	A Theoretical Limit on the Number of Effective Pixels that can be Optically Resolved on a Non-Planar Subject.	Robert Boman	2005	In normal imaging systems, the depth of field is inversely proportional to the lens aperture. If we assume the system is diffraction limited, then the maximum resolution (i.e. pixels per mm) is proportional to the lens aperture. Thus there is a tradeoff between depth of field and resolution. This tradeoff creates an upper limit on the number of pixels that can be resolved on a non-planar subject. This paper presents the theoretical limit on the number of pixels. The derivations of the limit shows that the limit is only a function of the subject size and depth. The subject distance, focal length, and sensor size do not matter. For small subjects the limit is well below the capabilities of modern imaging systems. For example a subject 15cm wide and 10cm deep can only be imaged with 300,000 pixels even though sensors with 10 times that many pixels are readily available. The resulting limit has obvious applications in machine vision, particularly when specifying optics and imaging sensors. Experimental results are provided to validate the main result of the paper.
ICCV	A Supervised Learning Framework for Generic Object Detection in Images.	Saad Ali,Mubarak Shah	2005	In recent years Kernel Principal Component Analysis (Kernel PCA) has gained much attention because of its ability to capture nonlinear image features, which are particularly important for encoding image structure. Boosting has been established as a powerful learning algorithm that can be used for feature selection. In this paper we present a novel framework for object class detection that combines the feature reduction and feature selection abilities of Kernel PCA and AdaBoost respectively. The classifier obtained in this way is able to handle change in object appearance, illumination conditions, and surrounding clutter. A nonlinear subspace is learned for positive and negative object classes using Kernel PCA. Features are derived by projecting example images onto the learned subspaces. Base learners are modeled using Bayes classifier. AdaBoost is then employed to discover the features that are most relevant for the object detection task at hand. The proposed method has been successfully tested on wide range of object classes (cars, airplanes, pedestrians, motorcycles, etc) using standard data sets and has shown remarkable performance. Using a small training set, a classifier learned in this way was able to generalize the intra-class variation while still maintaining high detection rate. In most object categories we achieved detection rates of above 95% with minimal false alarm rates. We demonstrate the effectiveness of our approach in terms of absolute performance parameters and comparative performance against current state of the art approaches.
ICCV	An Axis-Based Representation for Recognition.	Cagri Aslan,Sibel Tari	2005	We present a new axis-based shape representation scheme along with a matching framework to address the problem of generic shape recognition. The main idea is to define the relative spatial arrangement of local symmetry axes and their metric properties in a shape centered coordinate frame. The resulting descriptions are invariant to scale, rotation, small changes in viewpoint and articulations. Symmetry points are extracted from a surface whose level curves roughly mimic the motion by curvature. By increasing the amount of smoothing on the evolving curve, only those symmetry axes that correspond to the most prominent parts of a shape are extracted. The representation does not suffer from the common instability problems of the traditional connected skeletons. It captures the perceptual qualities of shapes well. Therefore finding the similarities and the differences among shapes becomes easier. The matching process gives highly successful results on a diverse database of 2D shapes.
ICCV	Multi-View Surface Reconstruction Using Polarization.	Gary Atkinson,Edwin R. Hancock	2005	A new technique for surface reconstruction is developed that uses polarization information from two views. One common problem arising from many multiple view techniques is that of finding correspondences between pixels on each image. In the new method, these correspondences are found by exploiting the spontaneous polarization of light caused by reflection to recover surface normals. These normals are then used to recover surface height. The similarity between reconstructed surface regions determines whether or not a pair of points correspond to each other. The technique is thus able to overcome the convex/concave ambiguity found in many single view techniques. Because the technique relies on smooth surface regions to detect correspondences, rather than feature detection, it is applicable to objects normally inaccessible to stereo vision. Also due to this fact, it is possible to remove noise without causing oversmoothing problems.
ICCV	Shape Recovery of 3D Data Obtained from a Moving Range Sensor by Using Image Sequences.	Atsuhiko Banno,Katsushi Ikeuchi	2005	For a large object, scanning from the air is one of the most efficient methods of obtaining 3D data. In the case of large cultural heritage objects, there are some difficulties in scanning with respect to safety and efficiency. To remedy these problems, we have been developing a novel 3D measurement system, the Floating Laser Range Sensor (FLRS), in which a range sensor is suspended beneath a balloon. The obtained data, however, have some distortions due to sensor-movements during the scanning process. In this paper, we propose a method to recover 3D range data obtained by a moving laser range sensor. This method is applicable not only to our FLRS, but also to a general moving range sensor. Using image sequences from a video camera mounted on the FLRS enables us to estimate the motion of the FLRS without any physical sensors such as gyros or GPS. In the first stage, the initial values of camera motion parameters are estimated by full-perspective factorization. The next stage refines camera motion parameters using the relationships between camera images and range data distortion. Finally, by using the refined parameters, the distorted range data are recovered. In addition, our method is applicable with an uncalibrated video camera and range sensor system. We applied this method to an actual scanning project, and the results showed the effectiveness of our method.
ICCV	Efficient Visual Event Detection Using Volumetric Features.	Yan Ke,Rahul Sukthankar,Martial Hebert	2005	This paper studies the use of volumetric features as an alternative to popular local descriptor approaches for event detection in video sequences. Motivated by the recent success of similar ideas in object detection on static images, we generalize the notion of 2D box features to 3D spatio-temporal volumetric features. This general framework enables us to do real-time video analysis. We construct a real-time event detector for each action of interest by learning a cascade of filters based on volumetric features thatefficiently scans video sequences in space and time. This event detector recognizes actions that are traditionally problematic for interest point methods ¿ such as smooth motions where insufficient space-time interest points are available. Our experiments demonstrate that the technique accurately detects actions on real-world sequences and is robust to changes in viewpoint, scale and action speed. We also adapt our technique to the related task of human action classification and confirm that it achieves performance comparable to a current interest point based human activity recognizer on a standard database of human activities.
ICCV	Quasiconvex Optimization for Robust Geometric Reconstruction.	Qifa Ke,Takeo Kanade	2005	Geometric reconstruction problems in computer vision are often solved by minimizing a cost function that combines the reprojection errors in the 2D images. In this paper, we show that, for various geometric reconstruction problems, their reprojection error functions share a common and quasiconvex formulation. Based on the quasiconvexity, we present a novel quasiconvex optimization framework in which the geometric reconstruction problems are formulated as a small number of small-scale convex programs that are ready to solve. Our final reconstruction algorithm is simple and has intuitive geometric interpretation. In contrast to existing local minimization approaches, our algorithm is deterministic and guarantees a predefined accuracy of the minimization result.The quasiconvexity also provides an intuitive method to handle directional uncertainties and outliers in measurements. When there are outliers in the measurements, our method provides a mechanism to locate the global minimum of a robust error function. For large scale problems and when computational resources are constrained, we provide an efficient approximation that gives a good upper bound (but not global minimum) on the reconstruction error. We demonstrate the effectiveness of our algorithm by experiments on both synthetic and real data.
ICCV	Efficient Learning of Relational Object Class Models.	Aharon Bar-Hillel,Tomer Hertz,Daphna Weinshall	2005	We present an efficient method for learning part-based object class models. The models include location and scale relations between parts, as well as part appearance. Models are learnt from raw object and background images, represented as an unordered set of features extracted using an interest point detector. The object class is generatively modeled using a simple Bayesian network with a central hidden node containing location and scale information, and nodes describing object parts. The model¿s parameters, however, are optimized to reduce a loss function which reflects training error, as in discriminative methods. Specifically, the optimization is done using a boosting-like technique with complexity linear in the number of parts and the number of features per image. This efficiency allows our method to learn relational models with many parts and features, and leads to improved results when compared with other methods. Extensive experimental results are described, using some common bench-mark datasets and three sets of newly collected data, showing the relative advantage of our method.
ICCV	Integrated Spatial and Frequency Domain 2D Motion Segmentation and Estimation.	Alexia Briassouli,Narendra Ahuja	2005	A video containing multiple objects in rotational and translational motion is analyzed through a combination of spatial and frequency domain representations. It is argued that the combined analysis can take advantage of the strengths of both representations. Initial estimates of constant, as well as time-varying, translation and rotation velocities are obtained from frequency analysis. Improved motion estimates and motion segmentation for the case of translation are achieved by integrating spatial and Fourier domain information. For combined rotational and translational motions, the frequency representation is used for motion estimation, but only spatial information can be used to separate and extract the independently moving objects. The proposed algorithms are tested on synthetic and real videos.
ICCV	Incorporating Visual Knowledge Representation in Stereo Reconstruction.	Adrian Barbu,Song Chun Zhu	2005	In this paper, we present a two-layer generative model that incorporates generic middle-level visual knowledge for dense stereo reconstruction. The visual knowledge is represented by a dictionary of surface primitives including various categories of boundary discontinuities and junctions in parametric form. Given a stereo pair, we first compute a primal sketch representation which decomposes the image into a structural part for object boundaries and high intensity contrast represented by a 2D sketch graph, and a structureless part represented by Markov random field on pixels. Then we label the sketch graph and compute the 3D sketch (like a wire-frame) by fitting the primitive dictionary to the sketch graph. The surfaces between the 3D sketches are filled in by computing the depth of the MRF model on the structureless part. These two levels interact closely since the MRF is used to propagate information between the primitives, and at the same time, the primitives act as boundary conditions for the MRF. The two processes maximize a Bayesian posterior probability jointly. We propose anMCMC algorithm that simultaneously infers the 3D primitive types and parameters and estimates the depth of the scene. Our experiments show that this representation can infer the depth map with sharp boundaries and junctions for textureless images, curve objects and free-form shapes.
ICCV	Regular Polygon Detection.	Nick Barnes,Gareth Loy,David Shaw,Antonio Robles-Kelly	2005	This paper describes a new robust regular polygon detector. The regular polygon transform is posed as a mixture of regular polygons in a five dimensional space. Given the edge structure of an image, we derive the a posteriori probability for a mixture of regular polygons, and thus the probability density function for the appearance of a mixture of regular polygons. Likely regular polygons can be isolated quickly by discretising and collapsing the search space into three dimensions. The remaining dimensions may be efficiently recovered subsequently using maximum likelihood at the locations of the most likely polygons in the subspace. This leads to an efficient algorithm. Also the a posteriori formulation facilitates inclusion of additional a priori information leading to real-time application to road sign detection. The use of gradient information also reduces noise compared to existing approaches such as the generalised Hough transform. Results are presented for images with noise to show stability. The detector is also applied to two separate applications: real-time road sign detection for on-line driver assistance; and feature detection, recovering stable features in rectilinear environments.
ICCV	Fundamental Matrix for Cameras with Radial Distortion.	João P. Barreto,Kostas Daniilidis	2005	When deploying a heterogeneous camera network or when we use cheap zoom cameras like in cell-phones, it is not practical, if not impossible to off-line calibrate the radial distortion of each camera using reference objects. It is rather desirable to have an automatic procedure without strong assumptions about the scene. In this paper, we present a new algorithm for estimating the epipolar geometry of two views where the two views can be radially distorted with different distortion factors. It is the first algorithm in the literature solving the case of different distortion in the left and right view linearly and without assuming the existence of lines in the scene. Points in the projective plane are lifted to a quadric in three-dimensional projective space. A radial distortion of the projective plane results to a matrix transformation in the space of lifted coordinates. The new epipolar constraint depends linearly on a 4x4 radial fundamental matrix which has 9 degrees of freedom. A complete algorithm is presented and tested on real imagery.
ICCV	Towards Ultimate Motion Estimation: Combining Highest Accuracy with Real-Time Performance.	Andrés Bruhn,Joachim Weickert	2005	Although variational methods are among the most accurate techniques for estimating the optical flow, they have not yet entered the field of real-time vision. Main reason is the great popularity of standard numerical schemes that are easy to implement, however, at the expense of being too slow for real-time performance. In our paper we address this problem in two ways: (i) We present an improved version of the highly accurate technique of Brox et al. [9]. Thereby we show that a separate robustification of the constancy assumptions is very useful, in particular if the 1-norm is used as penalizer. As a result, a method is obtained that yields the lowest angular errors in the literature. (ii) We develop an efficient numerical scheme for the proposed approach that allows real-time performance for sequences of size 160 × 120. To this end, we combine two hierarchical strategies: A coarse-to-fine warping strategy as implementation of a fixed point iteration for a non-convex optimisation problem and a nonlinear full multigrid method ¿ a so called full approximation scheme (FAS) ¿ for solving the highly nonlinear equation systems at each warping level. In the experimental section the advantage of the proposed approach becomes obvious: Outperforming standard numerical schemes by two orders of magnitude frame rates of six high quality flow fields per second are obtained on a 3.06 GHz Pentium4 PC.
ICCV	3D Object Reconstruction from a Single 2D Line Drawing without Hidden Lines.	Liangliang Cao,Jianzhuang Liu,Xiaoou Tang	2005	The human vision system can interpret a single 2D line drawing as a 3D object without much difficulty even if the hidden lines of the object are invisible. Several reconstruction approaches have tried to emulate this ability, but they cannot recover the complete object if the hidden lines of the object are not shown. This paper proposes a novel approach for reconstructing complete 3D objects from line drawings without hidden lines. First, we develop some constraints and properties for the inference of the topology of the invisible edges and vertices of an object. Then we present a reconstruction method based on perceptual symmetry and planarity of the object. We give a number of examples to demonstrate the ability of our approach.
ICCV	Large Deformation Diffeomorphic Metric Mapping of Fiber Orientations.	Yan Cao,Michael I. Miller,Raimond L. Winslow,Laurent Younes	2005	This paper proposes a method to match diffusion tensor magnetic resonance images (DT-MRI) through the large deformation diffeomorphic metric mapping of vector fields, focusing on the fiber orientations, considered as unit vector fields on the image volume. We study a suitable action of diffeomorphisms on such vector fields, and provide an extension of the Large Deformation Diffeomorphic Metric Mapping framework to this type of dataset, resulting in optimizing for geodesics on the space of diffeomorphisms connecting two images. Two different distance function of vector fields are considered. Existence of the minimizers under smoothness assumptions on the compared vector fields is proved, and coarse to fine hierarchical strategies are detailed, to reduce both ambiguities and computation load. This is illustrated by numerical experiments on DT-MRI heart and brain images.
ICCV	Class-Specific Material Categorisation.	Barbara Caputo,Eric Hayman,P. Mallikarjuna	2005	Although a considerable amount of work has been published on material classification, relatively little of it studies situations with considerable variation within each class. Many experiments use the exact same sample, or different patches from the same image, for training and test sets. Thus, such studies are vulnerable to effectively recognising one particular sample of a material as opposed to the material category. In contrast, this paper places firm emphasis on the capability to generalise to previously unseen instances of materials. We adopt an appearance-based strategy, and conduct experiments on a new database which contains several samples of each of eleven material categories, imaged under a variety of pose, illumination and scale conditions. Together, these sources of intra-class variation provide a stern challenge indeed for recognition. Somewhat surprisingly, the difference in performance between various state-of-the-art texture descriptors proves rather small in this task. On the other hand, we clearly demonstrate that very significant gains can be achieved via different SVM-based classification techniques. Selecting appropriate kernel parameters proves crucial. This motivates a novel recognition scheme based on a decision tree. Each node contains an SVM to split one class from all others with a kernel parameter optimal for that particular node. Hence, each decision is made using a different, optimal, class-specific metric. Experiments show the superiority of this approach over several state-of-the-art classifiers.
ICCV	Multi-Scale Gesture Recognition from Time-Varying Contours.	Hong Li,Michael A. Greenspan	2005	A novel method is introduced to recognize and estimate the scale of time-varying human gestures. It exploits the changes in contours along spatio-temporal directions. Each contour is first parameterized as a 2D function of radius vs. cumulative contour length, and a 3D surface is composed from a sequence of such functions. In a two-phase recognition process, Dynamic TimeWarping is employed to rule out significantly different gesture models, and then Mutual Information (MI) is applied for matching the remaining models. The system has been tested on 8 gestures performed by 5 subjects with varied time scales. The two-phase process is compared against exhaustively testing three similarity measures based upon MI, Correlation, and Nonparametric Kernel Density Estimation. Experimental results demonstrate that the exhaustive application of MI is the most robust with a recognition rate of 90.6%, however, the two-phase approach is much more computationally efficient with a comparable recognition rate of 90.0%.
ICCV	Shape Parameter Optimization for AdaBoosted Active Shape Model.	Yuanzhong Li,Wataru Ito	2005	Active Shape Model (ASM) has been shown to be a powerful tool to aid the interpretation of images, especially in face alignment. ASM local appearance model parameter estimation is based on the assumption that residuals between model fit and data have a Gaussian distribution. Moreover, to generate an allowable face shape, ASM truncates coefficients of shape principal components into the bounds determined by eigenvalues. In this paper, an algorithm of modeling local appearances, called AdaBoosted ASM, and a shape parameter optimization method are proposed. In the algorithm of modeling the local appearances, we describe our novel modeling method by using AdaBoosted histogram classifiers, in which the assumption of the Gaussian distribution is not necessary. In the shape parameter optimization, we describe that there is an inadequacy on controlling shape parameters in ASM, and our novel method on how to solve it. Experimental results demonstrate that the AdaBoosted histogram classifiers improve robustness of landmark displacement greatly, and the shape parameter optimization solves the inadequacy problem of ASM on shape constraint effectively.
ICCV	Robust Path-Based Spectral Clustering with Application to Image Segmentation.	Hong Chang,Dit-Yan Yeung	2005	Spectral clustering and path-based clustering are two recently developed clustering approaches that have delivered impressive results in a number of challenging clustering tasks. However, they are not robust enough against noise and outliers in the data. In this paper, based on M-estimation from robust statistics, we develop a robust path-based spectral clustering method by defining a robust path-based similarity measure for spectral clustering. Our method is significantly more robust than spectral clustering and path-based clustering. We have performed experiments based on both synthetic and real-world data, comparing our method with some other methods. In particular, color images from the Berkeley Segmentation Dataset and Benchmark are used in the image segmentation experiments. Experimental results show that our method consistently outperforms other methods due to its higher robustness.
ICCV	Image Statistics Based on Diffeomorphic Matching.	Guillaume Charpiat,Olivier D. Faugeras,Renaud Keriven	2005	We propose a new approach to deal with the first and second order statistics of a set of images. These statistics take into account the images characteristic deformations and their variations in intensity. The central algorithm is based on non-supervised diffeomorphic image matching (without landmarks or human intervention). As they convey the notion of the mean shape and colors of an object and the one of its common variations, such statistics of sets of images may be relevant in the context of object recognition, both in the segmentation of any of its representations and in the classification of them. The proposed approach has been tested on a small database of face images to compute a mean face and second order statistics. The results are very encouraging since, wheras the algorithm does not need any human intervention and is not specific to face image databases, the mean image looks like a real face and the characteristic modes of variation (deformation and intensity changes) are sensible.
ICCV	Designing Spatially Coherent Minimizing Flows for Variational Problems Based on Active Contours.	Guillaume Charpiat,Renaud Keriven,Jean-Philippe Pons,Olivier D. Faugeras	2005	This paper tackles an important aspect of the variational problems involving active contours, which has been largely overlooked so far: the optimization by gradient flows. Classically, the definition of a gradient depends directly on the choice of an inner product structure. This consideration is largely absent from the active contours literature. Most authors, overtly or covertly, assume that the space of admissible deformations is ruled by the canonical L2 inner product. The classical gradient flows reported in the literature are relative to this particular choice. In this paper, we investigate the relevance of using other inner products, yielding other gradient descents, and some other minimizing flows not deriving from any inner product. In particular, we show how to induce different degrees of spatial coherence into the minimizing flow, in order to decrease the probability of getting trapped into irrelevant local minima. We show with some numerical experiments that the sensitivity of the active contours method to initial conditions, which seriously limits its applicability and its efficiency, is alleviated by our application-specific spatially coherent minimizing flows.
ICCV	Lighting Normalization with Generic Intrinsic Illumination Subspace for Face Recognition.	Chia-Ping Chen,Chu-Song Chen	2005	In this paper, we introduce the concept of Intrinsic Illumination Subspace which is based on the intrinsic images. This intrinsic illumination subspace enables an analytic generation of the illumination images under varying lighting conditions. When objects of the same class are concerned, our method allows a class-based generic intrinsic illumination subspace to be constructed in advance. We propose a lighting normalization method based on the generic intrinsic illumination subspace, which is used as a bootstrap subspace for novel images. Face recognition experiments are performed to demonstrate the effectiveness of our method.
ICCV	Learning Effective Image Metrics from Few Pairwise Examples.	Hwann-Tzong Chen,Tyng-Luh Liu,Chiou-Shann Fuh	2005	We present a new approach to learning image metrics. The main advantage of our method lies in a formulation that requires only a few pairwise examples. Apparently, based on the little amount of side-information, it would take a very effective learning scheme to yield a useful image metric. Our algorithm achieves this goal by addressing two key issues. First, we establish a global-local (glocal) image representation that induces two structure-meaningful vector spaces to respectively describe the global and the local image properties. Second, we develop a metric optimization framework that finds an optimal bilinear transform to best explain the given side-information. We emphasize it is the glocal image representation that makes the use of bilinear transform more powerful. Experimental results on classifications of face images and visual tracking are included to demonstrate the contributions of the proposed method.
ICCV	Mutual Information Regularized Bayesian Framework for Multiple Image Restoration.	Yunqiang Chen,Hongcheng Wang,Tong Fang,Jason Tyan	2005	Bayesian methods have been extensively used in various applications. However, there are two intrinsic issues rarely addressed, namely generalization and validity. In the context of multiple image restoration, we show that traditional Bayesian methods are sensitive to model errors and cannot guarantee valid results satisfying the underlying prior knowledge, e.g. independent noise property. To improve the Bayesian framework¿s generalization, we propose to explicitly enforce the validity of the result. Independent noise prior is very important but largely under-utilized in previous literature. In this paper, we use mutual information (MI) to explicitly enforce the independence. Efficient approximations based on Taylor expansion are proposed to adapt MI into standard energy forms to regularize the Bayesian methods. The new regularized Bayesian framework effectively utilizes the traditional generative signal/noise models but is much more robust to various model errors, as demonstrated in experiments on some demanding imaging applications.
ICCV	Registration of Multimodal Fluorescein Images Sequence of the Retina.	Tae Eun Choe,Isaac Cohen	2005	In this study we present a Y-feature extraction method for registering color and fluorescein angiograms of the retina. The registration of multimodal fluorescein imagery requires the identification of strong geometric features in the retinal images that are invariant across modalities and to temporal grey level variations due to the propagation of the dye in the vessels. The most informative features, invariant across the considered modalities, are the locations of vessels¿ ramification: the so-called Y-features. We propose a Y-feature extraction method based on the local classification of image gradient information and an articulated model. An appropriate cost function is proposed for fitting the model using a gradient-based approach. The fitted Y-features are subsequently matched across the images for registering the color and fluorescein images. Experimental results obtained on a large database validate the proposed method.
ICCV	Background Estimation as a Labeling Problem.	Scott Cohen	2005	We present a new background estimation algorithm that constructs the background of an image sequence with moving objects by copying areas from input frames. The background estimation problem is formulated as an optimal labeling problem in which the label at an output pixel is the frame number from which to copy the background color. The costs of assigning labels encourage seamless copying from regions that are stationary over a period of time in such a way that implied motion boundaries occur at intensity edges. This is accomplished without explicitly tracking the moving objects or computing optical flow. Experiments demonstrate that our algorithm is effective in difficult areas where the background is visible for only a small fraction of time, and on inputs with both moving objects that are not always in motion and moving objects with textureless areas.
ICCV	A Stochastic Filter for Fluid Motion Tracking.	Anne Cuzol,Étienne Mémin	2005	In this paper we present a method for the tracking of fluid flows velocity fields. The technique we propose is formalized within sequential Bayesian filter framework. The filter we propose here combines an Itô diffusion process coming from a stochastic formulation of the vorticity-velocity form of Navier-Stokes equation and discrete measurements extracted from an image sequence. The resulting tracker provides robust and consistent estimations of instantaneous motion fields along the whole image sequence. In order to handle a state space of reasonable dimension for the stochastic filtering problem, we represent the motion field as a combination of adapted basis functions. The used basis functions ensue from a mollification of Biot-Savart integral and a discretization of the vorticity and divergence maps of the fluid vector field. The efficiency of the method is demonstrated on a long real world sequence showing a vortex launch at tip of airplane wing.
ICCV	Object Recognition in High Clutter Images Using Line Features.	Philip David,Daniel DeMenthon	2005	We present an object recognition algorithm that uses model and image line features to locate complex objects in high clutter environments. Finding correspondences between model and image features is the main challenge in most object recognition systems. In our approach, corresponding line features are determined by a three-stage process. The first stage generates a large number of approximate pose hypotheses from correspondences of one or two lines in the model and image. Next, the pose hypotheses from the previous stage are quickly ranked by comparing local image neighborhoods to the corresponding local model neighborhoods. Fast nearest neighbor and range search algorithms are used to implement a distance measure that is unaffected by clutter and partial occlusion. The ranking of pose hypotheses is invariant to changes in image scale, orientation, and partially invariant to affine distortion. Finally, a robust pose estimation algorithmis applied for refinement and verification, starting from the few best approximate poses produced by the previous stages. Experiments on real images demonstrate robust recognition of partially occluded objects in very high clutter environments.
ICCV	BRDF Invariant Stereo Using Light Transport Constancy.	James E. Davis,Ruigang Yang,Liang Wang	2005	Nearly all existing methods for stereo reconstruction assume that scene reflectance is Lambertian, and make use of color constancy as a matching invariant. We introduce a new invariant for stereo reconstruction called Light Transport Constancy, which allows completely arbitrary scene reflectance (BRDFs). This invariant can be used to formulate a rank constraint on multiview stereo matching when the scene is observed in several lighting configurations. In addition, we show that this multiview constraint can be used with as few as two cameras and two lighting configurations. Unlikely previous methods for BRDF invariant stereo, Light Transport Constancy does not require precisely configured or calibrated light sources, nor calibration objects in the scene. Importantly, the new constraint can be used to provide BRDF invariance to any existing stereo method, whenever appropriate lighting variation is available.
ICCV	Active Search for Real-Time Vision.	Andrew J. Davison	2005	In most cases when information is to be extracted from an image, there are priors available on the state of the world and therefore on the detailed measurements which will be obtained. While such priors are commonly combined with the actual measurements via Bayes¿ rule to calculate posterior probability distributions on model parameters, their additional value in guiding efficient image processing has almost always been overlooked. Priors tell us where to look for information in an image, how much computational effort we can expect to expend to extract it, and of how much utility to the task in hand it is likely to be. Such considerations are of importance in all practical real-time vision systems, where the processing resources available at each frame in a sequence are strictly limited ¿ and it is exactly in high frame-rate real-time systems such as trackers where strong priors are most likely to be available. In this paper, we use Shannon information theory to analyse the fundamental value of measurements using mutual information scores in absolute units of bits, specifically looking at the overwhelming case where uncertainty can be characterised by Gaussian probability distributions. We then compare these measurement values with the computational cost of the image processing required to obtain them. This theory puts on a firm footing for the first time principles of ¿active search¿ for efficient guided image processing, in which candidate features of possibly different types can be compared and selected automatically for measurement.
ICCV	"Avoiding the ""Streetlight Effect"": Tracking by Exploring Likelihood Modes."	David Demirdjian,Leonid Taycher,Gregory Shakhnarovich,Kristen Grauman,Trevor Darrell	2005	Classic methods for Bayesian inference effectively constrain search to lie within regions of significant probability of the temporal prior. This is efficient with an accurate dynamics model, but otherwise is prone to ignore significant peaks in the true posterior. A more accurate posterior estimate can be obtained by explicitly finding modes of the likelihood function and combining them with a weak temporal prior. In our approach modes are found using effi- cient example-based matching followed by local refinement to find peaks and estimate peak bandwidth. By reweighting these peaks according to the temporal prior we obtain an estimate of the full posterior model. We show comparative results on real and synthetic images in a high degree of freedom articulated tracking task.
ICCV	A Symmetric Patch-Based Correspondence Model for Occlusion Handling.	Yi Deng,Qiong Yang,Xueyin Lin,Xiaoou Tang	2005	Occlusion is one of the challenging problems in stereo. In this paper, we solve the problem in a segment-based style. Both images are segmented, and we propose a novel patch-based stereo algorithm that cuts the segments of one image using the segments of the other, and handles occlusion areas in a proper way. A symmetric graph-cuts optimization framework is used to find correspondence and occlusions simultaneously. The experimental results show superior performance of the proposed algorithm, especially on occlusions, untextured areas and discontinuities.
ICCV	Simultaneous Facial Action Tracking and Expression Recognition Using a Particle Filter.	Fadi Dornaika,Franck Davoine	2005	The recognition of facial gestures and expressions in image sequences is an important and challenging problem. Most of the existing methods adopt the following paradigm. First, facial actions/features are retrieved from the images, and then facial expressions are recognized based on the retrieved temporal parameters. Unlike this main stream, this paper introduces a new approach allowing the simultaneous recovery of facial actions and expression using a particle filter adopting multi-class dynamics that are conditioned on the expression. For each frame in the video sequence, our approach is split in two consecutive stages. In the first stage, the 3D head pose is recovered using a deterministic registration technique based on Online Appearance Models. In the second stage, the facial actions as well as the facial expression are simultaneously recovered using the stochastic framework with mixed states. The proposed fast scheme is either as robust as existing ones or more robust with respect to many regards. Experimental results show the feasibility and robustness of the proposed approach.
ICCV	On Optimal Light Configurations in Photometric Stereo.	Ondrej Drbohlav,Mike J. Chantler	2005	This paper develops new theory for the optimal placement of photometric stereo lighting in the presence of camera noise. We show that for three lights, any triplet of orthogonal light directions minimises the uncertainty in scaled normal computation. The assumptions are that the camera noise is additive and normally distributed, and uncertainty is defined as the expectation of squared distance of scaled normal to the ground truth. If the camera noise is of zero mean and variance ¿², the optimal (minimum) uncertainty in the scaled normal is 3¿². For case of n > 3 lights, we show that the minimum uncertainty is 9¿²/n, and identify sets of light configurations which reach this theoretical minimum.
ICCV	Can Two Specular Pixels Calibrate Photometric Stereo?.	Ondrej Drbohlav,Mike J. Chantler	2005	Lambertian photometric stereo with unknown light source parameters is ambiguous. Provided that the object imaged constitutes a surface, the ambiguity is represented by the group of Generalised Bas-Relief (GBR) transformations. We show that this ambiguity is resolved when specularreflection is present in two images taken under two different light source directions. We identify all configurations of the two directional lights which are singular and show that they can easily be tested for. While previous work used optimisation algorithms to apply the constraints implied by the specular reflectance component, we have developed a linear algorithm to achieve this goal. Our theory can be utilised to construct fast algorithms for automatic reconstruction of smooth glossy surfaces.
ICCV	Incremental Discovery of Object Parts in Video Sequences.	Stéphane Drouin,Patrick Hébert,Marc Parizeau	2005	This paper addresses the problem of automatically discovering the rigid parts of an initially unknown moving deformable object in a monocular video sequence. The parts are first extracted through motion-based segmentation, using a time scale automatically chosen with the quantity of motion concept. Tracking and reobservation reinforce these low-level segmentation results and further segmentation is performed only when and where no modeled parts can be tracked. Central to the system is the Modeler that minimizes the impacts of erroneous segmentations and departures in tracking. The sequential nature of the framework allows incremental modeling and segmentation of parts that need not simultaneously be visible or in motion, making it possible to circumvent the typical constraint of model initialization. The fundamental principles are strictly ensemblist and do not rely on any specific PDF. The interest of this framework is demonstrated on three types of video sequences including human and robot motion.
ICCV	Variational-Based Method to Extract Parametric Shapes from Images.	Moumen T. El-Melegy,Nagi H. Al-Ashwal,Aly A. Farag	2005	In this paper, we propose a variational method to segment image objects, which have a given parametric shape based on a level-set formulation of the Mumford-Shah functional, and the shape parameters. We define an energy functional composed by two complementary terms. The first one detects object boundaries using a Chan-Vese-like method. The second term constrains the contour to find a shape compatible with the parametric shape. The segmentation of the object of interest is given by the minimum of our energy functional. This minimum is computed with the calculus of variation and the gradient descent method that provide a system of evolution equations solved with the well-known level set method. We focus in this paper on the parametric category of image linear objects. Applications of the proposed model are presented on synthetic and real images.
ICCV	Feature Hierarchies for Object Classification.	Boris Epshtein,Shimon Ullman	2005	The paper describes a method for automatically extracting informative feature hierarchies for object classification, and shows the advantage of the features constructed hierarchically over previous methods. The extraction process proceeds in a top-down manner: informative top-level fragments are extracted first, and by a repeated application of the same feature extraction process the classification fragments are broken down successively into their own optimal components. The hierarchical decomposition terminates with atomic features that cannot be usefully decomposed into simpler features. The entire hierarchy, the different features and sub-features, and their optimal parameters, are learned during a training phase using training examples. Experimental comparisons show that these feature hierarchies are significantly more informative and better for classification compared with similar non-hierarchical features as well as previous methods for using feature hierarchies.
ICCV	"Identifying Individuals in Video by Combining ""Generative"" and Discriminative Head Models."	Mark Everingham,Andrew Zisserman	2005	The objective of this work is automatic detection and identification of individuals in unconstrained consumer video, given a minimal number of labelled faces as training data. Whilst much work has been done on (mainly frontal) face detection and recognition, current methods are notsufficiently robust to deal with the wide variations in pose and appearance found in such video. These include variations in scale, illumination, expression, partial occlusion, motion blur, etc. We describe two areas of innovation: the first is to capture the 3-D appearance of the entire head, rather than just the face region, so that visual features such as the hairline can be exploited. The second is to combine discriminative and ¿generative¿ approaches for detection and recognition. Images rendered using the head model are used to train a discriminative tree-structured classifier giving efficient detection and pose estimates over a very wide pose range with three degrees of freedom. Subsequent verification of the identity is obtained using the head model in a ¿generative¿ framework. We demonstrate excellent performance in detecting and identifying three characters and their poses in a TV situation comedy.
ICCV	Fast Recognition of Multi-View Faces with Feature Selection.	Zhi-Gang Fan,Bao-Liang Lu	2005	We propose a discriminative feature selection method utilizing support vector machines for the challenging task of multi-view face recognition. According to the statistical relationship between the two tasks, feature selection and multi-class classification, we integrate the two tasks into a single consistent framework and effectively realize the goal of discriminative feature selection. The classification process can be made faster without degrading the generalization performance through this discriminative feature selection method. On the UMIST multi-view face database, our experiments show that this discriminative feature selection method can speed up the multi-view face recognition process without degrading the correct rate and outperform the traditional kernel subspace methods.
ICCV	Probabilistic Contour Extraction Using Hierarchical Shape Representation.	Xin Fan,Chun Qi,Dequn Liang,Hua Huang	2005	In this paper, we address the issue of extracting contour of the object with a specific shape. A hierarchical graphical model is proposed to represent shape variations. A complex shape is decomposed into several components which are described as Principal Component Analysis (PCA) based models in various levels. The hierarchical representation allows for chain-like conditional dependency within a single level and bidirectional communication between different levels. Additionally, a Sequential Monte-Carlo (SMC) based inference algorithm that can explore the graphical structure is proposed to estimate the contour. The experiments performed on real-world hand and face images show that the proposed method is effective in combating occlusion and cluttered background. Moreover, it is possible to isolate the localization error to an individual component of a shape attributed to the hierarchical representation.
ICCV	Realtime IBR with Omnidirectional Crossed-Slits Projection.	Doron Feldman,Daphna Weinshall	2005	The Crossed-Slits (X-Slits) projection can be used to generate new views of a scene from a sequence of perspective images. Compared with other image-based rendering (IBR) techniques, X-Slits image generation is simple and requires a relatively small number of input images, which makes it suitable for realtime IBR. In this paper we extend this model to omnidirectional cameras and a circular slit. We show how it can be used for realtime image-based rendering of omnidirectional images, and how to optimize it for speed and quality. We analyze the inherent geometric distortions of the circular X-Slits projection, and describe a normalization mechanism to reduce distortions, creating a realistic virtual environment. Essentially the same mechanism is used to augment the X-Slits images with artificial objects, when using standard graphics tools which assume perspective projection.
ICCV	Bayesian Structural Content Abstraction for Region-Level Image Authentication.	Wei Feng,Zhi-Qiang Liu	2005	We present a hierarchical representation of image structure and use it for image content authentication. Firstly, we model the image with the Markov pixon random field. Within the Bayesian framework, the optimal label map and regional pixon map can be obtained, based on which wedefine an undirected graph, namely Bayesian Structural Content Abstraction (BaSCA). This representation captures the spatial topology information of homogeneous regions as well as their finest scale and interactions. Then, an efficient optimization scheme has been proposed to iteratively minimize the distance (or learning error) to all content-identical image samples generated by an acceptable operation setdefined by the user. In addition, we use the regional pixon map to remove spurious vertices and thus to establish a BaSCA hierarchy naturally. The BaSCA itself and its features can act as the signature of the protected image. Our experimental results show that the proposed approach has much less false positive and comparable false negative probability compared with the existing methods
ICCV	Building a Classification Cascade for Visual Identification from One Example.	Andras Ferencz,Erik G. Learned-Miller,Jitendra Malik	2005	"Object identification (OID) is specialized recognition where the category is known (e.g. cars) and the algorithm recognizes an object¿s exact identity (e.g. Bob¿s BMW). Two special challenges characterize OID. (1) Inter-class variation is often small (many cars look alike) and may be dwarfed by illumination or pose changes. (2) There may be many classes but few or just one positive ""training"" examples per class. Due to (1), a solution must locate possibly subtle object-specific salient features (a door handle) while avoiding distracting ones (a specular highlight). However, (2) rules out direct techniques of feature selection. We describe an online algorithm that takes one model image from a known category and builds an efficient ""same"" vs. ""different"" classification cascade by predicting the most discriminative feature set for that object. Our method not only estimates the saliency and scoring function for each candidate feature, but also models the dependency between features, building an ordered feature sequence unique to a specific model image, maximizing cumulative information content. Learned stopping thresholds make the classifier very efficient. To make this possible, category-specific characteristics are learned automatically in an off-line training procedure from labeled image pairs of the category, without prior knowledge about the category. Our method, using the same algorithm for both cars and faces, outperforms a wide variety of other methods."
ICCV	"Learning Object Categories from Google's Image Search."	Robert Fergus,Fei-Fei Li,Pietro Perona,Andrew Zisserman	2005	"Learning Object Categories from Google's Image Search."
ICCV	Discontinuity Preserving Stereo with Small Baseline Multi-Flash Illumination.	Rogerio Feris,Ramesh Raskar,Longbin Chen,Kar-Han Tan,Matthew Turk	2005	Currently, sharp discontinuities in depth and partial occlusions in multiview imaging systems pose serious challenges for many dense correspondence algorithms. However, it is important for 3D reconstruction methods to preserve depth edges as they correspond to important shape features like silhouettes which are critical for understanding the structure of a scene. In this paper we show how active illumination algorithms can produce a rich set of feature maps that are useful in dense 3D reconstruction. We start by showing a method to compute a qualitative depth map from a single camera, which encodes object relative distances and can be used as a prior for stereo. In a multiview setup, we show that along with depth edges, binocular half-occluded pixels can also be explicitly and reliably labeled. To demonstrate the usefulness of these feature maps, we show how they can be used in two different algorithms for dense stereo correspondence. Our experimental results show that our enhanced stereo algorithms are able to extract high quality, discontinuity preserving correspondence maps from scenes that are extremely challenging for conventional stereo methods.
ICCV	Fixed Point Probability Field for Complex Occlusion Handling.	François Fleuret,Richard Lengagne,Pascal Fua	2005	In this paper, we show that in a multi-camera context, we can effectively handle occlusions in real-time at each frame independently, even when the only available data comes from the binary output of a simple blob detector, and the number of present individuals is a priori unknown. We start from occupancy probability estimates in a top view and rely on a generative model to yield probability images to be compared with the actual input images. We then refine the estimates so that the probability images match the binary input images as well as possible. We demonstrate the quality of our results on several sequences involving complex occlusions.
ICCV	Fusion of Multi-View Silhouette Cues Using a Space Occupancy Grid.	Jean-Sébastien Franco,Edmond Boyer	2005	In this paper, we investigate what can be inferred from several silhouette probability maps, in multi-camera environments. To this aim, we propose a new framework for multi-view silhouette cue fusion. This framework uses a space occupancy grid as a probabilistic 3D representation of scene contents. Such a representation is of great interest for various computer vision applications in perception, or localization for instance. Our main contribution is to introduce the occupancy grid concept, popular in the robotics community, for multi-camera environments. The idea is to consider each camera pixel as a statistical occupancy sensor. All pixel observations are then used jointly to infer where, and how likely, matter is present in the scene. As our results illustrate, this simple model has various advantages. Most sources of uncertainty are explicitly modeled, and no premature decisions about pixel labeling occur, thus preserving pixel knowledge. Consequently, optimal scene object localization, and robust volume reconstruction, can be achieved, with no constraint on camera placement and object visibility. In addition, this representation allows to improve silhouette extraction in images.
ICCV	Robust Structure from Motion and Identified Dynamics.	Benjamin R. Fransen,Octavia I. Camps,Mario Sznaier	2005	This paper addresses robust recovery of structure and motion for rigid bodies in video sequences. For small interframe motion, feature appearance is commonly estimated according to the previous frame. However, in the case of occlusion or corrupt frames, the small inter-frame motion model fails. In this paper, we propose to use robustidentification techniques to estimate the motion dynamics based on a set of previous frames. These dynamics are then recursively used to robustly estimate object structure and motion and to predict the object appearance in future frames. Results are tested for 3d reconstructions of rigid bodies in real and synthetic image sequences.
ICCV	Integrating Representative and Discriminative Models for Object Category Detection.	Mario Fritz,Bastian Leibe,Barbara Caputo,Bernt Schiele	2005	Category detection is a lively area of research. While categorization algorithms tend to agree in using local descriptors, they differ in the choice of the classifier, with some using generative models and others discriminative approaches. This paper presents a method for object category detection which integrates a generative model with a discriminative classifier. For each object category, we generate an appearance codebook, which becomes a common vocabulary for the generative and discriminative methods. Given a query image, the generative part of the algorithm finds a set of hypotheses and estimates their support in location and scale. Then, the discriminative part verifies each hypothesis on the same codebook activations. The new algorithm exploits the strengths of both original methods, minimizing their weaknesses. Experiments on several databases show that our new approach performs better than its building blocks taken separately. Moreover, experiments on two challenging multi-scale databases show that our new algorithm outperforms previously reported results.
ICCV	Efficient Block Noise Removal Based on Nonlinear Manifolds.	Haoying Fu,Hongyuan Zha,Jesse L. Barlow	2005	The problem of block noise removal is considered. It is assumed that the original image is on or close to a sub-space of admissible images in the form of a low dimensional nonlinear manifold. We propose to use a close variant of the total variation regularizer for measuring block noise. Based on this noise measure, we present an effective approach that reconstructs the original image in the presence of block noise. Our main computational task is the solution of a quadratic programming problem, for which we propose a very efficient interior point method. The effectiveness and efficiency of our approach is demonstrated by an example.
ICCV	When Does a Camera See Rain?	Kshitiz Garg,Shree K. Nayar	2005	Rain produces sharp intensity fluctuations in images and videos, which degrade the performance of outdoor vision systems. These intensity fluctuations depend on various factors, such as the camera parameters, the properties of rain, and the brightness of the scene. We show that the properties of rain ¿ its small drop size, high velocity, and low density ¿ make its visibility strongly dependent on camera parameters such as exposure time and depth of field. We show that these parameters can be selected so as to reduce or even remove the effects of rain without altering the appearance of the scene. Conversely, the parameters of a camera can also be set to enhance the visual effects of rain. This can be used to develop an inexpensive and portable camera-based rain gauge that provides instantaneous rain rate measurements. The proposed methods serve to make vision algorithms more robust to rain without any necessity for post-processing. In addition, they can be used to control the visual effects of rain during the filming of movies.
ICCV	Vignette and Exposure Calibration and Compensation.	Dan B. Goldman,Jiun-Hung Chen	2005	"We discuss calibration and removal of ""vignetting"" (radial falloff) and exposure (gain) variations from sequences of images. Unique solutions for vignetting, exposure and scene radiances are possible when the response curve is known. When the response curve is unknown, an exponential ambiguity prevents us from recovering these parameters uniquely. However, the vignetting and exposure variations can nonetheless be removed from the images without resolving this ambiguity. Applications include panoramic image mosaics, photometry for material reconstruction, image-based rendering, and preprocessing for correlation-based vision algorithms."
ICCV	Shape and Spatially-Varying BRDFs from Photometric Stereo.	Dan B. Goldman,Brian Curless,Aaron Hertzmann,Steven M. Seitz	2005	This paper describes a photometric stereo method designed for surfaces with spatially-varying BRDFs, including surfaces with both varying diffuse and specular properties. Our method builds on the observation that most objects are composed of a small number of fundamental materials. This approach recovers not only the shape but also material BRDFs and weight maps, yielding compelling results for a wide variety of objects. We also show examples of interactive lighting and editing operations made possible by our method.
ICCV	A Semi-Supervised Framework for Mapping Data to the Intrinsic Manifold.	Haifeng Gong,Chunhong Pan,Qing Yang,Hanqing Lu,Songde Ma	2005	This paper presents a novel scheme for manifold learning. Different from the previous work reducing data to Euclidean space which cannot handle the looped manifold well, we map the scattered data to its intrinsic parameter manifold by semi-supervised learning. Given a set of partially labeled points, the map to a specified parameter manifold is computed by an iterative neighborhood average method called Anchor Points Diffusion procedure (APD). We explore this idea on the most frequently used close-formed manifolds, Stiefel manifolds whose special cases include hyper sphere and orthogonal group. The experiments show that APD can recover the underlying intrinsic parameters of points on scattered data manifold successfully.
ICCV	The Pyramid Match Kernel: Discriminative Classification with Sets of Image Features.	Kristen Grauman,Trevor Darrell	2005	"Discriminative learning is challenging when examples are sets of features, and the sets vary in cardinality and lack any sort of meaningful ordering. Kernel-based classification methods can learn complex decision boundaries, but a kernel over unordered set inputs must somehow solve for correspondences ¿ generally a computationally expensive task that becomes impractical for large set sizes. We present a new fast kernel function which maps unordered feature sets to multi-resolution histograms and computes a weighted histogram intersection in this space. This ""pyramid match"" computation is linear in the number of features, and it implicitly finds correspondences based on the finest resolution histogram cell where a matched pair first appears. Since the kernel does not penalize the presence of extra features, it is robust to clutter. We show the kernel function is positive-definite, making it valid for use in learning algorithms whose optimal solutions are guaranteed only for Mercer kernels. We demonstrate our algorithm on object recognition tasks and show it to be accurate and dramatically faster than current approaches."
ICCV	Degenerate Cases and Closed-form Solutions for Camera Calibration with One-Dimensional Objects.	Pär Hammarstedt,Peter F. Sturm,Anders Heyden	2005	Camera Calibration with one-dimensional objects is based on an algebraic constraint on the image of the absolute conic. We will give an alternative derivation to this constraint, allowing a geometrical interpretation. From this we derive the degenerate cases, or critical motions, where the calibration algorithm will fail. We also show that constraints on the intrinsic parameters lead to simplified closed-form solutions and a reduced set of critical motions. A simulation and a real data experiment is performed to evaluate the accuracy of the calibration result for motions close to being critical.
ICCV	On-Line Density-Based Appearance Modeling for Object Tracking.	Bohyung Han,Larry S. Davis	2005	Object tracking is a challenging problems in real-time computer vision due to variations of lighting condition, pose, scale, and view-point over time. However, it is exceptionally difficult to model appearance with respect to all of those variations in advance; instead, on-line update algorithms are employed to adapt to these changes. We present a new on-line appearance modeling technique which is based on sequential density approximation. This technique provides accurate and compact representations using Gaussian mixtures, in which the number of Gaussians is automatically determined. This procedure is performed in linear time at each time step, which we prove by amortized analysis. Features for each pixel and rectangular region are modeled together by the proposed sequential density approximation algorithm, and the target model is updated in scale robustly. We show the performance of our method by simulations and tracking in natural videos.
ICCV	On Optimizing Template Matching via Performance Characterization.	Tony X. Han,Visvanathan Ramesh,Ying Zhu,Thomas S. Huang	2005	Template matching is a fundamental operator in computer vision and is widely used in feature tracking, motion estimation, image alignment, and mosaicing. Under a certain parameterized warping model, the traditional template matching algorithm estimates the geometric warp parameters that minimize the SSD between the target and a warped template. The performance of the template matching can be characterized by deriving the distribution of warp parameter estimate as a function of the ideal template, the ideal warp parameters, and a given noise or perturbation model. In this paper, we assume a discretization of the warp parameter space and derive the theoretical expression for the probability mass function (PMF) of the parameter estimate. As the PMF is also a function of the template size, we can optimize the choice of the template or block size by determining the template/block size that gives the estimate with minimum entropy. Experimental results illustrate the correctness of the theory. An experiment involving feature point tracking in face video is shown to illustrate the robustness of the algorithm in a real-world problem.
ICCV	Bottom-up/Top-Down Image Parsing by Attribute Graph Grammar.	Feng Han,Song Chun Zhu	2005	In this paper, we present an attribute graph grammar for image parsing on scenes with man-made objects, such as buildings, hallways, kitchens, and living rooms. We choose one class of primitives ¿ 3D planar rectangles projected on images, and six graph grammar production rules. Each production rule not only expands a node into its components, but also includes a number of equations that constrain the attributes of a parent node and those of its children. Thus our graph grammar is context sensitive. The grammar rules are used recursively to produce a large number of objects and patterns in images and thus the whole graph grammar is a type of generative model. The inference algorithm integrates bottom-up rectangle detection which activates top-down prediction using the grammar rules. The final results are validated in a Bayesian framework. The output of the inference is a hierarchical parsing graph with objects, surfaces, rectangles, and their spatial relations. In the inference, the acceptance of a grammar rule means a recognition of an object, and actions are taken to pass the attributes between a node and its parent through the constraint equations associated with this production rule. When an attribute is passed from a child node to a parent node, it is called bottom-up, and the opposite is called top-down.
ICCV	Multiple Light Sources and Reflectance Property Estimation Based on a Mixture of Spherical Distributions.	Kenji Hara,Ko Nishino,Katsushi Ikeuchi	2005	In this paper, we propose a new method for simultaneously estimating the illumination of the scene and the reflectanceproperty of the object from a single image. We assume that the illumination consists of multiple point sources and the shape of the object is known. Unlike previous methods, we will recover not only the direction and intensity of the light sources, but also the number of light sources and the specular reflection parameter of the object. First, we represent the illumination on the surface of a unit sphere as a finite mixture of von Mises-Fisher distributions by deriving a spherical specular reflection model. Next, we estimate this mixture and the number of distributions. Finally, using this result as initial estimates, we refine the estimates using the original specular reflection model. We can use the results to render the object under novel lighting conditions.
ICCV	Parameter-Free Radial Distortion Correction with Centre of Distortion Estimation.	Richard I. Hartley,Sing Bing Kang	2005	We propose a method of simultaneously calibrating the radial distortion function of a camera alongwith the other internal calibration parameters. The method relies on the use of a planar (or alternatively non-planar) calibration grid, which is captured in several images. In this way, the determination of the radial distortion is an easy add-on to the popular calibration method proposed by Zhang [17]. The method is entirely non-iterative, and hence is extremely rapid and immune from the problem of local minima. Our method determines the radial distortion in a parameter-free way, not relying on any particular radial distortion model. This makes it applicable to a large range of cameras from narrow-angle to fish-eye lenses. The method also computes the centre of radial distortion, which we argue is important in obtaining optimal results. Experiments show that this point may be significantly displaced from the centre of the image, or the principal point of the camera.
ICCV	Sparse Image Coding Using a 3D Non-Negative Tensor Factorization.	Tamir Hazan,Simon Polak,Amnon Shashua	2005	"We introduce an algorithm for a non-negative 3D tensor factorization for the purpose of establishing a local parts feature decomposition from an object class of images. In the past such a decomposition was obtained using non-negative matrix factorization (NMF) where images were vectorized before being factored by NMF. A tensor factorization (NTF) on the other hand preserves the 2D representations of images and provides a unique factorization (unlike NMF which is not unique). The resulting ""factors"" from the NTF factorization are both sparse (like with NMF) but also separable allowing efficient convolution with the test image. Results show a superior decomposition to what an NMF can provide on all fronts ¿ degree of sparsity, lack of ghost residue due to invariant parts and efficiency of coding of around an order of magnitude better. Experiments on using the local parts decomposition for face detection using SVM and Adaboost classifiers demonstrate that the recovered features are discriminatory and highly effective for classification."
ICCV	Neighborhood Preserving Embedding.	Xiaofei He,Deng Cai,Shuicheng Yan,HongJiang Zhang	2005	Recently there has been a lot of interest in geometrically motivated approaches to data analysis in high dimensional spaces. We consider the case where data is drawn from sampling a probability distribution that has support on or near a submanifold of Euclidean space. In this paper, we propose a novel subspace learning algorithm called Neighborhood Preserving Embedding (NPE). Different from Principal Component Analysis (PCA) which aims at preserving the global Euclidean structure, NPE aims at preserving the local neighborhood structure on the data manifold. Therefore, NPE is less sensitive to outliers than PCA. Also, comparing to the recently proposed manifold learning algorithms such as Isomap and Locally Linear Embedding, NPE is defined everywhere, rather than only on the training data points. Furthermore, NPE may be conducted in the original space or in the reproducing kernel Hilbert space into which data points are mapped. This gives rise to kernel NPE. Several experiments on face database demonstrate the effectiveness of our algorithm.
ICCV	Learning Non-Negative Sparse Image Codes by Convex Programming.	Matthias Heiler,Christoph Schnörr	2005	Example-based learning of codes that statistically encode general image classes is of vital importance for computational vision. Recently, non-negative matrix factorization (NMF) was suggested to provide image codes that are both sparse and localized, in contrast to established non-local methods like PCA. In this paper we adopt and generalize this approach to develop a novel learning framework that allows to efficiently compute sparsity-controlled invariant image codes by a well-defined sequence of convex conic programs. Applying the corresponding parameter-free algorithm to various image classes results in semantically relevant and transformation-invariant image representations that are remarkably robust against noise and quantization.
ICCV	3D Shape Recognition and Reconstruction Based on Line Element Geometry.	Michael Hofer,Boris Odehnal,Helmut Pottmann,Tibor Steiner,Johannes Wallner	2005	This paper presents a new method for the recognition and reconstruction of surfaces from 3D data. Line element geometry, which generalizes both line geometry and the Laguerre geometry of oriented planes, enables us to recognize a wide class of surfaces (spiral surfaces, cones, helical surfaces, rotational surfaces, cylinders, etc.) by fitting linear subspaces in an appropriate seven-dimensional image space. In combination with standard techniques such as PCA and RANSAC, line element geometry is employed to effectively perform the segmentation of complex objects according to surface type. Examples show applications in reverse engineering of CAD models and testing mathematical hypotheses concerning the exponential growth of sea shells.
ICCV	Geometric Context from a Single Image.	Derek Hoiem,Alexei A. Efros,Martial Hebert	2005	Many computer vision algorithms limit their performance by ignoring the underlying 3D geometric structure in the image. We show that we can estimate the coarse geometric properties of a scene by learning appearance-based models of geometric classes, even in cluttered natural scenes. Geometric classes describe the 3D orientation of an image region with respect to the camera. We provide a multiple-hypothesis framework for robustly estimating scene structure from a single image and obtaining confidences for each geometric label. These confidences can then be used to improve the performance of many other applications. We provide a thorough quantitative evaluation of our algorithm on a set of outdoor images and demonstrate its usefulness in two applications: object detection and automatic single-view reconstruction.
ICCV	Combining Generative Models and Fisher Kernels for Object Recognition.	Alex Holub,Max Welling,Pietro Perona	2005	Learning models for detecting and classifying object categories is a challenging problem in machine vision. While discriminative approaches to learning and classification have, in principle, superior performance, generative approaches provide many useful features, one of which is the ability to naturally establish explicit correspondence between model components and scene features ¿ this, in turn, allows for the handling of missing data and unsupervised learning in clutter. We explore a hybrid generative/discriminativeapproach using ¿Fisher kernels¿ [1] which retains most of the desirable properties of generative methods, while increasing the classification performance through a discriminative setting. Furthermore, we demonstrate how this kernel framework can be used to combine different types of features and models into a single classifier. Our experiments, conducted on a number of popular benchmarks, show strong performance improvements over the corresponding generative approach and are competitive with the best results reported in the literature.
ICCV	A Multi-Scale Hybrid Linear Model for Lossy Image Representation.	Wei Hong,John Wright,Kun Huang,Yi Ma	2005	This paper introduces a simple and efficient representation for natural images. We partition an image into blocks and treat the blocks as vectors in a high-dimensional space. We then fit a piece-wise linear model (i.e. a union of affine subspaces) to the vectors at each down-sampling scale. We call this a multi-scale hybrid linear model of the image. The hybrid and hierarchical structure of this model allows us effectively to extract and exploit multi-modal correlations among the imagery data at different scales. It conceptually and computationally remedies limitations of many existing image representation methods that are based on either a fixed linear transformation (e.g. DCT, wavelets), an adaptive uni-modal linear transformation (e.g. PCA), or a multi-modal model at a single scale. We will justify both analytically and experimentally why and how such a simple multi-scale hybrid model is able to reduce simultaneously the model complexity and computational cost. Despite a small overhead for the model, our results show that this new model gives more compact representations for a wide variety of natural images under a wide range of signal-to-noise ratio than many existing methods, including wavelets.
ICCV	Vector Boosting for Rotation Invariant Multi-View Face Detection.	Chang Huang,Haizhou Ai,Yuan Li,Shihong Lao	2005	In this paper, we propose a novel tree-structured multi-view face detector (MVFD), which adopts the coarse-to-fine strategy to divide the entire face space into smaller and smaller subspaces. For this purpose, a newly extended boosting algorithm named Vector Boosting is developed to train the predictors for the branching nodes of the tree that have multi-components outputs as vectors. Our MVFD covers a large range of the face space, say, +/- 45° rotation in plane (RIP) and +/- 90° rotation off plane (ROP), and achieves high accuracy and amazing speed (about 40 ms per frame on a 320 × 240 video sequence) compared with previous published works. As a result, by simply rotating the detector 90°, 180° and 270°, a rotation invariant (360° RIP) MVFD is implemented that achieves real time performance (11 fps on a 320 × 240video sequence) with high accuracy.
ICCV	Reconstructing the Geometry of Flowing Water.	Ivo Ihrke,Bastian Goldlücke,Marcus A. Magnor	2005	We present a recording scheme, image formation model and reconstruction method that enables image-based modeling of flowing bodies of water from multi-video input data. The recorded water is dyed with a fluorescent chemical to measure the thickness of a column of water, which leads to an image formation model based on integrated emissivities along a viewing ray. This model allows for a photo-consistency based error measure for a weighted minimal surface, which is recovered using a PDE obtained from the Euler-Lagrangian formulation of the problem. The resulting equation is solved using the level set method.
ICCV	Ensuring Color Consistency across Multiple Cameras.	Adrian Ilie,Greg Welch	2005	Most multi-camera vision applications assume a single common color response for all cameras. However different cameras ¿ even of the same type ¿ can exhibit radically different color responses, and the differences can cause significant errors in scene interpretation. To address this problem we have developed a robust system aimed at inter-camera color consistency. Our method consists of two phases: an iterative closed-loop calibration phase that searches for the per-camera hardware register settings that best balance linearity and dynamic range, followed by a refinement phase that computes the per-camera parametric values for an additional software-based color mapping.
ICCV	Finding Tree Structures by Grouping Symmetries.	Hiroshi Ishikawa,Davi Geiger,Richard Cole	2005	The representation of objects in images as tree structures is of great interest to vision, as they can represent articulated objects such as people as well as other structured objects like arteries in human bodies, roads, circuit board patterns, etc. Tree structures are often related to the symmetry axis representation of shapes, which captures their local symmetries. Algorithms have been introduced to detect (i) open contours in images in quadratic time (ii) closed contours in images in cubic time, and (iii) tree structures from contours in quadratic time. The algorithms are based on dynamic programming and Single Source Shortest Path algorithms. However, in this paper, we show that the problem of finding tree structures in images in a principled manner is a much harder problem. We argue that the optimization problem of finding tree structures in images is essentially equivalent to a variant of the Steiner Tree problem, which is NP-hard. Nevertheless, an approximate polynomial-time algorithm for this problem exists: we apply a fast implementation of the Goemans-Williamson approximate algorithm to the problem of finding a tree representation after an image is transformed by a local symmetry mapping. Examples of extracting tree structures from images illustrate the idea and applicability of the approximate method.
ICCV	Multi-Modal Tensor Face for Simultaneous Super-Resolution and Recognition.	Kui Jia,Shaogang Gong	2005	Face images of non-frontal views under poor illumination with low resolution reduce dramatically face recognition accuracy. This is evident most compellingly by the very low recognition rate of all existing face recognition systems when applied to live CCTV camera input. In this paper, we present a Bayesian framework to perform multi-modal (such as variations in viewpoint and illumination) face image super-resolution for recognition in tensor space. Given a single modal low-resolution face image, we benefit from the multiple factor interactions of training tensor, and super-resolve its high-resolution reconstructions across different modalities for face recognition. Instead of performing pixel-domain super-resolution and recognition independently as two separate sequential processes, we integrate the tasks of super-resolution and recognition by directly computing a maximum likelihood identity parameter vector in high-resolution tensor space for recognition. We show results from multi-modal super-resolution and face recognition experiments across different imaging modalities, using low-resolution images as testing inputs and demonstrate improved recognition rates over standard tensorface and eigenface representations.
ICCV	Eliminating Structure and Intensity Misalignment in Image Stitching.	Jiaya Jia,Chi-Keung Tang	2005	The aim of this paper is to achieve seamless image stitching for eliminating obvious visual artifact caused by severe intensity discrepancy, image distortion and structure mis-alignment, given that the input images are globally registered. Our approach is based on structure deformation and propagation while maintaining the overall appearance affinity of the result to the input images. This new approach is proven to be effective in solving the above problems, and has found applications in mosaic deghosting, image blending and intensity correction. Our new method consists of the following main processes. First, salient features or structures are robustly detected and aligned along the optimal partitioning boundary between the input images. From these features, we derive sparse deformation vectors to uniformly encode the underlying structure and intensity misalignment. These sparse deformation cues will then be propagated robustly and smoothly into the interior of the target image by solving the associated Laplace equations in the image gradient domain. We present convincing results to show that our method can handle significant structure and intensity misalignment in image stitching.
ICCV	A Robust Algorithm for Point Set Registration Using Mixture of Gaussians.	Bing Jian,Baba C. Vemuri	2005	This paper proposes a novel and robust approach to the point set registration problem in the presence of large amounts of noise and outliers. Each of the point sets is represented by a mixture of Gaussians and the point set registration is treated as a problem of aligning the two mixtures. We derive a closed-form expression for the L¿ distance between two Gaussian mixtures, which in turn leads to a computationally efficient registration algorithm. This new algorithm has an intuitive interpretation, is simple to implement and exhibits inherent statistical robustness. Experimental results indicate that our algorithm achieves very good performance in terms of both robustness and accuracy.
ICCV	Detection of Concentric Circles for Camera Calibration.	Guang Jiang,Long Quan	2005	The geometry of plane-based calibration methods is well understood, but some user interaction is often needed in practice for feature detection. This paper presents a fully automatic calibration system that uses patterns of pairs of concentric circles. The key observation is to introduce a geometric method that constructs a sequence of points strictly convergent to the image of the circle center from an arbitrary point. The method automatically detects the points of the pattern features by the construction method, and identify them by invariants. It then takes advantage of homological constraints to consistently and optimally estimate the features in the image. The experiments demonstrate the robustness and the accuracy of the new method.
ICCV	Layered Active Appearance Models.	Eagle Jones,Stefano Soatto	2005	Active appearance models (AAMs) provide a framework for modeling the joint shape and texture of an image. An AAM is a compact representation of both factors in a conditionally linear model. However, the standard AAM framework does not handle images which have missing features, or allow modification of certain structures in the image while leaving neighboring ones undeformed. We introduce the layered active appearance model (LAAM), which allows for missing features, occlusion, substantial spatial rearrangement of features, and which provides a more general representation that extends the applicability of the Active Appearance Model.
ICCV	Creating Efficient Codebooks for Visual Recognition.	Frédéric Jurie,Bill Triggs	2005	Visual codebook based quantization of robust appearance descriptors extracted from local image patches is an effective means of capturing image statistics for texture analysis and scene classification. Codebooks are usually constructed by using a method such as k-means to cluster the descriptor vectors of patches sampled either densely (¿textons¿) or sparsely (¿bags of features¿ based on key-points or salience measures) from a set of training images. This works well for texture analysis in homogeneous images, but the images that arise in natural object recognition tasks have far less uniform statistics. We show that for dense sampling, k-means over-adapts to this, clustering centres almost exclusively around the densest few regions in descriptor space and thus failing to code other informative regions. This gives suboptimal codes that are no better than using randomly selected centres. We describe a scalable acceptance-radius based clusterer that generates better codebooks and study its performance on several image classification tasks. We also show that dense representations outperform equivalent keypoint based ones on these tasks and that SVM or Mutual Information based feature selection starting from a dense codebook further improves the performance.
ICCV	Multiple View Geometry and the L-norm.	Fredrik Kahl	2005	This paper presents a new framework for solving geometric structure and motion problems based on L_¿ -norm. Instead of using the common sum-of-squares cost-function, that is, theL_¿ -norm, the model-fitting errors are measured using the L_¿ -norm. Unlike traditional methods based on L_¿, our framework allows for efficient computation of global estimates. We show that a variety of structure and motion problems, for example, triangulation, camera resectioning and homography estimation can be recast as a quasi-convex optimization problem within this framework. These problems can be efficiently solved using Second Order Cone Programming (SOCP) which is a standard technique in convex optimization. The proposed solutions have been validated on real data in different settings with small and large dimensions and with excellent performance.
ICCV	Globally Optimal Estimates for Geometric Reconstruction Problems.	Fredrik Kahl,Didier Henrion	2005	We introduce a framework for computing statistically optimal estimates of geometric reconstruction problems. While traditional algorithms often suffer from either local minima or non-optimality - or a combination of both - we pursue the goal of achieving global solutions of the statistically optimal cost-function. Our approach is based on a hierarchy of convex relaxations to solve non-convex optimization problems with polynomials. These convex relaxations generate a monotone sequence of lower bounds and we show how one can detect whether the global optimum is attained at a given relaxation. The technique is applied to a number of classical vision problems: triangulation, camera pose, homography estimation and last, but not least, epipolar geometry estimation. Experimental validation on both synthetic and real data is provided. In practice, only a few relaxations are needed for attaining the global optimum.
ICCV	Detection and Tracking of Moving Objects from a Moving Platform in Presence of Strong Parallax.	Jinman Kang,Isaac Cohen,Gérard G. Medioni,Chang Yuan	2005	Detection and Tracking of Moving Objects from a Moving Platform in Presence of Strong Parallax.
ICCV	Consistent Surface Color for Texturing Large Objects in Outdoor Scenes.	Rei Kawakami,Katsushi Ikeuchi,Robby T. Tan	2005	Color appearance of an object is significantly influenced by the color of the illumination. When the illumination color changes, the color appearance of the object will change accordingly, causing its appearance to be inconsistent. To arrive at color constancy, we have developed a physics-based method of estimating and removing the illumination color. In this paper, we focus on the use of this method to deal with outdoor scenes, since very few physics-based methods have successfully handled outdoor color constancy. Our method is principally based on shadowed and non-shadowed regions. Previously researchers have discovered that shadowed regions are illuminated by sky light, while non-shadowed regions are illuminated by a combination of sky light and sunlight. Based on this difference of illumination, we estimate the illumination colors (both the sunlight and the sky light) and then remove them. To reliably estimate the illumination colors in outdoor scenes, we include the analysis of noise, since the presence of noise is inevitable in natural images. As a result, compared to existing methods, the proposed method is more effective and robust in handling outdoor scenes. In addition, the proposed method requires only a single input image, making it useful for many applications of computer vision.
ICCV	Dynamic Measurement Clustering to Aid Real Time Tracking.	Christopher Kemp,Tom Drummond	2005	We present a technique for clustering measurements such that high-dimensional parameter estimation problems can be simplified. The key idea is to find rows of the measurement Jacobian whose rank is significantly less than its width. Such a set of rows gives a cluster of measurements which is affected only by a subset of the parameter space. This cluster can be used independently from other measurements to isolate parameter decisions. Unlike static partitioning techniques, the method presented dynamically generates clusters at each step of the estimation. This achieves substantial computational reductions, even for problems which cannot be partitioned in the traditional sense. The technique is applied to the task of tracking camera motions in real-time and video sequences are used to compare the resulting system to previous methods.
ICCV	Passive Photometric Stereo from Motion.	Jongwoo Lim,Jeffrey Ho,Ming-Hsuan Yang,David J. Kriegman	2005	We introduce an iterative algorithm for shape reconstruction from multiple images of a moving (Lambertian) object illuminated by distant (and possibly time varying) lighting. Starting with an initial piecewise linear surface, the algorithm iteratively estimates a new surface based on the previous surface estimate and the photometric information available from the input image sequence. During each iteration, standard photometric stereo techniques are applied to estimate the surface normals up to an unknown generalized bas-relief transform, and a new surface is computed by integrating the estimated normals. The algorithm essentially consists of a sequence of matrix factorizations (of intensity values) followed by minimization using gradient descent (integration of the normals). Conceptually, the algorithm admits a clear geometric interpretation, which is used to provide a qualitative analysis of the algorithm¿s convergence. Implementation-wise, it is straightforward, being based on several established photometric stereo and structure from motion algorithms. We demonstrate experimentally the effectiveness of our algorithm using several videos of hand-held objects moving in front of a fixed light and camera.
ICCV	An Integrated Framework for Image Segmentation and Perceptual Grouping.	Zhuowen Tu	2005	"This paper presents an efficient algorithm for image segmentation and a framework for perceptual grouping. It makes an attempt to provide one way of combining bottom-up and top-down approaches. In image segmentation, it generalizes the Swendsen-Wang cut algorithm [1] (SWC) to make both 2-way and m-way cuts, and includes topology change processes (graph repartitioning and boundary diffusion). The method directly works at a low temperature without using annealing. We show that it is much faster than the DDMCMC approach [12] and more robust than the SWC method. The results are demonstrated on the Berkeley data set [7]. In perceptual grouping, it integrates discriminative model learning/computing, a belief propagation algorithm (BP) [15] , and SWC into a three-layer computing framework. These methods are realized as different levels of approximation to an ""ideal"" generative model. We demonstrate the algorithm on the problem of human body configuration."
ICCV	Probabilistic Boosting-Tree: Learning Discriminative Models for Classification, Recognition, and Clustering.	Zhuowen Tu	2005	In this paper, a new learning framework¿probabilistic boosting-tree (PBT), is proposed for learning two-class and multi-class discriminative models. In the learning stage, the probabilistic boosting-tree automatically constructs a tree in which each node combines a number of weak classifiers (evidence, knowledge) into a strong classifier (a conditional posterior probability). It approaches the target posterior distribution by data augmentation (tree expansion) through a divide-and-conquer strategy. In the testing stage, the conditional probability is computed at each tree node based on the learned classifier, which guides the probability propagation in its sub-trees. The top node of the tree therefore outputs the overall posterior probability by integrating the probabilities gathered from its sub-trees. Also, clustering is naturally embedded in the learning phase and each sub-tree represents a cluster of certain level. The proposed framework is very general and it has interesting connections to a number of existing methods such as the A* algorithm, decision tree algorithms, generative models, and cascade approaches. In this paper, we show the applications of PBT for classification, detection, object recognition. We have also applied the framework in segmentation.
ICCV	A Practical Single Image Based Approach for Estimating Illumination Distribution from Shadows.	Taeone Kim,Ki-Sang Hong	2005	This paper presents a practical method that estimates illumination distribution from shadows where the shadows are assumed to be cast on a textured, Lambertian surface. Previous methods usually require that the reflectance property of the surface be constant or uniform, or need an additional image to cancel out the effects of varying albedo of the textured surface. We deal with an estimation problem for which surface albedo information is not available. In this case, the estimation problem corresponds to an underdetermined one. We show that combination of regularization by correlation and some user-specified information can be a practical method for solving the problem. In addition, as an optimization tool for solving the problem, we develop a constrained Non-Negative Quadratic Programming (NNQP) technique into which not only regularization but also user-specified information are easily incorporated. We test and validate our method on both synthetic and real images and present some experimental results.
ICCV	Opaque Document Imaging: Building Images of Inaccessible Texts.	Yun Lin,W. Brent Seales	2005	This paper introduces a method for building a readable image of an opaque, rolled or folded text from a volumetric, penetrating scan. The problem is framed by localizing, constructing, and manipulating an image induced by a surface embedded in a 3D voxel space. There are two central contributions that lead to the demonstrated results. First is an energy-based texture formation algorithm, which is a function of voxel intensities and the geometry of the embedded surface. Second is a regularization algorithm based on a constrained mapping of the embedded surface to a regularized image plane. The mapping preserves angles and lengths, which minimizes the distortion of text in the image. The experimental results show readable images derived from custom, high resolution (X-ray-based) CT scans of rolled papyrus and ink samples. These methods are significant for scholars seeking to study inaccessible texts, and may lead to viable techniques for scanning everyday opaque objects (books) without opening them.
ICCV	Coupled Space Learning for Image Style Transformation.	Dahua Lin,Xiaoou Tang	2005	In this paper, we present a new learning framework for image style transforms. Considering that the images in different style representations constitute different vector spaces, we propose a novel framework called Coupled Space Learning to learn the relations between different spaces and use them to infer the images from one style to another style. Observing that for each style, only the components correlated to the space of the target style are useful for inference, we first develop the Correlative Component Analysis to pursue the embedded hidden subspaces that best preserve the inter-space correlation information. Then we develop the Coupled Bidirectional Transform algorithm to estimate the transforms between the two embedded spaces, where the coupling between the forward transform and the backward transform is explicitly taken into account. To enhance the capability of modelling complex data, we further develop the Coupled Gaussian Mixture Model to generalize our framework to a mixture-model architecture. The effectiveness of the framework is demonstrated in the applications including face super-resolution and bidirectional portrait style transforms.
ICCV	Deformation Invariant Image Matching.	Haibin Ling,David W. Jacobs	2005	We propose a novel framework to build descriptors of local intensity that are invariant to general deformations. In this framework, an image is embedded as a 2D surface in 3D space, with intensity weighted relative to distance in x-y. We show that as this weight increases, geodesic distances on the embedded surface are less affected by image deformations. In the limit, distances are deformation invariant. We use geodesic sampling to get neighborhood samples for interest points, then use a geodesic-intensity histogram (GIH) as a deformation invariant local descriptor. In addition to its invariance, the new descriptor automatically finds its support region. This means it can safely gather information from a large neighborhood to improve discriminability. Furthermore, we propose a matching method for this descriptor that is invariant to affine lighting changes. We have tested this new descriptor on interest point matching for two data sets, one with synthetic deformation and lighting change, another with real non-affine deformations. Our method shows promising matching results compared to several other approaches.
ICCV	Effciently Solving Dynamic Markov Random Fields Using Graph Cuts.	Pushmeet Kohli,Philip H. S. Torr	2005	In this paper we present a fast new fully dynamic algorithm for the st-mincut/max-flow problem. We show how this algorithm can be used to efficiently compute MAP estimates for dynamically changing MRF models of labelling problems in computer vision, such as image segmentation. Specifically, given the solution of the max-flow problem on a graph, we show how to efficiently compute the maximum flow in a modified version of the graph. Our experiments showed that the time taken by our algorithm is roughly proportional to the number of edges whose weights were different in the two graphs. We test the performance of our algorithmon one particular problem: the object-background segmentation problem for video and compare it with the best known st-mincut algorithm. The results show that the dynamic graph cut algorithm is much faster than its static counterpart and enables real time image segmentation. It should be noted that our method is generic and can be used to yield similar improvements in many other cases that involve dynamic change in the graph.
ICCV	An Expectation Maximization Approach to the Synergy between Image Segmentation and Object Categorization.	Iasonas Kokkinos,Petros Maragos	2005	In this work we deal with the problem of modelling and exploiting the interaction between the processes of image segmentation and object categorization. We propose a novel framework to address this problem that is based on the combination of the Expectation Maximization (EM) algorithm and generative models for object categories. Using a concise formulation of the interaction between these two processes, segmentation is interpreted as the E step, assigning observations to models, whereas object detection/analysis is modelled as the M-step, fitting models to observations. We present in detail the segmentation and detection processes comprising the E and M steps and demonstrate results on the joint detection and segmentation of the object categories of faces and cars.
ICCV	What Metrics Can Be Approximated by Geo-Cuts, Or Global Optimization of Length/Area and Flux.	Vladimir Kolmogorov,Yuri Boykov	2005	"In [3] we showed that graph cuts can find hyper-surfaces of globally minimal length (or area) under any Riemannian metric. Here we show that graph cuts on directed regular grids can approximate a significantly more general class of continuous non-symmetric metrics. Using submodularity condition [1, 11], we obtain a tight characterization of graph-representable metrics. Such ""submodular"" metrics have an elegant geometric interpretation via hyper-surface functionals combining length/area and flux. Practically speaking, we extend ""geo-cuts"" algorithm [3] to a wider class of geometrically motivated hypersurface functionals and show how to globally optimize any combination of length/area and flux of a given vector field. The concept of flux was recently introduced into computer vision by [13] but it was mainly studied within variational framework so far. We are first to show that flux can be integrated into graph cuts as well. Combining geometric concepts of flux and length/area within the global optimization framework of graph cuts allows principled discrete segmentation models and advances the state of the art for the graph cuts methods in vision. In particular, we address the ""shrinking"" problem of graph cuts, improve segmentation of long thin objects, and introduce useful shape constraints."
ICCV	A New Framework for Approximate Labeling via Graph Cuts.	Nikos Komodakis,Georgios Tziritas	2005	A new framework is presented that uses tools from duality theory of linear programming to derive graph-cut based combinatorial algorithms for approximating NP-hard classification problems. The derived algorithms include ¿-expansion graph cut techniques merely as a special case, have guaranteed optimality properties even in cases where á-expansion techniques fail to do so and can provide very tight per-instance suboptimality bounds in all occasions.
ICCV	Multi-View AAM Fitting and Camera Calibration.	Seth Koterba,Simon Baker,Iain Matthews,Changbo Hu,Jing Xiao,Jeffrey F. Cohn,Takeo Kanade	2005	In this paper we study the relationship between multiview Active Appearance Model (AAM) fitting and camera calibration. In the first part of the paper we propose an algorithm to calibrate the relative orientation of a set of N > 1 cameras by fitting an AAM to sets of N images. In essence, we use the human face as a (non-rigid) calibration grid. Our algorithm calibrates a set of 2 × 3 weak-perspective camera projection matrices, projections of the world coordinate system origin into the images, depths of the world coordinate system origin, and focal lengths. We demonstrate that the performance of this algorithm is comparable to a standard algorithm using a calibration grid. In the second part of the paper we show how calibrating the cameras improves the performance of multi-view AAM fitting.
ICCV	A Multilevel Banded Graph Cuts Method for Fast Image Segmentation.	Herve Lombaert,Yiyong Sun,Leo Grady,Chenyang Xu	2005	In the short time since publication of Boykov and Jolly¿s seminal paper [3], graph cuts have become well established as a leading method in 2D and 3D semi-automated image segmentation. Although this approach is computationally feasible for many tasks, the memory overhead and supralinear time complexity of leading algorithms results in an excessive computational burden for high-resolution data. In this paper, we introduce a multilevel banded heuristic for computation of graph cuts that is motivated by the well-known narrow band algorithm in level set computation. We perform a number of numerical experiments to show that this heuristic drastically reduces both the running time and the memory consumption of graph cuts while producing nearly the same segmentation result as the conventional graph cuts. Additionally, we are able to characterize the type of segmentation target for which our multilevel banded heuristic will yield different results from the conventional graph cuts. The proposed method has been applied to both 2D and 3D images with promising results.
ICCV	Is Levenberg-Marquardt the Most Efficient Optimization Algorithm for Implementing Bundle Adjustment?.	Manolis I. A. Lourakis,Antonis A. Argyros	2005	In order to obtain optimal 3D structure and viewing parameter estimates, bundle adjustment is often used as the last step of feature-based structure and motion estimation algorithms. Bundle adjustment involves the formulation of a large scale, yet sparse minimization problem, which is traditionally solved using a sparse variant of the Levenberg- Marquardt optimization algorithm that avoids storing and operating on zero entries. This paper argues that considerable computational benefits can be gained by substituting the sparse Levenberg-Marquardt algorithm in the implementation of bundle adjustment with a sparse variant of Powell¿s dog leg non-linear least squares technique. Detailed comparative experimental results provide strong evidence supporting this claim.
ICCV	Shapelets Correlated with Surface Normals Produce Surfaces.	Peter Kovesi	2005	This paper addresses the problem of deducing the surface shape of an object given just the surface normals. Many shape measurement algorithms such as shape from shading and shape from texture only return the surface normals of an object, often with an ambiguity of ¿ in the surface tilt. The surface shape has to be inferred from these normals, typically via some integration process. However, reconstruction through the integration of surface gradients is sensitive to noise and the choice of integration paths across the surface. In addition, existing techniques cannot accommodate ambiguities in tilt. This paper presents a new approach to the reconstruction of surfaces from surface normals using basis functions, referred to here as shapelets. The surface gradients of the shapelets are correlated with the gradients of the surface and the correlations summed to form the reconstruction. This results in a simple reconstruction process that is very robust to noise. Where there is an ambiguity of ¿ in the surface tilt, reconstructions of reduced quality are still possible up to a positive/negative shape ambiguity. Intriguingly, some form of reconstruction is also possible using just slant information.
ICCV	Appearance Modeling Under Geometric Context.	Jian Li,Shaohua Kevin Zhou,Rama Chellappa	2005	We propose a unified framework based on a general definition of geometric transform (GeT) for modeling appearance. GeT represents the appearance by applying designed functionals over certain geometric sets. We show that image warping, Radon transform, trace transform, etc. are special cases of our definition. Moreover, three different types of GeTs are designed to handle deformation, articulation and occlusion and applied to fingerprinting the appearance inside a contour. They include the contour-driven GeT, the feature curve based GeT and selecting functionals to model the appearance inside the convex hull of the contour. A multi-resolution representation that combines both shape and appearance information is also proposed. We apply our approach to image synthesis and object recognition. The proposed approach produces promising results when applied to fingerprinting the appearance of human and body parts despite the challenges due to articulated motion and deformations.
ICCV	Bayesian Autocalibration for Surveillance.	Nils Krahnstoever,Paulo R. S. Mendonça	2005	In the context of visual surveillance of human activity, knowledge about a camera¿s internal and external parameters is useful, as it allows for the establishment of a connection between image and world measurements. Unfortunately, calibration information is rarely available and difficult to obtain after a surveillance system has been installed. In this paper a method for camera autocalibration based on information gathered by tracking people is developed. It brings two main contributions: first, we show how a foot-to-head plane homology can be used to obtain the calibration parameters and then we show an approach how to efficiently estimate initial parameter estimates from measurements; second, we present a Bayesian solution to the calibration problem that can elegantly handle measurement uncertainties, outliers, as well as prior information. It is shown how the full posterior distribution of calibration parameters given the measurements can be estimated, which allows making statements about the accuracy of both the calibration parameters and the measurements involving them.
ICCV	A Hierarchical Field Framework for Unified Context-Based Classification.	Sanjiv Kumar,Martial Hebert	2005	We present a two-layer hierarchical formulation to exploit different levels of contextual information in images for robust classification. Each layer is modeled as a conditional field that allows one to capture arbitrary observation-dependent label interactions. The proposed framework has two main advantages. First, it encodes both the short-range interactions (e.g., pixelwise label smoothing) as well as the long-range interactions (e.g., relative configurations of objects or regions) in a tractable manner. Second, the formulation is general enough to be applied to different domains ranging from pixelwise image labeling to contextual object detection. The parameters of the model are learned using a sequential maximum-likelihood approximation. The benefits of the proposed framework are demonstrated on four different datasets and comparison results are presented.
ICCV	Learning Layered Motion Segmentation of Video.	M. Pawan Kumar,Philip H. S. Torr,Andrew Zisserman	2005	We present an unsupervised approach for learning a layered representation of a scene from a video for motion segmentation. Our method is applicable to any video containing piecewise parametric motion. The learnt model is a composition of layers, which consist of one or more segments. The shape of each segment is represented using a binary matte and its appearance is given by the rgb value for each point belonging to the matte. Included in the model are the effects of image projection, lighting, and motion blur. Furthermore, spatial continuity is explicitly modeled resulting in contiguous segments. Unlike previous approaches, our method does not use reference frame(s) for initialization. The two main contributions of our method are: (i) A novel algorithm for obtaining the initial estimate of the model by dividing the scene into rigidly moving components using efficient loopy belief propagation; and (ii) Refining the initial estimate using ¿ β-swap and ¿-expansion algorithms, which guarantee a strong local minima. Results are presented on several classes of objects with different types of camera motion, e.g. videos of a human walking shot with static or translating cameras. We compare our method with the state of the art and demonstrate significant improvements.
ICCV	Edge-Based Rich Representation for Vehicle Classification.	Xiaoxu Ma,W. Eric L. Grimson	2005	In this paper we propose an approach to vehicle classification under a mid-field surveillance framework. We develop a repeatable and discriminative feature based on edge points and modified SIFT descriptors, and introduce a rich representation for object classes. Experimental results show the proposed approach is promising for vehicle classification in surveillance videos despite great challenges such as limited image size and quality and large intra-class variations. Comparisons demonstrate the proposed approach outperforms other methods.
ICCV	Geometric Invariants and Applications under Catadioptric Camera Model.	Yihong Wu,Zhanyi Hu	2005	This paper presents geometric invariants of points and their applications under central catadioptric camera model. Although the image has severe distortions under the model, we establish some accurate projective geometric invariants of scene points and their image points. These invariants, being functions of principal point, are useful, from which a method for calibrating the camera principal point and a method for recovering planar scene structures are proposed. The main advantage of using these invariants for plane reconstruction is that neither camera motion nor the intrinsic parameters, except for the principal point, is needed. The theoretical correctness of the established invariants and robustness of the proposed methods are demonstrated by experiments. In addition, our results are found to be applicable to some more general camera models other than the catadioptric one.
ICCV	A Theory of Refractive and Specular 3D Shape by Light-Path Triangulation.	Kiriakos N. Kutulakos,Eron Steger	2005	"We investigate the feasibility of reconstructing an arbitrarily-shaped specular scene (refractive or mirror-like) from one or more viewpoints. By reducing shape recovery to the problem of reconstructing individual 3D light paths that cross the image plane, we obtain three key results. First, we show how to compute the depth map of a specular scene from a single viewpoint, when the scene redirects incoming light just once. Second, for scenes where incoming light undergoes two refractions or reflections, we show that three viewpoints are sufficient to enable reconstruction in the general case. Third, we show that it is impossible to reconstruct individual light paths when light is redirected more than twice. Our analysis assumes that, for every point on the image plane, we know at least one 3D point on its light path. This leads to reconstruction algorithms that rely on an ""environment matting"" procedure to establish pixel-to-point correspondences along a light path. Preliminary results for a variety of scenes (mirror, glass, etc.) are also presented."
ICCV	Beyond Trees: Common-Factor Models for 2D Human Pose Recovery.	Xiangyang Lan,Daniel P. Huttenlocher	2005	Tree structured models have been widely used for determining the pose of a human body, from either 2D or 3D data. While such models can effectively represent the kinematic constraints of the skeletal structure, they do not capture additional constraints such as coordination of the limbs. Tree structured models thus miss an important source of information about human body pose, as limb coordination is necessary for balance while standing, walking, or running, as well as being evident in other activities such as dancing and throwing. In this paper we consider the use of undirected graphical models that augment a tree structure with latent variables in order to account for coordination between limbs. We refer to these as common-factor models, since they are constructed by using factor analysis to identify additional correlations in limb position that are not accounted for by the kinematic tree structure. These common-factor models have an underlying tree structure and thus a variant of the standard Viterbi algorithm for a tree can be applied for efficient estimation. We present some experimental results contrasting common-factor models with tree models, and quantify the improvement in pose estimation for 2D image data.
ICCV	Periodic Motion Detection and Segmentation via Approximate Sequence Alignment.	Ivan Laptev,Serge J. Belongie,Patrick Pérez,Josh Wills	2005	A method for defecting and segmenting periodic motion is presented. We exploit periodicity as a cite and detect periodic motion in complex scenes where common methods for rnotion segmentation are likely to fail. We note that periodic motion detection can be seen as an approximate case of sequence alignment where an image sequence is matched to itself over one or more periods of time. To use this observation, we first consider alignment of two video sequences obtained by independently moving cameras. Under assumption of constant translation, the fundamental matrices and the homographies are shown to be time-linear matrix functions. These dynamic quantities can be estimated by matching corresponding space-time points with similar local motion and shape. For periodic motion, we match corresponding points across periods and develop a RANSAC procedure to simultaneously estimate the period and the dynamic geometric transformations between periodic views. Using this method, we demonstrate detection and segmentation of human periodic motion in complex scenes with non-rigid backgrounds, moving camera and motion parallax.
ICCV	A Maximum Entropy Framework for Part-Based Texture and Object Recognition.	Svetlana Lazebnik,Cordelia Schmid,Jean Ponce	2005	This paper presents a probabilistic part-based approach for texture and object recognition. Textures are represented using a part dictionary found by quantizing the appearance of scale- or affine-invariant keypoints. Object classes are represented using a dictionary of composite semi-local parts, or groups of neighboring keypoints with stable and distinctive appearance and geometric layout. A discriminative maximum entropy framework is used to learn the posterior distribution of the class label given the occurrences of parts from the dictionary in the training set. Experiments on two texture and two object databases demonstrate the effectiveness of this framework for visual classification.
ICCV	A Bilinear Illumination Model for Robust Face Recognition.	Jinho Lee,Baback Moghaddam,Hanspeter Pfister,Raghu Machiraju	2005	We present a technique to generate an illumination subspace for arbitrary 3D faces based on the statistics of measured illuminations under variable lighting conditions from many subjects. A bilinear model based on the higher-order singular value decomposition is used to create a compact illumination subspace given arbitrary shape parameters from a parametric 3D face model. Using a fitting procedure based on minimizing the distance of the input image to the dynamically changing illumination subspace, we reconstruct a shape-specific illumination subspace from a single photograph. We use the reconstructed illumination subspace in various face recognition experiments with variable lighting conditions and obtain accuracies which are very competitive with previous methods that require specific training sessions or multiple images of the subject.
ICCV	A Spectral Technique for Correspondence Problems Using Pairwise Constraints.	Marius Leordeanu,Martial Hebert	2005	We present an efficient spectral method for finding consistent correspondences between two sets of features. We build the adjacency matrix M of a graph whose nodes represent the potential correspondences and the weights on the links represent pairwise agreements between potential correspondences. Correct assignments are likely to establish links among each other and thus form a strongly connected cluster. Incorrect correspondences establish links with the other correspondences only accidentally, so they are unlikely to belong to strongly connected clusters. We recover the correct assignments based on how strongly they belong to the main cluster of M, by using the principal eigenvector ofM and imposing the mapping constraints required by the overall correspondence mapping (one-to-one or one-to-many). The experimental evaluation shows that our method is robust to outliers, accurate in terms of matching rate, while being much faster than existing methods.
ICCV	Adaptive Enhancement of Cardiac Magnetic Resonance (CMR) Images.	Erich T. H. Leung,John K. Tsotsos	2005	This paper presents a wavelet-based framework for enhancing the coherent structures attributable to the target organ in cardiac Magnetic Resonance (MR) images. Previous approaches focus on the Rician nature of noise in magnitude MR images. Image noise is but only one of the confounding factors that obscure the anatomical structures of the target organ. This paper models the image noise in a magnitude MR image in terms of two noise classes which occur over different ranges of signal intensity. An adaptive enhancement scheme is developed to achieve simultaneous attenuation of the effects of these factors and improvement in image contrast.
ICCV	Randomized RANSAC with Sequential Probability Ratio Test.	Jiri Matas,Ondrej Chum	2005	A randomized model verification strategy for RANSAC is presented. The proposed method finds, like RANSAC, a solution that is optimal with user-controllable probability n. A provably optimal model verification strategy is designed for the situation when the contamination of data by outliers is known, i.e. the algorithm is the fastest possible (on average) of all randomized RANSAC algorithms guaranteeing 1 - n confidence in the solution. The derivation of the optimality property is based on Wald¿s theory of sequential decision making. The R-RANSAC with SPRT, which does not require the a priori knowledge of the fraction of outliers and has results close to the optimal strategy, is introduced. We show experimentally that on standard test data the method is 2 to 10 times faster than the standard RANSAC and up to 4 times faster than previously published methods.
ICCV	Kernel-Based Multifactor Analysis for Image Synthesis and Recognition.	Yang Li,Yangzhou Du,Xueyin Lin	2005	In many vision problems, the appearances of the observed images, e.g. the human facial images, are often influenced by multiple underlying factors. In this paper, a kernel-based factorization framework is proposed to analyze a multifactor dataset. Specifically, we perform N-mode Singular Value Decomposition (N-mode SVD) in a higher dimensional feature space instead of the input space by using kernel approaches. Given an input sample, its specific underlying factors which may be all absent in the training set can be extracted and translated from one sample to another by using kernel-based ¿translation¿. Therefore our framework is suitable for tasks of new image synthesis and underlying factor recognition. We demonstrate the capabilities of our framework on ensembles of facial images subjected to different person identities, view-points and illuminations with high-quality synthetic faces and high face recognition accuracy.
ICCV	A Generative/Discriminative Learning Algorithm for Image Classification.	Yi Li,Linda G. Shapiro,Jeff A. Bilmes	2005	We have developed a two-phase generative/discriminative learning procedure for the recognition of classes of objects and concepts in outdoor scenes. Our method uses both multiple types of object features and context within the image. The generative phase normalizes the description length of images, which can have an arbitrary number of extracted features of each type. In the discriminative phase, a classifier learns which images, as represented by this fixed-length description, contain the target object. We have tested the approach by comparing it to several other approaches in the literature and by experimenting with several different data sets and combinations of features. Our results, using color, texture, and structure features, show a significant improvement over previously published results in image retrieval. Using salient region features, we are competitive with recent results in object recognition.
ICCV	Local Features for Object Class Recognition.	Krystian Mikolajczyk,Bastian Leibe,Bernt Schiele	2005	In this paper we compare the performance of local detectors and descriptors in the context of object class recognition. Recently, many detectors / descriptors have been evaluated in the context of matching as well as invariance to viewpoint changes [20]. However, it is unclear if these results can be generalized to categorization problems, which require different properties of features. We evaluate 5state-of-the-art scale invariant region detectors and 5 descriptors. Local features are computed for 20 object classes and clustered using hierarchical agglomerative clustering. We measure the quality of appearance clusters and location distributions using entropy as well as precision. We also measure how the clusters generalize from training set to novel test data. Our results indicate that extended SIFT descriptors [22] computed on Hessian-Laplace [20] regions perform best. Second score is obtained by Salient regions [11]. The results also show that these two detectors provide complementary features. The new detectors/descriptorssignificantly improve the performance of a state-of-the art recognition approach [16] in pedestrian detection task.
ICCV	Globally Optimal Solutions for Energy Minimization in Stereo Vision Using Reweighted Belief Propagation.	Talya Meltzer,Chen Yanover,Yair Weiss	2005	A wide range of low level vision problems have been formulated in terms of finding the most probable assignment of a Markov Random Field (or equivalently the lowest energy configuration). Perhaps the most successful example is stereo vision. For the stereo problem, it has been shown that finding the global optimum is NP hard but good results have been obtained using a number of approximate optimization algorithms. In this paper we show that for standard benchmark stereo pairs, the global optimum can be found in about 30 minutes using a variant of the belief propagation (BP) algorithm. We extend previous theoretical results on reweighted belief propagation to account for possible ties in the beliefs and using these results we obtain easily checkable conditions that guarantee that the BP disparities are the global optima. We verify experimentally that these conditions are typically met for the standard benchmark stereo pairs and discuss the implications of our results for further progress in stereo.
ICCV	Detection of Multiple, Partially Occluded Humans in a Single Image by Bayesian Combination of Edgelet Part Detectors.	Bo Wu,Ramakant Nevatia	2005	This paper proposes a method for human detection in crowded scene from static images. An individual human is modeled as an assembly of natural body parts. We introduce edgelet features, which are a new type of silhouette oriented features. Part detectors, based on these features, are learned by a boosting method. Responses of part detectors are combined to form a joint likelihood model that includes cases of multiple, possibly inter-occluded humans. The human detection problem is formulated as maximum a posteriori (MAP) estimation. We show results on a commonly used previous dataset as well as new data-sets that could not be processed by earlier methods.
ICCV	Joint Haar-like Features for Face Detection.	Takeshi Mita,Toshimitsu Kaneko,Osamu Hori	2005	In this paper, we propose a new distinctive feature, called joint Haar-like feature, for detecting faces in images. This is based on co-occurrence of multiple Haar-like features. Feature co-occurrence, which captures the structural similarities within the face class, makes it possible to construct an effective classifier. The joint Haar-like feature can be calculated very fast and has robustness against addition of noise and change in illumination. A face detector is learned by stagewise selection of the joint Haar-like features using AdaBoost. A small number of distinctive features achieve both computational efficiency and accuracy. Experimental results with 5,676 face images and 30,000 nonface images show that our detector yields higher classification performance than Viola and Jones¿ detector, which uses a single feature for each weak classifier. Given the same number of features, our method reduces the error by 37%. Our detector is 2.6 times as fast as Viola and Jones¿ detector to achieve the same performance.
ICCV	Shape and Appearance Repair for Incomplete Point Surfaces.	Seyoun Park,Xiaohu Guo,Hayong Shin,Hong Qin	2005	This paper presents a new surface content completion framework that can restore both shape and appearance from scanned, incomplete point set inputs. First, the geometric holes can be robustly identified from noisy and defective data sets without the need of any normal or orientation information, using the method of active deformable models. The geometry and texture information of the holes can then be determined either automatically from the models¿ context, or semi-automatically with minimal users¿ intervention. The central idea for this repair process is to establish a quantitative similarity measurement among local surface patches based on their local parameterizations and curvature computation. The geometry and texture information of each hole can be completed by warping the candidate region and gluing it to the hole. The displacement for the alignment process is computed by solving a Poisson equation in 2D. Our experiments show that the unified framework, founded upon the techniques of deformable models, local parameterization, and PDE modeling, can provide a robust and elegant solution for content completion of defective, complex point surfaces.
ICCV	Evaluation of Features Detectors and Descriptors Based on 3D Objects.	Pierre Moreels,Pietro Perona	2005	We explore the performance of a number of popular feature detectors and descriptors in matching 3D object features across viewpoints and lighting conditions. To this end we design a method, based on intersecting epipolar constraints, for providing ground truth correspondence automatically. These correspondences are based purely on geometric information, and do not rely on the choice of a specific feature appearance descriptor. We test detector-descriptor combinations on a database of 100 objects viewed from 144 calibrated viewpoints under three different lighting conditions. We find that the combination of Hessian-affine feature finder and SIFT features is most robust to viewpoint change. Harris-affine combined with SIFT and Hessian-affine combined with shape context descriptors were best respectively for lighting change and change in camera focal length. We also find that no detector-descriptor combination performs well with viewpoint changes of more than 25---30¿.
ICCV	A Bayesian Approach for Shadow Extraction from a Single Image.	Tai-Pang Wu,Chi-Keung Tang	2005	This paper addresses the problem of shadow extraction from a single image of a complex natural scene. No simplifying assumption on the camera and the light source other than the Lambertian assumption is used. Our method is unique because it is capable of translating very rough user-supplied hints into the effective likelihood and prior functions for our Bayesian optimization. The likelihood function requires a decent estimation of the shadowless image, which is obtained by solving the associated Poisson equation. Our Bayesian framework allows for the optimal extraction of smooth shadows while preserving texture appearance under the extracted shadow. Thus our technique can be applied to shadow removal, producing some best results to date compared with the current state-of-the-art techniques using a single input image. We propose related applications in shadow compositing and image repair using our Bayesian technique.
ICCV	Integration of Conditionally Dependent Object Features for Robust Figure/Background Segmentation.	Francesc Moreno-Noguer,Alberto Sanfeliu,Dimitris Samaras	2005	We propose a new technique for fusing multiple cues to robustly segment an object from its background in video sequences that suffer from abrupt changes of both illumination and position of the target. Robustness is achieved by the integration of appearance and geometric object features and by their description using particle filters. Previous approaches assume independence of the object cues or apply the particle filter formulation to only one of the features, and assume a smooth change in the rest, which can prove is very limiting, especially when the state of some features needs to be updated using other cues or when their dynamics follow non-linear and unpredictable paths. Our technique offers a general framework to model the probabilistic relationship between features. The proposed method is analytically justified and applied to develop a robust tracking system that adapts online and simultaneously the colorspace where the image points are represented, the color distributions, and the contour of the object. Results with synthetic data and real video sequences demonstrate the robustness and versatility of our method.
ICCV	Guiding Model Search Using Segmentation.	Greg Mori	2005	In this paper we show how a segmentation as preprocessing paradigm can be used to improve the efficiency and accuracy of model search in an image. We operationalize this idea using an over-segmentation of an image into superpixels. The problem domain we explore is human body pose estimation from still images. The superpixels prove useful in two ways. First, we restrict the joint positions in our human body model to lie at centers of superpixels, which reduces the size of the model search space. In addition, accurate support masks for computing features on half-limbs of the body model are obtained by using agglomerations of superpixels as half-limb segments. We present results on a challenging dataset of people in sports news images.
ICCV	Dynamic Refraction Stereo.	Nigel J. W. Morris,Kiriakos N. Kutulakos	2005	"In this paper we consider the problem of reconstructing the 3D position and surface normal of points on an unknown, arbitrarily-shaped refractive surface. We show that two viewpoints are sufficient to solve this problem in the general case, even if the refractive index is unknown. The key requirements are (1) knowledge of a function that maps each point on the two image planes to a known 3D point that refracts to it, and (2) light is refracted only once. We apply this result to the problem of reconstructing the time-varying surface of a liquid from patterns placed below it. To do this, we introduce a novel ""stereo matching"" criterion called refractive disparity, appropriate for refractive scenes, and develop an optimization-based algorithm for individually reconstructing the position and normal of each point projecting to a pixel in the input views. Results on reconstructing a variety of complex, deforming liquid surfaces suggest that our technique can yield detailed reconstructions that capture the dynamic behavior of free-flowing liquids."
ICCV	Preface.		2005	Preface.
ICCV	Objective Image Fusion Performance Characterisation.	Vladimir S. Petrovic,Costas S. Xydeas	2005	Image fusion as a way of combining multiple image signals into a single fused image has in recent years been extensively researched for a variety of multisensor applications. Choosing an optimal fusion approach for each application from the plethora of algorithms available however, remains a largely open issue. A small number of metrics proposed so far provide only a rough, numerical estimate of fusion performance with limited understanding of the relative merits of different fusion schemes. This paper proposes a method for comprehensive, objective, image fusion performance characterisation using a fusion evaluation framework based on gradient information representation. The method provides an in-depth analysis of fusion performance by quantifying: information contributions by each sensor, fusion gain, fusion information loss and fusion artifacts (artificial information created). It is demonstrated on the evaluation of an extensive dataset of multisensor images fused with a wide range of established image fusion algorithms. The results demonstrate and quantify a number of well known issues concerning the performance of these schemes and provide a useful insight into a number of more subtle yet important fusion performance effects not immediately accessible to an observer.
ICCV	Combining Image Regions and Human Activity for Indirect Object Recognition in Indoor Wide-Angle Views.	Patrick Peursum,Geoff A. W. West,Svetha Venkatesh	2005	Traditional methods of object recognition are reliant on shape and so are very difficult to apply in cluttered, wide-angle and low-detail views such as surveillance scenes. To address this, a method of indirect object recognition is proposed, where human activity is used to infer both the location and identity of objects. No shape analysis is necessary. The concept is dubbed ¿interaction signatures¿, since the premise is that a human will interact with objects in ways characteristic of the function of that object ¿ for example, a person sits in a chair and drinks from a cup. The human-centred approach means that recognition is possible in low-detail views and is largely invariant to the shape of objects within the same functional class. This paper implements a Bayesian network for classifying region patches with object labels, building upon our previous work in automatically segmenting and recognising a human¿s interactions with the objects. Experiments show that interaction signatures can successfully find and label objects in low-detail views and are equally effective at recognising test objects that differ markedly in appearance from the training objects.
ICCV	N-Dimensional Probablility Density Function Transfer and its Application to Colour Transfer.	François Pitié,Anil C. Kokaram,Rozenn Dahyot	2005	This article proposes an original method to estimate a continuous transformation that maps one N-dimensional distribution to another. The method is iterative, non-linear, and is shown to converge. Only 1D marginal distributions are used in the estimation process, hence involving low computation costs. As an illustration this mapping is applied to colour transfer between two images of different contents. The paper also serves as a central focal point for collecting together the research activity in this area and relating it to the important problem of Automated Colour Grading.
ICCV	A Shape-Based Segmentation Approach: An Improved Technique Using Level Sets.	Hossam E. Abd El Munim,Aly A. Farag	2005	"We propose a novel approach for shape-based segmentation based on a specially designed level set function format. This format permits us to better control the process of object registration which is an important part in the shapebased segmentation framework. The method depends on a set of training shapes used to build a parametric shape model. The color is taken into consideration besides the shape prior information. The shape model is fitted to the image volume by registration through an energy minimization problem. The approach overcomes the conventional methods problems like point correspondences and weighing coefficients tuning of the partial differential equations (PDE's). Also it is suitable for multi-dimensional data and computationally efficient. Results of extracting the 2D star fish and the brain ventricles in 3D demonstrate theefficiency of the approach."
ICCV	Efficient Model-Based 3D Tracking of Deformable Objects.	Enrique Muñoz,José Miguel Buenaposada,Luis Baumela	2005	Efficient incremental image alignment is a topic of renewed interest in the computer vision community because of its applications in model fitting and model-based object tracking. Successful compositional procedures for aligning 2D and 3D models under weak-perspective imaging conditions have already been proposed. Here we present a mixed compositional and additive algorithm which is applicable to the full projective camera case.
ICCV	Mesh Optimization Using an Inconsistency Detection Template.	Atsutada Nakatsuji,Yasuyuki Sugaya,Ken-ichi Kanatani	2005	We propose a new technique for optimizing a triangular mesh for polyhedral representation of the scene in a video stream. We introduce a specially designed template that can effectively detect color and texture discontinuities. Using real images, we demonstrate that our method is superior to existing methods.
ICCV	Structured Light in Scattering Media.	Srinivasa G. Narasimhan,Shree K. Nayar,Bo Sun,Sanjeev J. Koppal	2005	Virtually all structured light methods assume that the scene and the sources are immersed in pure air and that light is neither scattered nor absorbed. Recently, however, structured lighting has found growing application in underwater and aerial imaging, where scattering effects cannot be ignored. In this paper, we present a comprehensive analysis of two representative methods - light stripe range scanning and photometric stereo - in the presence of scattering. For both methods, we derive physical models for the appearances of a surface immersed in a scattering medium. Based on these models, we present results on (a) the condition for object detectability in light striping and (b) the number of sources required for photometric stereo. In both cases, we demonstrate that while traditional methods fail when scattering is significant, our methods accurately recover the scene (depths, normals, albedos) as well as the properties of the medium. These results are in turn used to restore the appearances of scenes as if they were captured in clear air. Although we have focused on light striping and photometric stereo, our approach can also be extended to other methods such as grid coding, gated and active polarization imaging.
ICCV	Organizers.		2005	Organizers.
ICCV	Shadow Flow: A Recursive Method to Learn Moving Cast Shadows.	Fatih Porikli,Jay Thornton	2005	We present a novel algorithm to detect and remove cast shadows in a video sequence by taking advantage of the statistical prevalence of the shadowed regions over the object regions. We model shadows using multivariate Gaussians. We apply a weak classifier as a pre-filter.We project shadow models into a quantized color space to update a shadow flow function. We use shadow flow, background models, and current frame to determine the shadow and object regions. This method has several advantages: It does not require a color space transformation. We pose the problem in the RGB color space, and we can carry out the same analysis in other Cartesian spaces as well. It is data-driven and adapts to the changing shadow conditions. In other words, accuracy of our method is not limited by the preset values. Furthermore, it does not assume any 3D models for the target objects or tracking of the cast shadows between frames. Our results show that the detection performance is superior than the benchmark method.
ICCV	Improved Sub-pixel Stereo Correspondences through Symmetric Refinement.	Diego Nehab,Szymon Rusinkiewicz,James Davis	2005	Most dense stereo correspondence algorithms start by establishing discrete pixel matches and later refine these matches to sub-pixel precision. Traditional sub-pixel refinement methods attempt to determine the precise location of points, in the secondary image, that correspond to discrete positions in the reference image. We show that this strategy can lead to a systematic bias associated with the violation of the general symmetry of matching cost functions. This bias produces random or coherent noise in the final reconstruction, but can be avoided by refining both image coordinates simultaneously, in a symmetric way. We demonstrate that the symmetric sub-pixel refinement strategy results in more accurate correspondences by avoiding bias while preserving detail.
ICCV	Detecting Rotational Symmetries.	V. Shiv Naga Prasad,Larry S. Davis	2005	We present an algorithm for detecting multiple rotational symmetries in natural images. Given an image, its gradient magnitude field is computed, and information from the gradients is spread using a diffusion process in the form of a Gradient Vector Flow (GVF) field. We construct a graph whose nodes correspond to pixels in the image, connecting points that are likely to be rotated versions of one another. The n-cycles present in the graph are made to vote for C_n symmetries, their votes being weighted by the errors in transformation between GVF in the neighborhood of the voting points, and the irregularity of the n-sided polygons formed by the voters. The votes are accumulated at the centroids of possible rotational symmetries, generating a confidence map for each order of symmetry. We tested the method with several natural images.
ICCV	Reviewers.		2005	Reviewers.
ICCV	An Enhanced Correlation-Based Method for Stereo Correspondence with Sub-Pixel Accuracy.	Emmanouil Z. Psarakis,Georgios D. Evangelidis	2005	The invariance of the similarity measure in photometric distortions as well as its capability in producing sub-pixel accuracy are two desired and often required features in most stereo vision applications. In this paper we propose a new correlation-based measure which incorporates both mentioned requirements. Specifically, by using an appropriate interpolation scheme in the candidate windows of the matching image, and using the classical zero mean normalized cross correlation function, we introduce a suitable measure. Although the proposed measure is a non-linear function of the sub-pixel displacement parameter, its maximization results in a closed form solution, resulting in reduced complexity for its use in matching techniques. Application of the proposed measure in a number of benchmark stereo pair images reveals its superiority over existing correlation-based techniques used for sub-pixel accuracy.
ICCV	Basic Gray Level Aura Matrices: Theory and its Application to Texture Synthesis.	Xuejie Qin,Yee-Hong Yang	2005	In this paper, we present a new mathematical framework for modeling texture images using independent Basic Gray Level Aura Matrices (BGLAMs). We prove that independent BGLAMs are the basis of Gray Level Aura Matrices (GLAMs), and that an image can be uniquely represented by its independent BGLAMs. We propose a new BGLAM distance measure for automatically evaluating synthesis results w.r.t. input textures to determine if the output is a successful synthesis of the input. For the application to texture synthesis, we present a new algorithm to synthesize textures by sampling only the independent BGLAMs of an input texture. With respect to synthesis of textures and evaluation of the results, the performance of our approach is extensively evaluated and compared with symmetric GLAMs that are used in existing techniques and with Gray Level Cooccurrence Matrices (GLCMs). Experimental results have shown that (1) our approach significantly outperforms both symmetric GLAMs and GLCMs; (2) the new BGLAM distance measure has the ability to evaluate synthesis results, which can be used to automate the conventional visual inspection process for determining whether or not the output texture is a successful synthesis of the input; and (3) a broad range of textures can be faithfully synthesized using independent BGLAMs and the synthesis results are comparable to existing techniques.
ICCV	Face Recognition by Stepwise Nonparametric Margin Maximum Criterion.	Xipeng Qiu,Lide Wu	2005	Linear Discriminant Analysis (LDA) is a popular feature extraction technique in face recognition. However, it often suffers from the small sample size problem when dealing with the high dimensional data. Moreover, while LDA is guaranteed to find the best directions when each class has a Gaussian density with a common covariance matrix, it can fail if the class densities are more general. In this paper, a new nonparametric linear feature extraction method, step-wise nonparametric margin maximum criterion(SNMMC), is proposed to find the most discriminant directions, which does not assume that the class densities belong to any particular parametric family and does not depend on the non-singularity of the within-class scatter matrix either. On three datasets from ATT and FERET face databases, our experimental results demonstrate that SNMMC outperforms other methods and is robust to variations of pose, illumination and expression.
ICCV	Using Eye Reflections for Face Recognition Under Varying Illumination.	Ko Nishino,Peter N. Belhumeur,Shree K. Nayar	2005	Face recognition under varying illumination remains a challenging problem. Much progress has been made toward a solution through methods that require multiple gallery images of each subject under varying illumination. Yet for many applications, this requirement is too severe. In this paper, we propose a novel method that requires only a single gallery image per subject taken under unknown lighting. The method builds upon two contributions. We first estimate the lighting from its reflection in the eyes. This allows us to explicitly recover the illumination in the single gallery images as well as the probe image. Next, we exploit the local linearity of face appearance variation across different people. We represent the gallery images as locally linear montages of images of many different faces taken under the same lighting (bootstrap images). Then, we transfer the estimated combination of bootstrap images to synthesize each subject¿s face under the probe lighting to accomplish recognition. Finally, we show through tests on the CMU PIE database that we can achieve better recognition results using our lighting estimation method and locally linear montages than the current state-of-the-art.
ICCV	Real-Time Interactively Distributed Multi-Object Tracking Using a Magnetic-Inertia Potential Model.	Wei Qu,Dan Schonfeld,Magdi A. Mohamed	2005	"This paper breaks with the common practice of using a joint state space representation and performing the joint data association in multi-object tracking. Instead, we present an interactively distributed framework with linear complexity for real-time applications. When objects do not interact on each other, our approach performs like multiple independent trackers. When the objects are in close proximity or present occlusions, we propose a magnetic-inertia potential model to handle the ""error merge"" and ""labelling"" problems in a particle filtering framework. Specifically, we propose to model the interactive likelihood densities by a ""gravitation"" and ""magnetic"" repulsion scheme and relax the common first-order Markov chain assumption by using an ""Inertia"" Markov chain. Our model represents the cumulative effect of virtual physical forces that objects undergo while interacting with others. It implicitly handles the ""error merge"" and ""labelling"" problems and thus solves the difficult object occlusion and data association problems using an innovative scheme. Our preliminary work has demonstrated that the proposed approach is far superior to existing methods not only in robustness but also in speed."
ICCV	Non-Parametric Self-Calibration.	David Nistér,Henrik Stewénius,Etienne Grossmann	2005	In this paper we develop a theory of non-parametric self-calibration. Recently, schemes have been devised for non-parametric laboratory calibration, but not for self-calibration. We allow an arbitrary warp to model the intrinsic mapping, with the only restriction that the camera is central and that the intrinsic mapping has a well-defined non-singular matrix derivative at a finite number of points under study. We give a number of theoretical results, both forinfinitesimal motion and finite motion, for a finite number of observations and when observing motion over a dense image, for rotation and translation. Our main result is that through observing the flow induced by three instantaneous rotations at a finite number of points of the distorted image, we can perform projective reconstruction of those image points on the undistorted image. We present some results with synthetic and real data.
ICCV	Recovering Photometric Properties of Multiple Strongly-Reflective, Partially-Transparent Surfaces from a Single Image.	José P. Queiroz-Neto,Rodrigo L. Carceroni,Lara C. R. Coelho	2005	This paper introduces a method to recover photometric parameters of a set of 3D surfaces from a single image with significant global-illumination effects such as inter-reflections and transparencies. Since this problem is ambiguous for arbitrary unknown scenes, our formulation assumes that the scene consists of a small set of photometrically homogeneous surfaces with known 3D shapes, illuminated by known light sources. We show that under these conditions, the system of non-linear equations that defines how the image is formed may be factorized into a vector composed only of products of some photometric parameters, and a matrix, whose elements depend non-linearly on both the known illumination, the known 3D shapes and the remaining photometric parameters. This factorization leads to an efficient optimization-based algorithm to compute all unknown photometric parameters from a single input image. Experiments with real data show that this algorithm is more stable and efficient than simpler alternatives.
ICCV	Modeling Scenes with Local Descriptors and Latent Aspects.	Pedro Quelhas,Florent Monay,Jean-Marc Odobez,Daniel Gatica-Perez,Tinne Tuytelaars,Luc J. Van Gool	2005	We present a new approach to model visual scenes in image collections, based on local invariant features and probabilistic latent space models. Our formulation provides answers to three open questions:(1) whether the invariant local features are suitable for scene (rather than object) classification; (2) whether unsupervised latent space models can be used for feature extraction in the classification task; and (3) whether the latent space formulation can discover visual co-occurrence patterns, motivating novel approaches for image organization and segmentation. Using a 9500-image dataset, our approach is validated on each of these issues. First, we show with extensive experiments on binary and multi-class scene classification tasks, that a bag-of-visterm representation, derived from local invariant descriptors, consistently outperforms state-of-the-art approaches. Second, we show that Probabilistic Latent Semantic Analysis (PLSA) generates a compact scene representation, discriminative for accurate classification, and significantly more robust when less training data are available. Third, we have exploited the ability of PLSA to automatically extract visually meaningful aspects, to propose new algorithms for aspect-based image ranking and context-sensitive image segmentation.
ICCV	Learning and Inference in Parametric Switching Linear Dynamical Systems.	Sang Min Oh,James M. Rehg,Tucker R. Balch,Frank Dellaert	2005	We introduce parametric switching linear dynamic systems (P-SLDS) for learning and interpretation of parametrized motion, i.e., motion that exhibits systematic temporal and spatial variations. Our motivating example is the honeybee dance: bees communicate the orientation and distance to food sources through the dance angles and waggle lengths of their stylized dances. Switching linear dynamic systems (SLDS) are a compelling way to model such complex motions. However, SLDS does not provide a means to quantify systematic variations in the motion. Previously, Wilson ¿ Bobick presented parametric HMMs [21], an extension to HMMs with which they successfully interpreted human gestures. Inspired by their work, we similarly extend the standard SLDS model to obtain parametric SLDS. We introduce additional global parameters that represent systematic variations in the motion, and present general expectation-maximization (EM) methods for learning and inference. In the learning phase, P-SLDS learns canonical SLDS model from data. In the inference phase, P-SLDS simultaneously quantifies the global parameters and labels the data. We apply these methods to the automatic interpretation of honey-bee dances, and present both qualitative and quantitative experimental results on actual bee-tracks collected from noisy video data.
ICCV	A Graph Cut Algorithm for Generalized Image Deconvolution.	Ashish Raj,Ramin Zabih	2005	The goal of deconvolution is to recover an image x from its convolution with a known blurring function. This is equivalent to inverting the linear system y = Hx. In this paper we consider the generalized problem where the system matrix H is an arbitrary non-negative matrix. Linear inverse problems can be solved by adding a regularization term to impose spatial smoothness. To avoid oversmoothing, the regularization term must preserve discontinuities; this results in a particularly challenging energy minimization problem. Where H is diagonal, as occurs in image denoising, the energy function can be solved by techniques such as graph cuts, which have proven to be very effective for problems in early vision. When H is non-diagonal, however, the data cost for a pixel to have a intensity depends on the hypothesized intensities of nearby pixels, so existing graph cut methods cannot be applied. This paper shows how to use graph cuts to obtain a discontinuity-preserving solution to a linear inverse system with an arbitrary non-negative system matrix. We use a dynamically chosen approximation to the energy which can be minimized by graph cuts; minimizing this approximation also decreases the original energy. Experimental results are shown for MRI reconstruction from fourier data.
ICCV	On the Equivalence of Common Approaches to Lighting Insensitive Recognition.	Margarita Osadchy,David W. Jacobs,Michael Lindenbaum	2005	Lighting variation is commonly handled by methods invariant to additive and multiplicative changes in image intensity. It has been demonstrated that comparing images using the direction of the gradient can produce broader insensitivity to changes in lighting conditions, even for 3D scenes. We analyze two common approaches to image comparison that are invariant, normalized correlation using small correlation windows, and comparison based on a large set of oriented difference of Gaussian filters. We show analytically that these methods calculate a monotonic (cosine) function of the gradient direction difference and hence are equivalent to the direction of gradient method. Our analysis is supported with experiments on both synthetic and real scenes.
ICCV	Prior-Based Segmentation by Projective Registration and Level Sets.	Tammy Riklin-Raviv,Nahum Kiryati,Nir A. Sochen	2005	Object detection and segmentation can be facilitated by the availability of a reference object. However, accounting for possible transformations between the different object views, as part of the segmentation process, remains a challenge. Recent works address this problem by using comprehensive training data. Other approaches are applicable only to limited object classes or can only accommodate similarity transformations. We suggest a novel variational approach to prior-based segmentation, which accounts for planar projective transformation, using a single reference object. The prior shape is registered concurrently with the segmentation process, without point correspondence. The algorithm detects the object of interest and correctly extracts its boundaries. The homography between the two object views is accurately recovered as well. Extending the Chan-Vese level set framework, we propose a region-based segmentation functional that includes explicit representation of the projective homography between the prior shape and the shape to segment. The formulation is derived from two-view geometry. Segmentation of a variety of objects is demonstrated and the recovered transformation is verified.
ICCV	Segmentation of Hybrid Motions via Hybrid Quadratic Surface Analysis.	Shankar Rao,Allen Y. Yang,Andrew Wagner,Yi Ma	2005	In this paper, we investigate the mathematical problem underlying segmentation of hybrid motions: Given a series of tracked feature correspondences between two (perspective) images, we seek to segment and estimate multiple motions, possibly of different types (e.g., affine, epipolar, and homography). In order to accomplish this task, we cast the problem into a more general mathematical framework of segmenting data samples drawn from a mixture of linear subspaces and quadratic surfaces. The result is a novel algorithm called Hybrid Quadratic Surface Analysis (HQSA). HQSA uses both the derivatives and Hessians of fitting poly-nomials for the data to separate linear data samples from quadratic data samples. These derivatives and Hessians also lead to important necessary conditions, based on the so-called mutual contraction subspace, to separate data samples on different quadratic surfaces. The algebraic solution we derive is non-iterative and numerically stable. It tolerates moderate noise and can be used in conjunction with outlier removal techniques. We show how to solve the hybrid motion segmentation problem using HQSA, and demonstrate its performance on simulated data with noise and on real perspective images.
ICCV	An Ensemble Prior of Image Structure for Cross-Modal Inference.	S. Ravela,Antonio Torralba,William T. Freeman	2005	In cross-modal inference, we estimate complete fields from noisy and missing observations of one sensory modality using structure found in another sensory modality. This inference problem occurs in several areas including texture reconstruction and reconstruction of geophysical fields. We propose a method for cross-modal inference that simultaneously learns shape recipes between two modalities and estimates missing information by using a prior on image structure gleaned from the alternate modality. In the absence of a physical basis for representing image priors, we use a statistical one that represents correlations in differential features. This is done efficiently using a perturbation sampling scheme. Using just one example of the alternate modality, we produce a factorized ensemble representation of feature correlations that yields efficient solutions to large-sized spatial inference problems. We demonstrate the utility of this approach on cross-modal inference with depth and spectral data.
ICCV	Recovering Human Body Configurations Using Pairwise Constraints between Parts.	Xiaofeng Ren,Alexander C. Berg,Jitendra Malik	2005	The goal of this work is to recover human bodyconfigurations from static images. Without assuming a priori knowledge of scale, pose or appearance, this problem is extremely challenging and demands the use of all possible sources of information. We develop a framework which can incorporate arbitrary pairwise constraints between body parts, such as scale compatibility, relative position, symmetry of clothing and smooth contour connections between parts. We detect candidate body parts from bottom-up using parallelism, and use various pairwise configuration constraints to assemble them together into body configurations. To find the most probable configuration, we solve an Integer Quadratic Programming problem with a standard technique using linear approximations. Approximate IQP allows us to incorporate much more information than the traditional dynamic programming and remains computationally efficient. 15 hand-labeled images are used to train the low-level part detector and learn the pairwise constraints. We show test results on a variety of images.
ICCV	Scale-Invariant Contour Completion Using Conditional Random Fields.	Xiaofeng Ren,Charless Fowlkes,Jitendra Malik	2005	We present a model of curvilinear grouping using piece-wise linear representations of contours and a conditional random field to capture continuity and the frequency of different junction types. Potential completions are generated by building a constrained Delaunay triangulation (CDT) over the set of contours found by a local edge detector. Maximum likelihood parameters for the model are learned from human labeled groundtruth. Using held out test data, we measure how the model, by incorporating continuity structure, improves boundary detection over the local edge detector. We also compare performance with a baseline local classifier that operates on pairs of edgels. Both algorithms consistently dominate the low-level boundary detector at all thresholds. To our knowledge, this is the first time that curvilinear continuity has been shown quantitatively useful for a large variety of natural images. Better boundary detection has immediate application in the problem of object detection and recognition.
ICCV	Vehicle Identification between Non-Overlapping Cameras without Direct Feature Matching.	Ying Shan,Harpreet S. Sawhney,Rakesh Kumar	2005	We propose a novel method for identifying road vehicles between two non-overlapping cameras. The problem is formulated as a same-different classification problem: probability of two vehicle images from two distinct cameras being from the same vehicle or from different vehicles. The key idea is to compute the probability without matching the two vehicle images directly, which is a process vulnerable to drastic appearance and aspect changes. We represent each vehicle image as an embedding amongst representative exemplars of vehicles within the same camera. The embedding is computed as a vector each of whose components is a non-metric distance for a vehicle to an exemplar. The non-metric distances are computed using robust matching of oriented edge images. A set of truthed training examples of same-different vehicle pairings across the two cameras is used to learn a classifier that encodes the probability distributions. A pair of the embeddings representing two vehicles across two cameras are then used to compute the same-different probability. In order for the vehicle exemplars to be representative for both cameras, we also propose a method for jointly selection of corresponding exemplars using the training data. Experiments on observations of over 400 vehicles under drastically illumination and camera conditions demonstrate promising results.
ICCV	Behaviour Understanding in Video: A Combined Method.	Neil Robertson,Ian D. Reid	2005	In this paper we develop a system for human behaviour recognition in video sequences. Human behaviour is modelled as a stochastic sequence of actions. Actions are described by a feature vector comprising both trajectory information (position and velocity), and a set of local motion descriptors. Action recognition is achieved via probabilistic search of image feature databases representing previously seen actions. A HMM which encodes the rules of the scene is used to smooth sequences of actions. High-level behaviour recognition is achieved by computing the likelihood that a set of predefined Hidden Markov Models explains the current action sequence. Thus, human actions and behaviour are represented using a hierarchy of abstraction: from simple actions, to actions with spatio-temporal context, to action sequences and finally general behaviours. While the upper levels all use (parametric) Bayes networks and belief propagation, the lowest level uses non-parametric sampling from a previously learned database of actions. The combined method represents a general framework for human behaviour modelling. In this paper we demonstrate the results chiefly on broadcast tennis sequences for automated video annotation.
ICCV	Face Recognition with MRC-Boosting.	Xun Xu,Thomas S. Huang	2005	In this paper, a novel classification algorithm called MRC-Boosting is proposed. Through aggregating Maximal-Rejection-Classifier features under boosting framework, this algorithm can deal with complicated two-class classification problem, especially for the category called target detection problem where a target class should be discriminated from the surrounding clutter class. MRC-Boosting is efficient since unlike many other boosting based algorithms, at each iteration the optimal feature is computed in closed-form, with neither exhaustive search nor time-consuming numerical optimization. Furthermore, a variant of MRC-Boosting is derived and applied to face recognition. This variant MRC-Boosting algorithm is able to utilize large amount of training samples efficiently, overcoming the difficulty faced by other algorithms like AdaBoost. The effectiveness of the proposed algorithm is validated by face recognition experiments on CMU-PIE database.
ICCV	Integrating the Effects of Motion, Illumination and Structure in Video Sequences.	Yilei Xu,Amit K. Roy Chowdhury	2005	"Most work in computer vision has concentrated on studying the individual effects of motion and illumination on a 3D object. In this paper, we present a theory for combining the effects of motion, illumination, 3D structure, albedo, and camera parameters in a sequence of images obtained by a perspective camera. We show that the set of all Lambertian reflectance functions of a moving object, illuminated by arbitrarily distant light sources, lies ""close"" to a bilinear subspace consisting of nine illumination variables and six motion variables. This result implies that, given an arbitrary video sequence, it is possible to recover the 3D structure, motion and illumination conditions simultaneously using the bilinear subspace formulation. The derivation is based on the intuitive notion that, given an illumination direction, the images of a moving surface cannot change suddenly over a short time period. We experimentally compare the images obtained using our theory with ground truth data and show that the difference is small and acceptable. We also provide experimental results on real data by synthesizing video sequences of a 3D face with various combinations of motion and illumination directions."
ICCV	Phase Field Models and Higher-Order Active Contours.	Marie Rochery,Ian Jermyn,Josiane Zerubia	2005	The representation and modelling of regions is an important topic in computer vision. In this paper, we represent a region via a level set of a ¿phase field¿ function. The function is not constrained, e.g. to be a distance function; nevertheless, phase field energies equivalent to classical active contour energies can be defined. They represent an advantageous alternative to other methods: a linear representation space; ease of implementation (a PDE with no reinitialization); neutral initialization; greater topological freedom. We extend the basic phase field model with terms that reproduce ¿higher-order active contour¿ energies, a powerful way of including prior geometric knowledge in the active contour framework via nonlocal interactions between contour points. In addition to the above advantages, the phase field greatly simplifies the analysis and implementation of the higher-order terms. We define a phase field model that favours regions composed of thin arms meeting at junctions, combine this with image terms, and apply the model to the extraction of line networks from remote sensing images.
ICCV	Object Tracking across Multiple Independently Moving Aerial Cameras.	Yaser Sheikh,Mubarak Shah	2005	A camera mounted on an aerial vehicle provides an excellent means for monitoring large areas of a scene. Utilizing several such cameras on different aerial vehicles allows further flexibility, in terms of increased visual scope and in the pursuit of multiple targets. In this paper, we address the problem of tracking objects across multiple moving airborne cameras. Since the cameras are moving and often widely separated, direct appearance-based or proximity-based constraints cannot be used. Instead, we exploit geometric constraints on the relationship between the motion of each object across cameras, to test multiple correspondence hypotheses, without assuming any prior calibration information. We propose a statistically and geometrically meaningful means of evaluating a hypothesized correspondence between two observations in different cameras. Second, since multiple cameras exist, ensuring coherency in correspondence, i.e. transitive closure is maintained between more than two cameras, is an essential requirement. To ensure such coherency we pose the problem of object tracking across cameras as a k-dimensional matching and use an approximation to find the Maximum Likelihood assignment of correspondence. Third, we show that as a result of tracking objects across the cameras, a concurrent visualization of multiple aerial video streams is possible. Results are shown on a number of real and controlled scenarios with multiple objects observed by multiple cameras, validating our qualitative models.
ICCV	Exploring the Space of a Human Action.	Yaser Sheikh,Mumtaz Sheikh,Mubarak Shah	2005	One of the fundamental challenges of recognizing actions is accounting for the variability that arises when arbitrary cameras capture humans performing actions. In this paper, we explicitly identify three important sources of variability: (1) viewpoint, (2) execution rate, and (3) anthropometry of actors, and propose a model of human actions that allows us to investigate all three. Our hypothesis is that the variability associated with the execution of an action can be closely approximated by a linear combination of action bases in joint spatio-temporal space. We demonstrate that such a model bounds the rank of a matrix of image measurements and that this bound can be used to achieve recognition of actions based only on imaged data. A test employing principal angles between subspaces that is robust to statistical fluctuations in measurement data is presented to find the membership of an instance of an action. The algorithm is applied to recognize several actions, and promising results have been obtained.
ICCV	Fast Global Kernel Density Mode Seeking with Application to Localisation and Tracking.	Chunhua Shen,Michael J. Brooks,Anton van den Hengel	2005	We address the problem of seeking the global mode of a density function using the mean shift algorithm. Mean shift, like other gradient ascent optimisation methods, is susceptible to local maxima, and hence often fails to find the desired global maximum. In this work, we propose a multi-bandwidth mean shift procedure that alleviates this problem, which we term annealed mean shift, as it shares similarities with the annealed importance sampling procedure. The bandwidth of the algorithm plays the same role as the temperature in annealing. We observe that the over-smoothed density function with a sufficiently large bandwidth is uni-modal. Using a continuation principle, the influence of the global peak in the density function is introduced gradually. In this way the global maximum is more reliably located. Generally, the price of this annealing-like procedure is that more iterations are required. Since it is imperative that the computation complexity is minimal in real-time applications such as visual tracking. We propose an accelerated version of the mean shift algorithm. Compared with the conventional mean shift algorithm, the accelerated mean shift can significantly decrease the number of iterations required for convergence. The proposed algorithm is applied to the problems of visual tracking and object localisation. We empirically show on various data sets that the proposed algorithm can reliably find the true object location when the starting position of mean shift is far away from the global maximum, in contrast with the conventional mean shift algorithm that will usually get trapped in a spurious local maximum.
ICCV	Fusing Points and Lines for High Performance Tracking.	Edward Rosten,Tom Drummond	2005	This paper addresses the problem of real-time 3D model-based tracking by combining point-based and edge-based tracking systems. We present a careful analysis of the properties of these two sensor systems and show that this leads to some non-trivial design choices that collectively yield extremely high performance. In particular, we present a method for integrating the two systems and robustly combining the pose estimates they produce. Further we show how on-line learning can be used to improve the performance of feature tracking. Finally, to aid real-time performance, we introduce the FAST feature detector which can perform full-frame feature detection at 400Hz. The combination of these techniques results in a system which is capable of tracking average prediction errors of 200 pixels. This level of robustness allows us to track very rapid motions, such as 50.camera shake at 6Hz.
ICCV	On the Spatial Statistics of Optical Flow.	Stefan Roth,Michael J. Black	2005	We develop a method for learning the spatial statistics of optical flow fields from a novel training database. Training flow fields are constructed using range images of natural scenes and 3D camera motions recovered from hand-held and car-mounted video sequences. A detailed analysis of optical flow statistics in natural scenes is presented and machine learning methods are developed to learn a Markov random field model of optical flow. The prior probability of a flow field is formulated as a Field-of-Experts model that captures the higher order spatial statistics in overlapping patches and is trained using contrastive divergence. This new optical flow prior is compared with previous robust priors and is incorporated into a recent, accurate algorithm for dense optical flow computation. Experiments with natural and synthetic sequences illustrate how the learned optical flow prior quantitatively improves flow accuracy and how it captures the rich spatial structure found in natural scene motion.
ICCV	Learning Non-Generative Grammatical Models for Document Analysis.	Michael Shilman,Percy Liang,Paul A. Viola	2005	We present a general approach for the hierarchical segmentation and labeling of document layout structures. This approach models document layout as a grammar and performs a global search for the optimal parse based on a grammatical cost function. Our contribution is to utilize machine learning to discriminatively select features and set all parameters in the parsing process. Therefore, and unlike many other approaches for layout analysis, ours can easily adapt itself to a variety of document analysis problems. One need only specify the page grammar and provide a set of correctly labeled pages. We apply this technique to two document image analysis tasks: page layout structure extraction and mathematical expression interpretation. Experiments demonstrate that the learned grammars can be used to extract the document structure in 57 files from the UWIII document image database. We also show that the same framework can be used to automatically interpret printed mathematical expressions so as to recreate the original LaTeX.
ICCV	Non-Negative Lighting and Specular Object Recognition.	Sameer Shirdhonkar,David W. Jacobs	2005	Recognition of specular objects is particularly difficult because their appearance is much more sensitive to lighting changes than that of Lambertian objects. We consider an approach in which we use a 3D model to deduce the lighting that best matches the model to the image. In this case, an important constraint is that incident lighting should be non-negative everywhere. In this paper, we propose a new method to enforce this constraint and explore its usefulness in specular object recognition, using the spherical harmonic representation of lighting. The method follows from a novel extension of Szego¿s eigenvalue distribution theorem to spherical harmonics, and uses semidefinite programming to perform a constrained optimization. The new method is faster as well as more accurate than previous methods. Experiments on both synthetic and real data indicate that the constraint can improve recognition of specular objects by better separating the correct and incorrect models.
ICCV	Contour-Based Learning for Object Detection.	Jamie Shotton,Andrew Blake,Roberto Cipolla	2005	We present a novel categorical object detection scheme that uses only local contour-based features. A two-stage, partially supervised learning architecture is proposed: a rudimentary detector is learned from a very small set of segmented images and applied to a larger training set of un-segmented images; the second stage bootstraps these detections to learn an improved classifier while explicitly training against clutter. The detectors are learned with a boosting algorithm which creates a location-sensitive classifier using a discriminative set of features from a randomly chosen dictionary of contour fragments. We present results that are very competitive with other state-of-the-art object detection schemes and show robustness to object articulations, clutter, and occlusion. Our major contributions are the application of boosted local contour-based features for object detection in a partially supervised learning framework, and an efficient new boosting procedure for simultaneously selecting features and estimating per-feature parameters.
ICCV	Fast Texture-Based Tracking and Delineation Using Texture Entropy.	Ali Shahrokni,Tom Drummond,Pascal Fua	2005	We propose a fast texture-segmentation approach to the problem of 2-D and 3-D model-based contour tracking, which is suitable for real-time or interactive applications. Our approach relies on detecting texture boundaries in the direction normal to the contour boundaries and on using a Hidden Markov model to link these boundary points in the other direction. The probabilities that appear in this computation closely relate to texture entropy and Kullback-Leibler divergence, a property we use to compute and update dynamic texture models. We demonstrate results both in the context of interactive 2-D delineation and fast 3-D tracking.
ICCV	Multiperspective Projection and Collineation.	Jingyi Yu,Leonard McMillan	2005	We present theories of multiperspective projection and collineation. Given an arbitrary multiperspective imaging system that captures smoothly varying set of rays, we show how to map the rays onto a 2D ray manifold embedded in a 4D linear vector space. The characteristics of this imaging system, such as its projection, collineation, and image distortions can be analyzed by studying the 2-D tangent planes of this ray manifold. These tangent planes correspond to the recently proposed General Linear Camera (GLC) model. In this paper, we study the imaging process of the GLCs. We show the GLC imaging process can be broken down into two separate stages: the mapping of 3D geometry to rays and the sampling of those rays over an image plane. We derive a closed-form solution to projecting 3D points in a scene to rays in a GLC. A GLC image is created by sampling these rays over an image plane. We develop a notion of GLC collineation analogous to pinhole cameras. GLC collineation describes the transformation between the images of a single GLC due to changes in sampling and image plane selection. We show that general GLC collineations can be characterized by a quartic (4th order) rational function. GLC projection and collineation provides a basis for developing new computer vision algorithms suitable for analyzing a wider range of imaging systems than current methods, based on simple pinhole projection models, permit.
ICCV	Visual Speech Recognition with Loosely Synchronized Feature Streams.	Kate Saenko,Karen Livescu,Michael Siracusa,Kevin Wilson,James R. Glass,Trevor Darrell	2005	We present an approach to detecting and recognizing spoken isolated phrases based solely on visual input. We adopt an architecture that first employs discriminative detection of visual speech and articulatory features, and then performs recognition using a model that accounts for the loose synchronization of the feature streams. Discriminative classifiers detect the subclass of lip appearance corresponding to the presence of speech, and further decompose it into features corresponding to the physical components of articulatory production. These components often evolve in a semi-independent fashion, and conventional viseme-based approaches to recognition fail to capture the resulting co-articulation effects. We present a novel dynamic Bayesian network with a multi-stream structure and observations consisting of articulatory feature classifier scores, which can model varying degrees of co-articulation in a principled way. We evaluate our visual-only recognition system on a command utterance task. We show comparative results on lip detection and speech/nonspeech classification, as well as recognition performance against several baseline systems.
ICCV	Fitting Globally Stabilized Algebraic Surfaces to Range Data.	H. Türker Sahin,Mustafa Unel	2005	Fitting Globally Stabilized Algebraic Surfaces to Range Data.
ICCV	Separating Transparent Layers of Repetitive Dynamic Behaviors.	Bernard Sarel,Michal Irani	2005	In this paper we present an approach for separating two transparent layers of complex non-rigid scene dynamics. The dynamics in one of the layers is assumed to be repetitive, while the other can have any arbitrary dynamics. Such repetitive dynamics includes, among other, human actions in video (e.g., a walking person), or a repetitive musical tune in audio signals. We use a global-to-local space-time alignment approach to detect and align the repetitive behavior. Once aligned, a median operator applied to space-time derivatives is used to recover the intrinsic repeating behavior, and separate it from the other transparent layer. We show results on synthetic and real video sequences. In addition, we show the applicability of our approach to separating mixed audio signals (from a single source).
ICCV	Multi-View Reconstruction Using Photo-consistency and Exact Silhouette Constraints: A Maximum-Flow Formulation.	Sudipta N. Sinha,Marc Pollefeys	2005	This paper describes a novel approach for reconstructing a closed continuous surface of an object from multiple calibrated color images and silhouettes. Any accurate reconstruction must satisfy (1) photo-consistency and (2) silhouette consistency constraints. Most existing techniques treat these cues identically in optimization frameworks where silhouette constraints are traded off against photo-consistency and smoothness priors. Our approach strictly enforces silhouette constraints, while optimizing photo-consistency and smoothness in a global graph-cut framework. We transform the reconstruction problem into computing max-flow/mincut in a geometric graph, where any cut corresponds to a surface satisfying exact silhouette constraints (its silhouettes should exactly coincide with those of the visual hull); a minimum cut is the most photo-consistent surface amongst them. Our graph-cut formulation is based on the rim mesh, (the combinatorial arrangement of rims or contour generators from many views) which can be computed directly from the silhouettes. Unlike other methods, our approach enforces silhouette constraints without introducing a bias near the visual hull boundary and also recovers the rim curves. Results are presented for synthetic and real datasets.
ICCV	Discovering Objects and their Localization in Images.	Josef Sivic,Bryan C. Russell,Alexei A. Efros,Andrew Zisserman,William T. Freeman	2005	We seek to discover the object categories depicted in a set of unlabelled images. We achieve this using a model developed in the statistical text literature: probabilistic Latent Semantic Analysis (pLSA). In text analysis this is used to discover topics in a corpus using the bag-of-words document representation. Here we treat object categories as topics, so that an image containing instances of several categories is modeled as a mixture of topics. The model is applied to images by using a visual analogue of a word, formed by vector quantizing SIFT-like region descriptors. The topic discovery approach successfully translates to the visual domain: for a small set of objects, we show that both the object categories and their approximate spatial layout are found without supervision. Performance of this unsupervised method is compared to the supervised approach of Fergus et al. [8] on a set of unseen images containing only one object per image. We also extend the bag-of-words vocabulary to include ¿doublets¿ which encode spatially local co-occurring regions. It is demonstrated that this extended vocabulary gives a cleaner image segmentation. Finally, theclassification and segmentation methods are applied to a set of images containing multiple objects per image. These results demonstrate that we can successfully build object class models from an unsupervised analysis of images.
ICCV	Using Extended Light Sources for Modeling Object Appearance under Varying Illumination.	Imari Sato,Takahiro Okabe,Yoichi Sato,Katsushi Ikeuchi	2005	In this study, we demonstrate the effectiveness of using extended light sources for modeling the appearance of an object for varying illumination. Extended light sources have a radiance distribution that is similar to that of the Gaussian function and have the potential of functioning as a low-pass filter when the appearance of an object is sampled under them. This enables us to obtain a set of basis images of an object for variable illumination from input images of the object taken under those light sources without suffering aliasing caused by insufficient sampling of its appearance. Furthermore, extended light sources are useful in terms of reducing high contrast in image intensities due to specular and diffuse reflection components. This helps us observe both specular and diffuse reflection components of an object in the same image taken with a single shutter speed. We have tested our proposed approach based on extended light sources with objects of complex appearance that are generally difficult to model using image-based modeling techniques.
ICCV	Conditional Random Fields for Contextual Human Motion Recognition.	Cristian Sminchisescu,Atul Kanaujia,Zhiguo Li,Dimitris N. Metaxas	2005	We present algorithms for recognizing human motion in monocular video sequences, based on discriminative Conditional Random Field (CRF) and Maximum Entropy Markov Models (MEMM). Existing approaches to this problem typically use generative (joint) structures like the Hidden Markov Model (HMM). Therefore they have to make simplifying, often unrealistic assumptions on the conditional independence of observations given the motion class labels and cannot accommodate overlapping features or long term contextual dependencies in the observation sequence. In contrast, conditional models like the CRFs seamlessly represent contextual dependencies, support efficient, exact inference using dynamic programming, and their parameters can be trained using convex optimization. We introduce conditional graphical models as complementary tools for human motion recognition and present an extensive set of experiments that show how these typically outperform HMMs in classifying not only diverse human activities like walking, jumping, running, picking or dancing, but also for discriminating among subtle motion styles like normal walk and wander walk.
ICCV	Recovering Facial Shape and Albedo Using a Statistical Model of Surface Normal Direction.	William A. P. Smith,Edwin R. Hancock	2005	This paper describes how facial shape can be modelled using a statistical model that captures variations in surface normal direction. To construct this model we make use of the azimuthal equidistant projection to map surface normals from the unit sphere to points on a local tangent plane. The variations in surface normal direction are captured using the covariance matrix for the projected point positions. This allows us to model variations in face shape using a standard point distribution model. We train the model on fields of surface normals extracted from range data and show how to fit the model to intensity data using constraints on the surface normal direction provided by Lambert¿s law. We demonstrate that this process yields accurate facial shape recovery and allows an estimate of the albedo map to be made from single, real world face images.
ICCV	TemporalBoost for Event Recognition.	Paul Smith,Niels da Vitoria Lobo,Mubarak Shah	2005	This paper contributes a new boosting paradigm to achieve detection of events in video. Previous boosting paradigms in vision focus on single frame detection and do not scale to video events. Thus new concepts need to be introduced to address questions such as determining if an event has occurred, localizing the event, handling same action performed at different speeds, incorporating previous classifier responses into current decision, using temporal consistency of data to aid detection and recognition. The proposed method has the capability to improve weak classifiers by allowing them to use previous history in evaluating the current frame. A learning mechanism built into the boosting paradigm is also given which allows event level decisions to be made. This is contrasted with previous work in boosting which uses limited higher level temporal reasoning and essentially makes object detection decisions at the frame level. Our approach makes extensive use of temporal continuity of video at the classifier and detector levels. We also introduce a relevant set of activity features. Features are evaluated at multiple zoom levels to improve detection. We show results for a system that is able to recognize 11 actions.
ICCV	A Model-Based Vehicle Segmentation Method for Tracking.	Xuefeng Song,Ramakant Nevatia	2005	Our goal is to detect and track moving vehicles on a road observed from cameras placed on poles or buildings. Inter-vehicle occlusion is significant under these conditions and traditional blob tracking methods will be unable to separate the vehicles in the merged blobs. We use vehicle shape models, in addition to camera calibration and ground plane knowledge, to detect, track and classify moving vehicles in presence of occlusion. We use a 2-stage approach. In the first stage, hypothesis for vehicle types, positions and orientations are formed by a coarse search, which is then refined by a data driven Markov Chain Monte Carlo (DDMCMC) process. We show results and evaluations on some real urban traffic video sequence using three types of vehicle models.
ICCV	Manifold Clustering.	Richard Souvenir,Robert Pless	2005	Manifold learning has become a vital tool in data driven methods for interpretation of video, motion capture, and handwritten character data when they lie on a low dimensional, non-linear manifold. This work extends manifold learning to classify and parameterize unlabeled data which lie on multiple, intersecting manifolds. This approach significantly increases the domain to which manifold learning methods can be applied, allowing parameterization of example manifolds such as figure eights and intersecting paths which are quite common in natural data sets. This approach introduces several technical contributions which may be of broader interest, including node-weighted multi-dimensional scaling and a fast algorithm for weighted low-rank approximation for rank-one weight matrices. We show examples for intersecting manifolds of mixed topology and dimension and demonstrations on human motion capture data.
ICCV	Convex Grouping Combining Boundary and Region Information.	Joachim S. Stahl,Song Wang	2005	Convexity is an important geometric property of many natural and man-made structures. Prior research has shown that it is imperative to many perceptual-organization and image-understanding tasks. This paper presents a new grouping method for detecting convex structures from noisy images in a globally optimal fashion. Particularly, this method combines both region and boundary information: the detected structural boundary is closed and well aligned with detected edges while the enclosed region has good intensity homogeneity. We introduce a ratio-form cost function for measuring the structural desirability, which avoids a possible bias to detect small structures. A new fragment-pruning algorithm is developed to achieve the structural convexity. The proposed method can also be extended to detect open boundaries, which correspond to the structures that are partially cropped by the image perimeter, and incorporate a human-computer interaction for detecting a convex boundary around a specified point. We test the proposed method on a set of real images and compare it with the Jacobs¿ convex-grouping method.
ICCV	Spherical Matching for Temporal Correspondence of Non-Rigid Surfaces.	Jonathan Starck,Adrian Hilton	2005	This paper introduces spherical matching to estimate dense temporal correspondence of non-rigid surfaces with genus-zero topology. The spherical domain gives a consistent 2D parameterisation of non-rigid surfaces for matching. Non-rigid 3D surface correspondence is formulated as the recovery of a bijective mapping between two surfaces in the 2D domain. Formulating matching as a 2D bijection guarantees a continuous one-to-one surface correspondence without overfolding. This overcomes limitations of direct estimation of non-rigid surface correspondence in the 3D domain. A multiple resolution coarse-to-fine algorithm is introduced to robustly estimate the dense correspondence which minimises the disparity in shape and appearance between two surfaces. Spherical matching is applied to derive the temporal correspondence between non-rigid surfaces reconstructed at successive frames from multiple view video sequences of people. Dense surface correspondence is recovered across complete motion sequences for both textured and uniform regions, without the requirement for a prior model of human shape or kinematic structure for tracking.
ICCV	A Theory of Inverse Light Transport.	Steven M. Seitz,Yasuyuki Matsushita,Kiriakos N. Kutulakos	2005	"In this paper we consider the problem of computing and removing interreflections in photographs of real scenes. Towards this end, we introduce the problem of inverse light transport ¿ given a photograph of an unknown scene, decompose it into a sum of n-bounce images, where each image records the contribution of light that bounces exactly n times before reaching the camera. We prove the existence of a set of interreflection cancelation operators that enable computing each n-bounce image by multiplying the photograph by a matrix. This matrix is derived from a set of ""impulse images"" obtained by probing the scene with a narrow beam of light. The operators work under unknown and arbitrary illumination, and exist for scenes that have arbitrary spatially-varying BRDFs. We derive a closed-form expression for these operators in the Lambertian case and present experiments with textured and untextured Lambertian scenes that confirm our theory¿s predictions."
ICCV	Efficiently Registering Video into Panoramic Mosaics.	Drew Steedly,Chris Pal,Richard Szeliski	2005	We present an automatic and efficient method to register and stitch thousands of video frames into a large panoramic mosaic. Our method preserves the robustness and accuracy of image stitchers that match all pairs of images while utilizing the ordering information provided by video. We reduce the cost of searching for matches between video frames by adaptively identifying key frames based on the amount of image-to-image overlap. Key frames are matched to all other key frames, but intermediate video frames are only matched to temporally neighboring key frames and intermediate frames. Image orientations can be estimated from this sparse set of matches in time quadratic to cubic in the number of key frames but only linear in the number of intermediate frames. Additionally, the matches between pairs of images are compressed by replacing measurements within small windows in the image with a single representative measurement. We show that this approach substantially reduces the time required to estimate the image orientations with minimal loss of accuracy. Finally, we demonstrate both the efficiency and quality of our results by registering several long video sequences.
ICCV	How Hard is 3-View Triangulation Really?	Henrik Stewénius,Frederik Schaffalitzky,David Nistér	2005	We present a solution for optimal triangulation in three views. The solution is guaranteed to find the optimal solution because it computes all the stationary points of the (maximum likelihood) objective function. Internally, the solution is found by computing roots of multi-variate polynomial equations, directly solving the conditions for stationarity. The solver makes use of standard methods from computational commutative algebra to convert the root-finding problem into a 47 × 47 non-symmetric eigen-problem. Although there are in general 47 roots, counting both real and complex ones, the number of real roots is usually much smaller. We also show experimentally that the number of stationary points that are local minima and lie in front of each camera is small but does depend on the scene geometry.
ICCV	Learning Hierarchical Models of Scenes, Objects, and Parts.	Erik B. Sudderth,Antonio Torralba,William T. Freeman,Alan S. Willsky	2005	We describe a hierarchical probabilistic model for the detection and recognition of objects in cluttered, natural scenes. The model is based on a set of parts which describe the expected appearance and position, in an object centered coordinate frame, of features detected by a low-level interest operator. Each object category then has its own distribution over these parts, which are shared between objects. We learn the parameters of this model via a Gibbs sampler which uses the graphical model¿s structure to analytically average over many parameters. Applied to a database of images of isolated objects, the sharing of parts among objects improves detection accuracy when few training examples are available. We also extend this hierarchical framework to scenes containing multiple objects.
ICCV	Geometric and Photometric Restoration of Distorted Documents.	Mingxuan Sun,Ruigang Yang,Yun Lin,George V. Landon,W. Brent Seales,Michael S. Brown	2005	We present a system to restore the 2D content printed on distorted documents. Our system works by acquiring a 3D scan of the document¿s surface together with a high-resolution image. Using the 3D surface information and the 2D image, we can ameliorate unwanted surface distortion and effects from non-uniform illumination. Our system can process arbitrary geometric distortions, not requiring any pre-assumed parametric models for the document¿s geometry. The illumination correction uses the 3D shape to distinguish content edges from illumination edges to recover the 2D content¿s reflectance image while making no assumptions about light sources and their positions. Results are shown for real objects, demonstrating a complete framework capable of restoring geometric and photometric artifacts on distorted documents.
ICCV	Bi-Directional Tracking Using Trajectory Segment Analysis.	Jian Sun,Weiwei Zhang,Xiaoou Tang,Heung-Yeung Shum	2005	In this paper, we present a novel approach to keyframe-based tracking, called bi-directional tracking. Given two object templates in the beginning and ending keyframes, the bi-directional tracker outputs the MAP (Maximum A Posterior) solution of the whole state sequence of the target object in the Bayesian framework. First, a number of 3D trajectory segments of the object are extracted from the input video, using a novel trajectory segment analysis. Second, these disconnected trajectory segments due to occlusion are linked by a number of inferred occlusion segments. Last, the MAP solution is obtained by trajectory optimization in a coarse-to-fine manner. Experimental results show the robustness of our approach with respect to sudden motion, ambiguity, and short and long periods of occlusion.
ICCV	More-Than-Topology-Preserving Flows for Active Contours and Polygons.	Ganesh Sundaramoorthi,Anthony J. Yezzi	2005	Active contour and active polygon models have been used widely for image segmentation. In some applications, the topology of the object(s) to be detected from an image is known a priori, despite an unknown complex geometry, and it is important that the active contour or polygon maintain the desired topology. In this work, we construct a novel geometric flow that can be added to image based evolutions of active contours and polygons so that the topology of the initial contour or polygon is preserved. Indeed, the proposed geometric flow ensures more than just correct topology; it ensures that the active contour or polygon is, in some sense, kept far away from a topology change. Smoothness properties similar to curvature flow are also guaranteed by the proposed geometric flow. The proposed topology preserving geometric flow is the gradient flow arising from an energy that is based on electrostatic principles. The evolution of a single point on the contour depends on all other points of the contour, which is different from traditional curve evolutions in computer vision literature.
ICCV	"Common Pattern Discovery Using Earth Mover's Distance and Local Flow Maximization."	Hung-Khoon Tan,Chong-Wah Ngo	2005	In this paper, we present a novel segmentation-insensitive approach for mining common patterns from 2 images. We develop an algorithm using the Earth Movers Distance (EMD) framework, unary and adaptive neighborhood color similarity. We then propose a novel local flow maximization approach to provide the best estimation of location and scale of the common pattern. This is achieved by performing an iterative optimization in search of the most stable flows¿ centroid. Common pattern discovery isdifficult owing to the huge search space and problem domain. We intend to solve this problem by reducing the search space through identifying the location and a reduced spatial space for common pattern discovery. Experimental results justify the effectiveness and the potential of the approach.
ICCV	Photometric Stereo under Perspective Projection.	Ariel Tankus,Nahum Kiryati	2005	Photometric stereo is a fundamental approach in Computer Vision. At its core lies a set of image irradiance equations each taken with a different illumination. The vast majority of studies in this field have assumed orthography as the projection model. This paper re-examines the basic set of equations of photometric stereo, under an assumption of perspective projection. We show that the resulting system is linear (as is the case under the orthographic model; Nevertheless, the unknowns are different in the perspective case). We then suggest a simple reconstruction algorithm based on the perspective formulae, and compare it to its orthographic counterpart on synthetic as well as real images. This algorithm obtained lower error rates than the orthographic one in all of the error measures. These findings strengthen the hypothesis that a more realistic set of assumptions, the perspective one, improves reconstruction significantly.
ICCV	Non-Orthogonal Binary Subspace and Its Applications in Computer Vision.	Hai Tao,Ryan Crabb,Feng Tang	2005	This paper presents a novel approach that represents an image or a set of images using a non-orthogonal binary subspace (NBS) spanned by box-like base vectors. These base vectors possess the property that the inner product operation with them can be computed very efficiently. We investigate the optimized orthogonal matching pursuit method for finding the best NBS base vectors. It is demonstrated in this paper how the NBS based expansion can be applied to speed up several common computer vision algorithms, including normalized cross correlation (NCC), sum of squared difference (SSD) matching, appearance subspace projection and subspace-based object recognition. Promising experimental results on facial and natural images are demonstrated in this paper.
ICCV	Modelling Shapes with Uncertainties: Higher Order Polynomials, Variable Bandwidth Kernels and Non Parametric Density Estimation.	Maxime Taron,Nikos Paragios,Marie-Pierre Jolly	2005	In this paper, we introduce a new technique for shape modelling in the space of implicit polynomials. Registration consists of recovering an optimal one-to-one transformation of a higher order polynomial along with uncertainties measures that are determined according to the covariance matrix of the correspondences at the zero isosurface. In the modelling phase, these measures are used to weight the importance of the training samples phase according to a variable bandwidth non-parametric density estimation process. The selection of the most appropriate kernels to represent the training set is done through the maximum likelihood criterion. Excellent results for patterns of digits, related with the registration and the modelling aspects of our approach demonstrate the potentials of our method.
ICCV	Shape Classifer Based on Generalized Probabilistic Descent Method with Hidden Markov Descriptor.	Ninad Thakoor,Jean Gao	2005	The goal of this paper is to present a weighted likelihood discriminant for minimum error shape classification. Different from traditional Maximum Likelihood (ML) methods, in which classification is based on probabilities from independent individual class models as is the case for general Hidden Markov Model (HMM) methods, proposed method utilizes information from all classes to minimize classification error. The proposed approach uses a HMM for shape curvature as its 2-D shape descriptor. In this contribution we introduce a weighted likelihood discriminant function and present a minimum error classification strategy based on Generalized Probabilistic Descent (GPD) method. We believe our sound theory based implementation reduces classification error by combining HMM with GPD theory. We show comparative results obtained with our approach and classic ML classification alongwith Fourier descriptor and Zernike moments based classification for fighter planes and vehicle shapes.
ICCV	Multi-View Geometry of 1D Radial Cameras and its Application to Omnidirectional Camera Calibration.	SriRam Thirthala,Marc Pollefeys	2005	We study the multi-view geometry of 1D radial cameras. A broad a class of both central and non-central cameras, such as fish-eye and catadioptric cameras, can be reduced to 1D radial cameras under the assumption of known center of radial distortion. For cameras in general configuration, we introduce a quadrifocal tensor that can be computed linearly from 15 or more features seen in four views. From this tensor a metric reconstruction of the 1D cameras as well as the observed features can be obtained. In a second phase this reconstruction can then be used as a calibration object to estimate a non-parametric non-central model for the cameras. We study some degenerate cases, including pure rotation. In the case of a purely rotating camera we obtain a trifocal tensor that can be estimated linearly from 7 points in three views. This allows us to obtain a metric reconstruction of the plane at infinity. Next, we use the plane at infinity as a calibration device to non-parametrically estimate the radial distortion. We demonstrate the results of our approach on real and synthetic images.
ICCV	Shape from Symmetry.	Sebastian Thrun,Ben Wegbreit	2005	We describe a technique for reconstructing probable occluded surfaces from 3-D range images. The technique exploits the fact that many objects possess shape symmetries that can be recognized even from partial 3-D views. Our approach identifies probable symmetries and uses them to extend the partial 3-D shape model into the occluded space. To accommodate objects consisting of multiple parts, we describe a technique for segmenting objects into parts characterized by different symmetries. Results are provided for a real-world database of 3-D range images of common objects, acquired through an active stereo rig.
ICCV	Inference of Non-Overlapping Camera Network Topology by Measuring Statistical Dependence.	Kinh Tieu,Gerald Dalley,W. Eric L. Grimson	2005	We present an approach for inferring the topology of a camera network by measuring statistical dependence between observations in different cameras. Two cameras are considered connected if objects seen departing in one camera are seen arriving in the other. This is captured by the degree of statistical dependence between the cameras. The nature of dependence is characterized by the distribution of observation transformations between cameras, such as departure to arrival transition times, and color appearance. We show how to measure statistical dependence when the correspondence between observations in different cameras is unknown. This is accomplished by non-parametric estimates of statistical dependence and Bayesian integration of the unknown correspondence. Our approach generalizes previous work which assumed restricted parametric transition distributions and only implicitly dealt with unknown correspondence. Results are shown on simulated and real data. We also describe a technique for learning the absolute locations of the cameras with Global Positioning System (GPS) side information.
ICCV	Simultaneous Multiple 3D Motion Estimation via Mode Finding on Lie Groups.	Oncel Tuzel,Raghav Subbarao,Peter Meer	2005	We propose a new method to estimate multiple rigid motions from noisy 3D point correspondences in the presence of outliers. The method does not require prior specification of number of motion groups and estimates all the motion parameters simultaneously. We start with generating samples from the rigid motion distribution. The motion parameters are then estimated via mode finding operations on the sampled distribution. Since rigid motions do not lie on a vector space, classical statistical methods can not be used for mode finding. We develop a mean shift algorithm which estimates modes of the sampled distribution using the Lie group structure of the rigid motions. We also show that proposed mean shift algorithm is general and can be applied to any distribution having a matrix Lie group structure. Experimental results on synthetic and real image data demonstrate the superior performance of the algorithm.
ICCV	Priors for People Tracking from Small Training Sets.	Raquel Urtasun,David J. Fleet,Aaron Hertzmann,Pascal Fua	2005	We advocate the use of Scaled Gaussian Process Latent Variable Models (SGPLVM) to learn prior models of 3D human pose for 3D people tracking. The SGPLVM simultaneously optimizes a low-dimensional embedding of the high-dimensional pose data and a density function that both gives higher probability to points close to training data and provides a nonlinear probabilistic mapping from the low-dimensional latent space to the full-dimensional pose space. The SGPLVM is a natural choice when only small amounts of training data are available. We demonstrate our approach with two distinct motions, golfing and walking. We show that the SGPLVM sufficiently constrains the problem such that tracking can be accomplished with straighforward deterministic optimization.
ICCV	KALMANSAC: Robust Filtering by Consensus.	Andrea Vedaldi,Hailin Jin,Paolo Favaro,Stefano Soatto	2005	We propose an algorithm to perform causal inference of the state of a dynamical model when the measurements are corrupted by outliers. While the optimal (maximum-likelihood) solution has doubly exponential complexity due to the combinatorial explosion of possible choices of inliers, we exploit the structure of the problem to design a sampling-based algorithm that has constant complexity. We derive our algorithm from the equations of the optimal filter, which makes our approximation explicit. Our work is motivated by real-time tracking and the estimation of structure from motion (SFM). We test our algorithm for on-line outlier rejection both for tracking and for SFM. We show that our approach can tolerate a large proportion of outliers, whereas previous causal robust statistical inference methods failed with less than half as many. Our work can be thought of as the extension of random sample consensus algorithms to dynamic data, or as the implementation of pseudo-Bayesian filtering algorithms in a sampling framework.
ICCV	Features for Recognition: Viewpoint Invariance for Non-Planar Scenes.	Andrea Vedaldi,Stefano Soatto	2005	Most current local feature detectors/descriptors implicitly assume that the scene is (locally) planar, an assumption that is violated at surface discontinuities. We show that this restriction is, at least in theory, unnecessary, as one can construct local features that are viewpoint-invariant for generic non-planar scenes. However, we show that any such feature necessarily sacrifices shape information, in the sense of being non shape-discriminative. Finally, we show that if viewpoint is factored out as part of the matching process, rather than explicitly in the representation, then shape is discriminative indeed. We illustrate our theoretical results empirically by showing that, even for simple scenes, current affine descriptors fail where even a naive 3-D viewpoint invariant succeeds in matching.
ICCV	Using Frontier Points to Recover Shape, Reflectance and Illumunation.	George Vogiatzis,Paolo Favaro,Roberto Cipolla	2005	We describe a method to recover the surface reflectance and the 3-d shape of a non-Lambertian object as well as illumination, from a collection of images. It is based on the so-called frontier points, which are extracted from the outlines of an object. Frontier points provide 3-d locations on the object surface where the surface normal is known. This information is exploited to infer the surface reflectance of the object and the light distribution of the scene both under varying illumination and fixed vantage point, and under varying vantage point and fixed illumination. We also show how to apply frontier points for shape recovery in photometric stereo. The effectiveness of frontier points for recovering reflectance, illumination and shape is confirmed by a number of experiments on both real and synthetic data.
ICCV	Learning Models for Predicting Recognition Performance.	Rong Wang,Bir Bhanu	2005	This paper addresses one of the fundamental problems encountered in performance prediction for object recognition. In particular we address the problems related to estimation of small gallery size that can give good error estimates and their confidences on large probe sets and populations. We use a generalized two-dimensional prediction model that integrates a hypergeometric probability distribution model with a binomial model explicitly and considers the distortion problem in large populations. We incorporate learning in the prediction process in order to find the optimal small gallery size and to improve its performance. The Chernoff and Chebychev inequalities are used as a guide to obtain the small gallery size. During the prediction we use the expectation-maximum (EM) algorithm to learn the match score and the non-match score distributions (the number of components, their weights, means and covariances) that are represented as Gaussian mixtures. By learning we find the optimal size of small gallery and at the same time provide the upper bound and the lower bound for the prediction on large populations. Results are shown using real-world databases.
ICCV	Perceptual Scale Space and its Applications.	Yizhou Wang,Siavosh Bahrami,Song Chun Zhu	2005	In this paper, we study a perceptual scale space by constructing a so-called sketch pyramid which augments the Gaussian and Laplacian pyramid representations in traditional image scale space theory. Each level of this sketch pyramid is a generic attributed graph ¿ called the primal sketch which is inferred from the corresponding image at the same level of the Gaussian pyramid. When images are viewed at increasing resolutions, more details are revealed. This corresponds to perceptual transitions which are represented by topological changes in the sketch graph in terms of a graph grammar. We compute the sketch or perceptual pyramid by Bayesian inference upwards-downwards the pyramid using Markov Chain Monte Carlo reversible jumps. We show two example applications of this perceptual scale space: (1) motion tracking of objects over scales, and (2) adaptive image displays which can efficiently show a large high-resolution image in a small screen (of a PDA for example) through a selective tour of its image pyramid. Other potential applications include super-resolution and multi-resolution object recognition.
ICCV	An Iterative Optimization Approach for Unified Image Segmentation and Matting.	Jue Wang,Michael F. Cohen	2005	Separating a foreground object from the background in a static image involves determining both full and partial pixel coverages, also known as extracting a matte. Previous approaches require the input image to be pre-segmented into three regions: foreground, background and unknown, which is called a trimap. Partial opacity values are then computed only for pixels inside the unknown region. This pre-segmentation based approach fails for images with large portions of semi-transparent foreground where the trimap is difficult to create even manually. In this paper we combine the segmentation and matting problem together and propose a unified optimization approach based on Belief Propagation. We iteratively estimate the opacity value for every pixel in the image, based on a small sample of foreground and background pixels marked by the user. Experimental results show that compared with previous approaches, our method is more efficient to extract high quality mattes for foregrounds with significant semi-transparent regions.
ICCV	Mutual Information-Based 3D Surface Matching with Applications to Face Recognition and Brain Mapping.	Yalin Wang,Ming-Chang Chiang,Paul M. Thompson	2005	Face recognition and many medical imaging applications require the computation of dense correspondence vector fields that match one surface with another. In brain imaging, surface-based registration is useful for tracking brain change, and for creating statistical shape models of anatomy. Based on surface correspondences, metrics can also be designed to measure differences in facial geometry and expressions. To avoid the need for a large set of manually-defined landmarks to constrain these surface correspondences, we developed an algorithm to automate the matching of surface features. It extends the mutual information method to automatically match general 3D surfaces (including surfaces with a branching topology). We use diffeomorphic flows to optimally align the Riemann surface structures of two surfaces. First, we use holomorphic1-forms to induce consistent conformal grids on both surfaces. High genus surfaces are mapped to a set of rectangles in the Euclidean plane, and closed genus-zero surfaces are mapped to the sphere. Next, we compute stable geometric features (mean curvature and conformal factor) and pull them back as scalar fields onto the 2D parameter domains. Mutual information is used as a cost functional to drive a fluid flow in the parameter domain that optimally aligns these surface features. A diffeomorphic surface-to-surface mapping is then recovered that matches surfaces in 3D. Lastly, we present a spectral method that ensures that the grids induced on the target surface remain conformal when pulled through the correspondence field. Using the chain rule, we express the gradient of the mutual information between surfaces in the conformal basis of the source surface. This finite-dimensional linear space generates all conformal reparameterizations of the surface. Illustrative experiments apply the method to face recognition and to the registration of brain structures, such as the hippocampus in 3D MRI scans, a key step in understanding brain shape alterations in Alzheimer¿s disease and schizophrenia.
ICCV	Retrieval with Knowledge-driven Kernel Design: An Approach to Improving SVM-Based CBIR with Relevance Feedback.	Lei Wang,Yan Gao,Kap Luk Chan,Ping Xue,Wei-Yun Yau	2005	The performance of SVM-based image retrieval is often constrained by the scarcity of training samples. The total number of image samples labelled by users in a retrieval session is very limited, and these small number of labelled samples cannot effectively represent the true distributions of positive and negative image classes, especially for the negative image class. This paper proposes a novel approach to deal with this problem. Instead of treating it as a problem, the mere existence of the small number of labelled images and their desired distribution in the kernel space is considered as prior knowledge from image retrieval to aid the design of the kernel used by SVMs. This is achieved by maximizing a criterion, such as one based on scatter matrices, through gradient-based search methods, incurring very little computational overhead to real-time retrieval process. Experimental results on two benchmark image databases demonstrate the improved retrieval performance by the dynamically designed kernel and hence the effectiveness of the proposed approach for SVM based image retrieval.
ICCV	Surface Parameterization Using Riemann Surface Structure.	Yalin Wang,Xianfeng Gu,Kiralee M. Hayashi,Tony F. Chan,Paul M. Thompson,Shing-Tung Yau	2005	We propose a general method that parameterizes general surfaces with complex (possible branching) topology using Riemann surface structure. Rather than evolve the surface geometry to a plane or sphere, we instead use the fact that all orientable surfaces are Riemann surfaces and admit conformal structures, which induce special curvilinear coordinate systems on the surfaces. We can then automatically partition the surface using a critical graph that connects zero points in the global conformal structure on the surface. The trajectories of iso-parametric curves canonically partition a surface into patches. Each of these patches is either a topological disk or a cylinder and can be conformally mapped to a parallelogram by integrating a holomorphic 1-form defined on the surface. The resulting surface subdivision and the parameterizations of the components are intrinsic and stable. For surfaces with similar topology and geometry, we show that the parameterization results are consistent and the subdivided surfaces can be matched to each other using constrained harmonic maps. The surface similarity can be measured by direct computation of distance between each pair of corresponding points on two surfaces. To illustrate the technique, we computed conformal structures for anatomical surfaces in MRI scans of the brain and human face surfaces. We found that the resulting parameterizations were consistent across subjects, even for branching structures such as the ventricles, which are otherwise difficult to parameterize. Our method provides a surface-based framework for statistical comparison of surfaces and for generating grids on surfaces for PDE-based signal processing.
ICCV	High Resolution Tracking of Non-Rigid 3D Motion of Densely Sampled Data Using Harmonic Maps.	Yang Wang,Mohit Gupta,Song Zhang,Sen Wang,Xianfeng Gu,Dimitris Samaras,Peisen Huang	2005	We present a novel fully automatic method for high resolution, non-rigid dense 3D point tracking. High quality dense point clouds of non-rigid geometry moving at video speeds are acquired using a phase-shifting structured light ranging technique. To use such data for the temporal study of subtle motions such as those seen in facial expressions, an efficient non-rigid 3D motion tracking algorithm is needed to establish inter-frame correspondences. The novelty of this paper is the development of an algorithmic framework for 3D tracking that unifies tracking of intensity and geometric features, using harmonic maps with added feature correspondence constraints. While the previous uses of harmonic maps provided only global alignment, the proposed introduction of interior feature constraints guarantees that non-rigid deformations will be accurately tracked as well. The harmonic map between two topological disks is a diffeomorphism with minimal stretching energy and bounded angle distortion. The map is stable, insensitive to resolution changes and is robust to noise. Due to the strong implicit and explicit smoothness constraints imposed by the algorithm and the high-resolution data, the resultingregistration/deformation field is smooth, continuous and gives dense one-to-one inter-frame correspondences. Our method is validated through a series of experiments demonstrating its accuracy and efficiency.
ICCV	Separating Reflections in Human Iris Images for Illumination Estimation.	Huiqiong Wang,Stephen Lin,Xiaopei Liu,Sing Bing Kang	2005	A method is presented for separating corneal reflections in an image of human irises to estimate illumination from the surrounding scene. Previous techniques for reflection separation have demonstrated success in only limited cases, such as for uniform colored lighting and simple object textures, so they are not applicable to irises which exhibit intricate textures and complicated reflections of the environment. To make this problem feasible, we present a method that capitalizes on physical characteristics of human irises to obtain an illumination estimate that encompasses the prominent light contributors in the scene. Results of this algorithm are presented for eyes of different colors, including light colored eyes for which reflection separation is necessary to determine a valid illumination estimate.
ICCV	Patch Based Blind Image Super Resolution.	Qiang Wang,Xiaoou Tang,Harry Shum	2005	In this paper, a novel method for learning based image super resolution (SR) is presented. The basic idea is to bridge the gap between a set of low resolution (LR) images and the corresponding high resolution (HR) image using both the SR reconstruction constraint and a patch based image synthesis constraint in a general probabilistic framework. We show that in this framework, the estimation of the LR image formation parameters is straightforward. The whole framework is implemented via an annealed Gibbs sampling method. Experiments on SR on both single image and image sequence input show that the proposed method provides an automatic and stable way to compute super-resolution and the achieved result is encouraging for both synthetic and real LR images.
ICCV	Space-Time Scene Manifolds.	Yonatan Wexler,Denis Simakov	2005	The space of images is known to be a non-linear sub-space that is difficult to model. This paper derives an algorithm that walks within this space. We seek a manifold through the video volume that is constrained to lie locally in this space. Every local neighborhood within the manifold resembles some image patch. We call this the Scene Manifold because the solution traces the scene outline. For a broad class of inputs the problem can be posed as finding the shortest path in a graph and can thus be solved efficiently to produce the globally optimal solution. Constraining appearance rather than geometry gives rise to numerous new capabilities. Here we demonstrate the usefulness of this approach by posing the well-studied problem of mosaicing in a new way. Instead of treating it as geometrical alignment, we pose it as an appearance optimization. Since the manifold is constrained to lie in the space of valid image patches, the resulting mosaic is guaranteed to have the least distortions possible. Any small part of it can be seen in some image even though the manifold spans the whole video. Thus it can deal seamlessly with both static and dynamic scenes, with or without 3D parallax. Essentially, the method simultaneously solves two problems that have been solved only separately until now: alignment and mosaicing.
ICCV	Object Categorization by Learned Universal Visual Dictionary.	John M. Winn,Antonio Criminisi,Thomas P. Minka	2005	This paper presents a new algorithm for the automatic recognition of object classes from images (categorization). Compact and yet discriminative appearance-based object class models are automatically learned from a set of training images. The method is simple and extremely fast, making it suitable for many applications such as semantic image retrieval, web search, and interactive image editing. Itclassifies a region according to the proportions of different visual words (clusters in feature space). The specific visual words and the typical proportions in each object are learned from a segmented training set. The main contribution of this paper is two fold: i) an optimally compact visual dictionary is learned by pair-wise merging of visual words from an initially large dictionary. The final visual words are described by GMMs. ii) A novel statistical measure of discrimination is proposed which is optimized by each merge operation. High classification accuracy is demonstrated for nine object classes on photographs of real objects viewed under general lighting conditions, poses and viewpoints. The set of test images used for validation comprise: i) photographs acquired by us, ii) images from the web and iii) images from the recently released Pascal dataset. The proposed algorithm performs well on both texture-rich objects (e.g. grass, sky, trees) and structure-rich ones (e.g. cars, bikes, planes).
ICCV	LOCUS: Learning Object Classes with Unsupervised Segmentation.	John M. Winn,Nebojsa Jojic	2005	We address the problem of learning object class models and object segmentations from unannotated images. We introduce LOCUS (Learning Object Classes with Unsupervised Segmentation) which uses a generative probabilistic model to combine bottom-up cues of color and edge with top-down cues of shape and pose. A key aspect of this model is that the object appearance is allowed to vary from image to image, allowing for significant within-class variation. By iteratively updating the belief in the object¿s position, size, segmentation and pose, LOCUS avoids making hard decisions about any of these quantities and so allows for each to be refined at any stage. We show that LOCUS successfully learns an object class model from unlabeled images, whilst also giving segmentation accuracies that rival existing supervised methods. Finally, we demonstrate simultaneous recognition and segmentation in novel images using the learned models for a number of object classes, as well as unsupervised object discovery and tracking in video.
ICCV	8-Point Algorithm Revisited: Factorized 8-Point Algorithm.	F. C. Wu,Z. Y. Hu,F. Q. Duan	2005	In this paper, a novel algorithm for the fundamental matrix estimation, called factorized 8-point algorithm, is presented. The factorized 8-point algorithm is composed of three steps: (1) The measurement matrix in the traditional 8-point algorithm is decomposed into two factor matrices; (2) By introducing some auxiliary variables, a new linear minimization problem is formed, where every element of its associated measurement matrix is simply either a measurement datum or a constant; (3) The fundamental matrix is determined by solving this minimization problem by a least squares method. Like the traditional 8-point algorithm and Hartley¿s normalized 8-point algorithm, the factorized8-point algorithm is also completely linear. But unlike the normalized 8-point algorithm, the factorized 8-point algorithm does not need any pre-normalization step. Since every element of the measurement matrix in the factorized8-point algorithm is a measurement datum or a constant, no amplification of measurement error is involved; the factorized 8-point algorithm can boost effectively the robustness of the estimation. Large numbers of experiments show that the factorized 8-point algorithm consistently outperforms the traditional 8-point algorithm. In addition, although the factorized 8-point algorithm is specially designed for fundamental matrix estimation, its basic principle can be generalized to other estimation problems in computer vision, such as camera projection matrix estimation, homography estimation, focus of expansion estimation, and trifocal tensor estimation.
ICCV	Visual Learning Given Sparse Data of Unknown Complexity.	Tao Xiang,Shaogang Gong	2005	This study addresses the problem of unsupervised visual learning. It examines existing popular model order selection criteria before proposes two novel criteria for improving visual learning given sparse data and without any knowledge about model complexity. In particular, a rectified Bayesian Information Criterion (BICr) and a Completed Likelihood Akaike¿s Information Criterion (CL-AIC) are formulated to estimate the optimal model order (complexity) for learning the dynamic structure of a visual scene. Both criteria are designed to overcome poor model selection by existing popular criteria when the data sample size varies from very small to large. Extensive experiments on learning a dynamic scene structure are carried out to demonstrate the effectiveness of BICr and CL-AIC, compared to that of BIC [15], AIC [1], ICL [3] and a MML based criterion [7].
ICCV	Video Behaviour Profiling and Abnormality Detection without Manual Labelling.	Tao Xiang,Shaogang Gong	2005	A novel framework is developed for automatic behaviour profiling and abnormality sampling/detection without any manual labelling of the training dataset. Natural grouping of behaviour patterns is discovered through unsupervised model selection and feature selection on the eigen-vectors of a normalised affinity matrix. Our experiments demonstrate that a behaviour model trained using an unlabelled dataset is superior to those trained using the same but labelled dataset in detecting abnormality from an unseen video.
ICCV	Uncalibrated Perspective Reconstruction of Deformable Structures.	Jing Xiao,Takeo Kanade	2005	Reconstruction of 3D structures from uncalibrated image sequences has a wealthy history. Most work has been focused on rigid objects or static scenes. This paper studies the problem of perspective reconstruction of deformable structures such as dynamic scenes from an uncalibrated image sequence. The task requires decomposing the image measurements into a composition of three factors: 3D deformable structures, rigid rotations and translations, and intrinsic camera parameters. We develop a factorization algorithm that consists of two steps. In the first step we recover the projective depths iteratively using the sub-space constraints embedded in the image measurements of the deformable structures. In the second step, we scale the image measurements by the reconstructed projective depths. We then extend the linear closed-form solution for weakperspective reconstruction [23] to factorize the scaled measurements and simultaneously reconstruct the deformable shapes and underlying shape model, the rigid motions, and the varying camera parameters such as focal lengths. The accuracy and robustness of the proposed method is demonstrated quantitatively on synthetic data and qualitatively on real image sequences.
ICCV	Automatic 3D Face Modeling from Video.	Le Xin,Qiang Wang,Jianhua Tao,Xiaoou Tang,Tieniu Tan,Harry Shum	2005	In this paper, we develop an efficient technique for fully automatic recovery of accurate 3D face shape from videos captured by a low cost camera. The method is designed to work with a short video containing a face rotating from frontal view to profile view. The whole approach consists of three components. First, automatic initialization is performed in the first frame with approximately frontal face. Then, to handle the case of low quality image captured by low cost camera, the 2D feature matching, head poses and underlying 3D face shape are estimated and refined iteratively in an efficient way based on image sequence segmentation. Finally, to take advantage of the sparse structure of the proposed algorithm, sparse bundle adjustment technique is further employed to speed up the computation. We demonstrate the accuracy and robustness of the algorithm using a set of experiments.
ICCV	Okapi-Chamfer Matching for Articulated Object Recognition.	Hanning Zhou,Thomas S. Huang	2005	Recent years have witnessed the rise of many effective text information retrieval systems. By treating local visual features as terms, training images as documents and input images as queries, we formulate the problem of object recognition into that of text retrieval. Our formulation opens up the opportunity to integrate some powerful text retrieval tools with computer vision techniques. In this paper, we propose to improve the efficiency of articulated object recognition by an Okapi-Chamfer matching algorithm. The algorithm is based on the inverted index technique. The inverted index is a widely used way to effectively organize a collection of text documents. With the inverted index, only documents that contain query terms are accessed and used for matching. To enable inverted indexing in an image database, we build a lexicon of local visual features by clustering the features extracted from the training images. Given a query image, we extract visual features and quantize them based on the lexicon, and then look up the inverted index to identify the subset of training images with non-zero matching score. To evaluate the matching scores in the subset, we combined the modified Okapi weighting formula with the Chamfer distance. The performance of the Okapi-Chamfer matching algorithm is evaluated on a hand posture recognition system. We test the system with both synthesized and real world images. Quantitative results demonstrate the accuracy and efficiency of our system.
ICCV	Detection, Analysis and Matching of Hair.	Yaser Yacoob,Larry S. Davis	2005	We develop computational models for measuring hair appearance for comparing different people. The models and methods developed have applications to person recognition and face image indexing. An automatic hair detection algorithm is described and results reported. A multidimensional representation of hair appearance is presented and computational algorithms are described. Results on a dataset of 524 subjects are reported. Identification of people using hair attributes is compared to Eigenface-based recognition along with a joint, Eigenface-hair based identification.
ICCV	Learning a Sparse, Corner-Based Representation for Time-varying Background Modeling.	Qiang Zhu,Shai Avidan,Kwang-Ting Cheng	2005	"Time-varying phenomenon, such as ripples on water, trees waving in the wind and illumination changes, produces false motions, which significantly compromises the performance of an outdoor-surveillance system. In this paper, we propose a corner-based background model to effectively detect moving-objects in challenging dynamic scenes. Specifically, the method follows a three-step process. First, we detect feature points using a Harris corner detector and represent them as SIFT-like descriptors. Second, we dynamically learn a background model and classify each extracted feature as either a background or a foreground feature. Last, a ""Lucas-Kanade"" feature tracker is integrated into this framework to differentiate motion-consistent foreground objects from background objects with random or repetitive motion. The key insight of our work is that a collection of SIFT-like features can effectively represent the environment and account for variations caused by natural effects with dynamic movements. Features that do not correspond to the background must therefore correspond to foreground moving objects. Our method is computational efficient and works in real-time. Experiments on challenging video clips demonstrate that the proposed method achieves a higher accuracy in detecting the foreground objects than the existing methods."
ICCV	Fast Multiple Object Tracking via a Hierarchical Particle Filter.	Changjiang Yang,Ramani Duraiswami,Larry S. Davis	2005	A very efficient and robust visual object tracking algorithm based on the particle filter is presented. The method characterizes the tracked objects using color and edge orientation histogram features. While the use of more features and samples can improve the robustness, the computational load required by the particle filter increases. To accelerate the algorithm while retaining robustness we adopt several enhancements in the algorithm. The first is the use of integral images [34] for efficiently computing the color features and edge orientation histograms, which allows a large amount of particles and a better description of the targets. Next, the observation likelihood based on multiple features is computed in a coarse-to-fine manner, which allows the computation to quickly focus on the more promising regions. Quasi-random sampling of the particles allows the filter to achieve a higher convergence rate. The resulting tracking algorithm maintains multiple hypotheses and offers robustness against clutter or short period occlusions. Experimental results demonstrate the efficiency and effectiveness of the algorithm for single and multiple object tracking.
ICCV	Hilbert Functions and Applications to the Estimation of Subspace Arrangements.	Allen Y. Yang,Robert M. Fossum	2005	This paper develops a new mathematical framework for studying the subspace-segmentation problem. We examine some important algebraic properties of subspace arrangements that are closely related to the subspace-segmentation problem. More specifically, we introduce an important class of invariants given by the Hilbert functions. We show that there exist rich relations between subspace arrangements and their corresponding Hilbert functions. We propose a new subspace-segmentation algorithm, and showcase two applications to demonstrate how the new theoretical revelation may solve subspace segmentation and model selection problems under less restrictive conditions with improved results.
ICCV	Consistent Segmentation for Optical Flow Estimation.	C. Lawrence Zitnick,Nebojsa Jojic,Sing Bing Kang	2005	In this paper, we propose a method for jointly computing optical flow and segmentating video while accounting for mixed pixels (matting). Our method is based on statistical modeling of an image pair using constraints on appearance and motion. Segments are viewed as overlapping regions with fractional (á) contributions. Bidirectional motion is estimated based on spatial coherence and similarity of segment colors. Our model is extended to video by chaining the pairwise models to produce a joint probability distribution to be maximized. To make the problem more tractable, we factorize the posterior distribution and iteratively minimize its parts. We demonstrate our method on frame interpolation.
ICCV	Learning the Probability of Correspondences without Ground Truth.	Qingxiong Yang,R. Matt Steele,David Nistér,Christopher O. Jaynes	2005	We present a quality assessment procedure for correspondence estimation based on geometric coherence rather than ground truth. The procedure can be used for performance evaluation of correspondence extraction schemes developed by researchers, as well as for online learning and adaptation aimed at better system performance. A very important aspect of the proposed procedure is that it considers uncertainty in the correspondence extraction, and encourages the evaluated methods to deal correctly with uncertainty. Other important strengths of the procedure are that it does not use any manual work, and that it does not put any strong constraints on the scene, but rather relies on geometric coherence in the motion. Thanks to these strengths, it can therefore be used with large amounts of real, potentially application specific data, or even data acquired during system operation. In the evaluation the correspondence extractor is handled as a black box producing a probability distribution for the local motion vector between a pair of image patches. The procedure is therefore quite general. We are making the evaluation procedure available for public use.
ICCV	Is ICA Significantly Better than PCA for Face Recognition?.	Jian Yang,David Zhang,Jing-Yu Yang	2005	The standard PCA was always used as baseline algorithm to evaluate ICA-based face recognition systems in the previous research. In this paper, we examine the two architectures of ICA for image representation and find that ICA Architecture I involves a PCA process by vertically centering (PCA I), while ICA Architecture II involves a whitened PCA process by horizontally centering (PCA II). So, it is reasonable to use these two PCA versions as baseline algorithms to revaluate the ICA-based face recognition systems. The experiments were performed on the FERET face database. The experimental results show there is no significant performance differences between ICA Architecture I (II) and PCA I (II), although ICA Architecture II significantly outperforms the standard PCA. It can be concluded that the performance of ICA strongly depends on its involved PCA process. The pure ICA projection has little effect on the performance of face recognition.
ICCV	Object Detection in Aerial Imagery Based on Enhanced Semi-Supervised Learning.	Jian Yao,Zhongfei (Mark) Zhang	2005	Object detection in aerial imagery has been well studied in computer vision for years. However, given the complexity of large variations of the appearance of the object and the background in a typical aerial image, a robust and efficient detection is still considered as an open and challenging problem. In this paper, we present the Enhanced Semi-Supervised Learning (ESL) framework and apply this framework to revising an object detection methodology we have developed in a previous effort. Theoretic analysis and experimental evaluation using the UCI machine learning repository clearly indicate the superiority of the ESL framework. The performance evaluations of the revised object detection methodology against the original one clearly demonstrate the promise and superiority of this approach.
ICCV	"Conformal Metrics and True ""Gradient Flows"" for Curves."	Anthony J. Yezzi,Andrea Mennucci	2005	"We wish to endow the manifold M of smooth curves in ¿¿ with a Riemannian metric that allows us to treat continuous morphs (homotopies) between two curves c¿ and c¿ as trajectories with computable lengths which are independent of the parameterization or representation of the two curves (and the curves making up the morph between them). We may then define the distance between the two curves using the trajectory of minimal length (geodesic) between them, assuming such a minimizing trajectory exists. At first we attempt to utilize the metric structure implied rather unanimously by the past twenty years or so of shape optimization literature in computer vision. This metric arises as the unique metric which validates the common references to a wide variety of contour evolution models in the literature as ""gradient flows"" to various formulated energy functionals. Surprisingly, this implied metric yields a pathological and useless notion of distance between curves. In this paper, we show how this metric can be minimally modified using conformal factors the depend upon a curve¿s total arclength. A nice property of these new conformal metrics is that all active contour models that have been called ""gradient flows"" in the past will constitute true gradient flows with respect to these new metrics under specfic time reparameterizations."
ICCV	Recognizing Human Actions in Videos Acquired by Uncalibrated Moving Cameras.	Alper Yilmaz,Mubarak Shah	2005	Most work in action recognition deals with sequences acquired by stationary cameras with fixed viewpoints. Due to the camera motion, the trajectories of the body parts contain not only the motion of the performing actor but also the motion of the camera. In addition to the camera motion, different viewpoints of the same action in different environments result in different trajectories, which can not be matched using standard approaches. In order to handle these problems, we propose to use the multi-view geometry between two actions. However, well known epipolar geometry of the static scenes where the cameras are stationary is not suitable for our task. Thus, we propose to extend the standard epipolar geometry to the geometry of dynamic scenes where the cameras are moving. We demonstrate the versatility of the proposed geometric approach for recognition of actions in a number of challenging sequences.
ICCV	Linear Approaches to Camera Calibration from Sphere Images or Active Intrinsic Calibration Using Vanishing Points.	Xianghua Ying,Hongbin Zha	2005	Spherical objects and vanish points are often used for camera calibration. An occluding contour of a sphere is projected to a conic in the perspective image, and using a moving active camera, the trajectory of a vanishing point in the perspective images is also a conic when the camera is rotated about a fixed 3D axis whereas the translation of the camera is arbitrary. In fact, the problems of camera calibration using conics from spheres or vanishing points can be described by same mathematic representations. Two linear approaches to the problems are proposed in this paper: one based on the geometric interpretation of the relation between image conics and the image of the absolute conic, and the other using the special structure of the problems in algebra. Only three such conics are needed for the two linear approaches, and the minimum number for previous nonlinear optimization methods is also three. All five intrinsic parameters are recovered linearly without making assumptions, such as, zero-skew or unitary aspect ratio which are often used in previous methods. The two linear algorithms have been tested in extensive experiments with respect to noise sensitivity and also made comparisons with recent calibration techniques.
ICCV	A Unifying Approach to Hard and Probabilistic Clustering.	Ron Zass,Amnon Shashua	2005	"We derive the clustering problem from first principles showing that the goal of achieving a probabilistic, or ""hard"", multi class clustering result is equivalent to the algebraic problem of a completely positive factorization under a doubly stochastic constraint. We show that spectral clustering, normalized cuts, kernel K-means and the various normalizations of the associated affinity matrix are particular instances and approximations of this general principle. We propose an efficient algorithm for achieving a completely positive factorization and extend the basic clustering scheme to situations where partial label information is available."
ICCV	Image Based Regression Using Boosting Method.	Shaohua Kevin Zhou,Bogdan Georgescu,Xiang Sean Zhou,Dorin Comaniciu	2005	We present a general algorithm of image based regression that is applicable to many vision problems. The proposed regressor that targets a multiple-output setting is learned using boosting method. We formulate a multiple-output regression problem in such a way that overfitting is decreased and an analytic solution is admitted. Because we represent the image via a set of highly redundant Haar-like features that can be evaluated very quickly and select relevant features through boosting to absorb the knowledge of the training data, during testing we require no storage of the training data and evaluate the regression function almost in no time. We also propose an efficient training algorithm that breaks the computational bottleneck in the greedy feature selection process. We validate the efficiency of the proposed regressor using three challenging tasks of age estimation, tumor detection, and endocardial wall localization and achieve the best performance with a dramatic speed, e.g., more than 1000 times faster than conventional data-driven techniques such as support vector regressor in the experiment of endo-cardial wall localization.
ICCV	Squaring the Circles in Panoramas.	Lihi Zelnik-Manor,Gabriele Peters,Pietro Perona	2005	Pictures taken by a rotating camera cover the viewing sphere surrounding the center of rotation. Having a set of images registered and blended on the sphere what is left to be done, in order to obtain a flat panorama, is projecting the spherical image onto a picture plane. This step is unfortunately not obvious ¿ the surface of the sphere may not be flattened onto a page without some form of distortion. The objective of this paper is discussing the difficulties and opportunities that are connected to the projection from viewing sphere to image plane. We first explore a number of alternatives to the commonly used linear perspective projection. These are ¿global¿ projections and do not depend on image content. We then show that multiple projections may coexist successfully in the same mosaic: these projections are chosen locally and depend on what is present in the pictures. We show that such multi-view projections can produce more compelling results than the global projections.
ICCV	Progressive Surface Reconstruction from Images Using a Local Prior.	Gang Zeng,Sylvain Paris,Long Quan,François X. Sillion	2005	This paper introduces a new method for surface reconstruction from multiple calibrated images. The primary contribution of this work is the notion of local prior to combine the flexibility of the carving approach with the accuracy of graph-cut optimization. A progressive refinement scheme is used to recover the topology and reason the visibility of the object. Within each voxel, a detailed surface patch is optimally reconstructed using a graph-cut method. The advantage of this technique is its ability to handle complex shape similarly to level sets while enjoying a higher precision. Compared to carving techniques, the addressed problem is well-posed, and the produced surface does not suffer from aliasing. In addition, our approach seamlessly handles complete and partial reconstructions: If the scene is only partially visible, the process naturally produces an open surface; otherwise, if the scene is fully visible, it creates a complete shape. These properties are demonstrated on real image sequences.
ICCV	A General Framework for Temporal Video Scene Segmentation.	Yun Zhai,Mubarak Shah	2005	Videos are composed of many shots caused by different camera operations, e.g., on/off operations and switching between cameras. One important goal in video analysis is to group the shots into temporal scenes, such that all the shots in a single scene are related to a particular physical setting, an on-going action or a theme. In this paper, we present a general framework for temporal scene segmentation for various video types. The proposed method is formulated in a statistical fashion and uses the Markov chain Monte Carlo (MCMC) technique to determine the boundaries between video scenes. In this approach, an arbitrary number of scene boundaries are randomly initialized and automatically updated using two types of updates: diffuse and jumps. The posterior probability on the number of scenes and their boundary locations is computed based on the model priors and the data likelihood. The updates of the model parameters are controlled by the hypothesis ratio test in the MCMC process. The proposed framework has been experimented on two types of videos, home videos and feature films, and accurate results have been obtained.
ICCV	Bayesian Body Localization Using Mixture of Nonlinear Shape Models.	Jiayong Zhang,Robert T. Collins,Yanxi Liu	2005	"We present a 2Dmodel-based approach to localizing human body in images viewed from arbitrary and unknown angles. The central component is a statistical shape representation of the nonrigid and articulated body contours, where a non-linear deformation is decomposed based on the concept of parts. Several image cues are combined to relate the body configuration to the observed image, with self-occlusion explicitly treated. To accommodate large viewpoint changes, a mixture of view-dependent models is employed. Inference is done by direct sampling of the posterior mixture, using Sequential Monte Carlo (SMC) simulation enhanced with annealing and kernel move. The fitting method is independent of the number of mixture components, and does not require the preselection of a ""correct"" viewpoint. The models were trained on a large number of interactively labeled gait images. Preliminary tests demonstrated the feasibility of the proposed approach."
ICCV	Local Gabor Binary Pattern Histogram Sequence (LGBPHS): A Novel Non-Statistical Model for Face Representation and Recognition.	Wenchao Zhang,Shiguang Shan,Wen Gao,Xilin Chen,Hongming Zhang	2005	"For years, researchers in face recognition area have been representing and recognizing faces based on subspace discriminant analysis or statistical learning. Nevertheless, these approaches are always suffering from the generalizability problem. This paper proposes a novel non-statistics based face representation approach, Local Gabor Binary Pattern Histogram Sequence (LGBPHS), in which training procedure is unnecessary to construct the face model, so that the generalizability problem is naturally avoided. In this approach, a face image is modeled as a ""histogram sequence"" by concatenating the histograms of all the local regions of all the local Gabor magnitude binary pattern maps. For recognition, histogram intersection is used to measure the similarity of different LGBPHSes and the nearest neighborhood is exploited for final classification. Additionally, we have further proposed to assign different weights for each histogram piece when measuring two LGBPHSes. Our experimental results on AR and FERET face database show the validity of the proposed approach especially for partially occluded face images, and more impressively, we have achieved the best result on FERET face database."
ICCV	A Probabilistic Semantic Model for Image Annotation and Multi-Modal Image Retrieva.	Ruofei Zhang,Zhongfei (Mark) Zhang,Mingjing Li,Wei-Ying Ma,HongJiang Zhang	2005	This paper addresses automatic image annotation problem and its application to multi-modal image retrieval. The contribution of our work is three-fold. (1) We propose a probabilistic semantic model in which the visual features and the textual words are connected via a hidden layer which constitutes the semantic concepts to be discovered to explicitly exploit the synergy among the modalities. (2) The association of visual features and textual words is determined in a Bayesian framework such that the confidence of the association can be provided. (3) Extensive evaluation on a large-scale, visually and semantically diverse image collection crawled from Web is reported to evaluate the prototype system based on the model. In the proposed probabilistic model, a hidden concept layer which connects the visual feature and the word layer is discovered by fitting a generative model to the training image and annotation words through an Expectation-Maximization (EM) based iterative learning procedure. The evaluation of the prototype system on 17,000 images and 7,736 automatically extracted annotation words from crawled Web pages for multi-modal image retrieval has indicated that the proposed semantic model and the developed Bayesian framework are superior to a state-of-the-art peer system in the literature.
ICCV	Closely Coupled Object Detection and Segmentation.	Liang Zhao,Larry S. Davis	2005	We propose a closely coupled object detection and segmentation algorithm for enhancing both processes in a cooperative and iterative manner. Figure-ground segmentation reduces the effect of background clutter on template matching; the matched template provides shape constraints on segmentation. More precisely, we estimate the probability of each pixel belonging to the foreground by a weighted sum of the estimates based on shape and color alone. The weight on the shape-based estimate is related to the probability that a familiar object is present and is updated dynamically so that we enforce shape constraints only where the object is present. Experiments on detecting people in images of cluttered scenes demonstrate that the proposed algorithm improves both segmentation and detection. More accurate object boundaries are extracted; higher object detection rates and lower false alarm rates are achieved than performing the two processes separately or sequentially.
ICCV	Robust Point Matching for Two-Dimensional Nonrigid Shapes.	Yefeng Zheng,David S. Doermann	2005	Recently, nonrigid shape matching has received more and more attention. For nonrigid shapes, most neighboring points cannot mow independently under deformation due to physical constraints. Furthermore, the rough structure of a shape should be preserved under deformation, otherwise even people cannot match shapes reliably. Therefore, though the absolute distance between two points may change sign@cantl>;, the neighborhood of a point is well preserved in general. Based on this observation, we formulate point matching as a graph matching problem. Each point is a node in the graph, and two nodes are connected by an edge if their Euclidean distance is less than a threshold. The optimal match between two graphs is the one that maximizes the number of matched edges. The shape context distance is used to initialize the graph matching, followed by relaxation labeling for rejnement. Nonrigid deformation is overcome by bringing one shape closer 10 the other in each iteration using deformation parameters estimated from the current point correspondence. Experiments demonstrate the effectiveness of our approach: it outperforms the shape context and TPS-RPM algorithms under nonrigid deformation and noise on a public data set.
VAST	Viewpoint quality and scene understanding.	Dmitry Sokolov,Dimitri Plemenos	2005	Viewpoint quality and scene understanding.
VAST	Rapid Visualization of Large Point-Based Surfaces.	Tamy Boubekeur,Florent Duguet,Christophe Schlick	2005	Rapid Visualization of Large Point-Based Surfaces.
VAST	VAST 2005: The 6th International Symposium on Virtual Reality, Archaeology and Intelligent Cultural Heritage, Pisa, Italy, 2005	Mark Mudge,Nick Ryan,Roberto Scopigno	2005	VAST 2005: The 6th International Symposium on Virtual Reality, Archaeology and Intelligent Cultural Heritage, Pisa, Italy, 2005
VAST	Participating Media for High-Fidelity Cultural Heritage.	Veronica Sundstedt,Diego Gutierrez,Fermin Gomez,Alan Chalmers	2005	Participating Media for High-Fidelity Cultural Heritage.
VAST	3D Face Modeling from Ancient Kabuki Drawings.	Weiwei Xu,Ryo Akama,Hiromi T. Tanaka	2005	3D Face Modeling from Ancient Kabuki Drawings.
VAST	3D Modeling for Non-Expert Users with the Castle Construction Kit v0.5.	Björn Gerth,René Berndt,Sven Havemann,Dieter W. Fellner	2005	3D Modeling for Non-Expert Users with the Castle Construction Kit v0.5.
VAST	Visualizing Temporal Uncertainty in 3D Virtual Reconstructions.	Torre Zuk,M. Sheelagh T. Carpendale,W. D. Glanzman	2005	Visualizing Temporal Uncertainty in 3D Virtual Reconstructions.
VAST	Interface Evaluation for Cultural Heritage Applications: the case of FERRUM exhibition.	Aurkene Alzua-Sorzabal,María Teresa Linaza,Marina Abad,L. Arretxea,Ana Susperregui	2005	Interface Evaluation for Cultural Heritage Applications: the case of FERRUM exhibition.
VAST	Exploring Digitized Artworks by Pointing Posture Recognition.	C. Malerczyk,Patrick Dähne,M. Schnaider	2005	Exploring Digitized Artworks by Pointing Posture Recognition.
VAST	Digital Restoration of Medieval Tapestries.	Sonja Schär,Hanspeter Bieri,Xiaoyi Jiang	2005	Digital Restoration of Medieval Tapestries.
VAST	"Visitors' Evaluations of ICTs Used in Cultural Heritage."	Ruth Owen,Dimitrios Buhalis,Daniel Pletinckx	2005	"Visitors' Evaluations of ICTs Used in Cultural Heritage."
VAST	Passive reconstruction of high quality textured 3D models of works of art.	Nicola Brusco,Luca Ballan,Guido M. Cortelazzo	2005	Passive reconstruction of high quality textured 3D models of works of art.
VAST	Investigating the Structural Validity of Virtual Reconstructions of Prehistoric Maltese Temples.	Alan Chalmers,Kurt Debattista	2005	Investigating the Structural Validity of Virtual Reconstructions of Prehistoric Maltese Temples.
VAST	Rapid Synchronous Acquisition of Geometry and Appearance of Cultural Heritage Artefacts.	Gero Müller,Gerhard H. Bendels,Reinhard Klein	2005	Rapid Synchronous Acquisition of Geometry and Appearance of Cultural Heritage Artefacts.
VAST	Detail-Preserving Surface Inpainting.	Gerhard H. Bendels,Ruwen Schnabel,Reinhard Klein	2005	Detail-Preserving Surface Inpainting.
VAST	Reflection Transformation Imaging and Virtual Representations of Coins from the Hospice of the Grand St. Bernard.	Mark Mudge,Jean-Pierre Voutaz,Carla Schroer,Marlin Lum	2005	Reflection Transformation Imaging and Virtual Representations of Coins from the Hospice of the Grand St. Bernard.
VAST	Real-time Shader Rendering for Crowds in Virtual Heritage.	Pablo de Heras Ciechomski,Sébastien Schertenleib,Jonathan Maïm,Damien Maupu,Daniel Thalmann	2005	Real-time Shader Rendering for Crowds in Virtual Heritage.
VAST	A Texture Based Approach to Reconstruction of Archaeological Finds.	Mahmut Samil Sagiroglu,Aytül Erçil	2005	A Texture Based Approach to Reconstruction of Archaeological Finds.
IEEE Visualization	Using Visual Design Expertise to Characterize the Effectiveness of 2D Scientific Visualization Methods.	Daniel Acevedo Feliz,David H. Laidlaw,Fritz Drury	2005	Using Visual Design Expertise to Characterize the Effectiveness of 2D Scientific Visualization Methods.
IEEE Visualization	Build-by-Number: Rearranging the Real World to Visualize Novel Architectural Spaces.	Daniel R. Bekins,Daniel G. Aliaga	2005	Build-by-Number: Rearranging the Real World to Visualize Novel Architectural Spaces.
IEEE Visualization	Visualizing Large Scale Laser-Plasma Interaction 3D Simulations Using Parallel VTK and Extensions on Clusters.	Daniel Aguilera,Thierry Carrard,Guillaume Colin de Verdiere,Jean-Philippe Nomine	2005	Visualizing Large Scale Laser-Plasma Interaction 3D Simulations Using Parallel VTK and Extensions on Clusters.
IEEE Visualization	Fast and Reproducible Fiber Bundle Selection in DTI Visualization.	Jorik Blaas,Charl P. Botha,Bart Peters,Frans Vos,Frits H. Post	2005	Fast and Reproducible Fiber Bundle Selection in DTI Visualization.
IEEE Visualization	View Selection for Volume Rendering.	Udeepta Bordoloi,Han-Wei Shen	2005	View Selection for Volume Rendering.
IEEE Visualization	A Shader-Based Parallel Rendering Framework.	Jérémie Allard,Bruno Raffin	2005	A Shader-Based Parallel Rendering Framework.
IEEE Visualization	Texture-Based Visualization of Uncertainty in Flow Fields.	Ralf P. Botchen,Daniel Weiskopf	2005	Texture-Based Visualization of Uncertainty in Flow Fields.
IEEE Visualization	Marching Diamonds for Unstructured Meshes.	John C. Anderson,Janine Bennett,Kenneth I. Joy	2005	Marching Diamonds for Unstructured Meshes.
IEEE Visualization	Dynamic Volume Rendering for Intensity Modulated Radiation Therapy (IMRT) Treatment.	Raj Arangarasan,Sungeun Kim,Seza Orcun	2005	Dynamic Volume Rendering for Intensity Modulated Radiation Therapy (IMRT) Treatment.
IEEE Visualization	Visual Exploration of Multidimensional Feature Space of Biological Data.	Tomasz Arodz,Krzysztof Boryczko,Witold Dzwinel,Marcin Kurdziel,David A. Yuen	2005	Visual Exploration of Multidimensional Feature Space of Biological Data.
IEEE Visualization	Level Set and PDE Methods for Visualization.	David E. Breen,Mike Kirby,Aaron E. Lefohn,Ken Museth,Tobias Preusser,Guillermo Sapiro,Ross T. Whitaker	2005	Level Set and PDE Methods for Visualization.
IEEE Visualization	VolumeShop: An Interactive System for Direct Volume Illustration.	Stefan Bruckner,M. Eduard Gröller	2005	VolumeShop: An Interactive System for Direct Volume Illustration.
IEEE Visualization	View-Dependent Rendering of Multiresolution Texture-Atlases.	Henrik Buchholz,Jürgen Döllner	2005	View-Dependent Rendering of Multiresolution Texture-Atlases.
IEEE Visualization	VisTrails: Enabling Interactive Multiple-View Visualizations.	Louis Bavoil,Steven P. Callahan,Carlos Eduardo Scheidegger,Huy T. Vo,Patricia Crossno,Cláudio T. Silva,Juliana Freire	2005	VisTrails: Enabling Interactive Multiple-View Visualizations.
IEEE Visualization	Noise-Adjusted Principle Component Analysis For Hyperspectral Remotely Sensed Imagery Visualization.	Shangshu Cai,Qian Du,Robert Moorhead,Mahnas Jean Mohammadi-Aragh,Derek Irby	2005	Noise-Adjusted Principle Component Analysis For Hyperspectral Remotely Sensed Imagery Visualization.
IEEE Visualization	Interactive Rendering of Large Unstructured Grids Using Dynamic Level-of-Detail.	Steven P. Callahan,João Luiz Dihl Comba,Peter Shirley,Cláudio T. Silva	2005	Interactive Rendering of Large Unstructured Grids Using Dynamic Level-of-Detail.
IEEE Visualization	Visualizing the tightening of knots.	Jason Cantarella,Michael Piatek,Eric Rawdon	2005	Visualizing the tightening of knots.
IEEE Visualization	Topology-driven Surface Mappings with Robust Feature Alignment.	Christopher Carner,Miao Jin,Xianfeng Gu,Hong Qin	2005	Topology-driven Surface Mappings with Robust Feature Alignment.
IEEE Visualization	Illustrative Rendering Techniques for Visualization: Future of Visualization or Just Another Technique?.	Dirk Bartz,Hans Hagen,Victoria Interrante,Kwan-Liu Ma,Bernhard Preim	2005	Illustrative Rendering Techniques for Visualization: Future of Visualization or Just Another Technique?.
IEEE Visualization	Visual Medicine: Part One - Medical Imaging.	Dirk Bartz,Gordon L. Kindlmann,Klaus Mueller,Bernhard Preim,Markus Wacker	2005	Visual Medicine: Part One - Medical Imaging.
IEEE Visualization	Visual Medicine: Part Two - Advanced Applications of Medical Imaging.	Dirk Bartz,Gordon L. Kindlmann,Klaus Mueller,Bernhard Preim,Markus Wacker	2005	Visual Medicine: Part Two - Advanced Applications of Medical Imaging.
IEEE Visualization	VolumeExplorer: Roaming Large Volumes to Couple Visualization and Data Processing for Oil and Gas Exploration.	Laurent Castanie,Bruno Lévy,Fabien Bosquet	2005	VolumeExplorer: Roaming Large Volumes to Couple Visualization and Data Processing for Oil and Gas Exploration.
IEEE Visualization	COTS Cluster-based Sort-last Rendering: Performance Evaluation and Pipelined Implementation.	Xavier Cavin,Christophe Mion,Alain Filbois	2005	COTS Cluster-based Sort-last Rendering: Performance Evaluation and Pipelined Implementation.
IEEE Visualization	Phonon Tracing for Auralization and Visualization of Sound.	Martin Bertram,Eduard Deines,Jan Mohring,Jevgenijs Jegorovs,Hans Hagen	2005	Phonon Tracing for Auralization and Visualization of Sound.
IEEE Visualization	OpenGL Multipipe SDK: A Toolkit for Scalable Parallel Rendering.	Praveen Bhaniramka,Philippe C. D. Robert,Stefan Eilemann	2005	OpenGL Multipipe SDK: A Toolkit for Scalable Parallel Rendering.
IEEE Visualization	Quality Mesh Generation for Molecular Skin Surfaces Using Restricted Union of Balls.	Ho-Lun Cheng,Xinwei Shi	2005	Quality surface meshes for molecular models are desirable in the studies of protein shapes and functionalities. However, there is still no robust software that is capable to generate such meshes with good quality. In this paper, we present a Delaunay-based surface triangulation algorithm generating quality surface meshes for the molecular skin model. We expand the restricted union of balls along the surface and generate an @e-sampling of the skin surface incrementally. At the same time, a quality surface mesh is extracted from the Delaunay triangulation of the sample points. The algorithm supports robust and efficient implementation and guarantees the mesh quality and topology as well. Our results facilitate molecular visualization and have made a contribution towards generating quality volumetric tetrahedral meshes for the macromolecules.
IEEE Visualization	A Contract Based System For Large Data Visualization.	Henry R. Childs,Eric Brugger,Kathleen S. Bonnell,Jeremy S. Meredith,Mark Miller,Brad Whitlock,Nelson Max	2005	A Contract Based System For Large Data Visualization.
IEEE Visualization	Batched Multi Triangulation.	Paolo Cignoni,Fabio Ganovelli,Enrico Gobbetti,Fabio Marton,Federico Ponchio,Roberto Scopigno	2005	Batched Multi Triangulation.
IEEE Visualization	A Visualization Tool for fMRI Data Mining.	Nicu D. Cornea,Ulukbek Ibraev,Deborah Silver,Paul B. Kantor,Ali Shokoufandeh,Jeff Abrahamson,Sven J. Dickinson	2005	A Visualization Tool for fMRI Data Mining.
IEEE Visualization	Curve-Skeleton Applications.	Nicu D. Cornea,Deborah Silver,Patrick Min	2005	Curve-Skeleton Applications.
IEEE Visualization	Hardware-Accelerated 3D Visualization of Mass Spectrometry Data.	Jose De Corral,Hanspeter Pfister	2005	Hardware-Accelerated 3D Visualization of Mass Spectrometry Data.
IEEE Visualization	Dataset Traversal with Motion-Controlled Transfer Functions.	Carlos D. Correa,Deborah Silver	2005	Dataset Traversal with Motion-Controlled Transfer Functions.
IEEE Visualization	Visualization of White Matter Tracts with Wrapped Streamlines.	Frank Enders,Natascha Sauber,Dorit Merhof,Peter Hastreiter,Christopher Nimsky,Marc Stamminger	2005	Visualization of White Matter Tracts with Wrapped Streamlines.
IEEE Visualization	Prefiltered Gaussian Reconstruction for High-Quality Rendering of Volumetric Data sampled on a Body-Centered Cubic Grid.	Balázs Csébfalvi	2005	Prefiltered Gaussian Reconstruction for High-Quality Rendering of Volumetric Data sampled on a Body-Centered Cubic Grid.
IEEE Visualization	Clustering Techniques for Out-of-Core Multi-resolution Modeling.	Emanuele Danovaro,Leila De Floriani,Enrico Puppo,Hanan Samet	2005	Clustering Techniques for Out-of-Core Multi-resolution Modeling.
IEEE Visualization	Illustrative Display of Hidden Iso-Surface Structures.	Jan Fischer,Dirk Bartz,Wolfgang Straßer	2005	Illustrative Display of Hidden Iso-Surface Structures.
IEEE Visualization	Example-based Volume Illustrations.	Aidong Lu,David S. Ebert	2005	Example-based Volume Illustrations.
IEEE Visualization	The Visualization Process: The Path from Data to Insight.	Kelly P. Gaither,David S. Ebert,Daniel Weiskopf,Pat Hanrahan	2005	The Visualization Process: The Path from Data to Insight.
IEEE Visualization	Distributed Data Management for Large Volume Visualization.	Jinzhu Gao,Jian Huang,C. Ryan Johnson,Scott Atchley,James Arthur Kohl	2005	Distributed Data Management for Large Volume Visualization.
IEEE Visualization	A Concurrent Visualization System for High-Performance Computational Simulations.	Joe Groner,Matthew Lee,Joel P. Martin,Robert J. Moorhead II,James Newman	2005	A Concurrent Visualization System for High-Performance Computational Simulations.
IEEE Visualization	Topology-based Simplification for Feature Extraction from 3D Scalar Fields.	Attila Gyulassy,Vijay Natarajan,Valerio Pascucci,Peer-Timo Bremer,Bernd Hamann	2005	Topology-based Simplification for Feature Extraction from 3D Scalar Fields.
IEEE Visualization	The Visible Radio: Process Visualization of a Software-Defined Radio.	Matthew Hall,Alex Betts,Donna Cox,David Pointer,Volodymyr V. Kindratenko	2005	The Visible Radio: Process Visualization of a Software-Defined Radio.
IEEE Visualization	Multimodal Exploration of the Fourth Dimension.	Andrew J. Hanson,Hui Zhang	2005	Multimodal Exploration of the Fourth Dimension.
IEEE Visualization	Integrated 3D Visualization of fMRI and DTI tractography.	Jan Hardenbergh,Bradley R. Buchbinder,Shirley A. M. Thurston,Jonathan W. Lombardi,Gordon J. Harris	2005	Integrated 3D Visualization of fMRI and DTI tractography.
IEEE Visualization	On the Optimization of Visualizations of Complex Phenomena.	Donald H. House,Alethea Bair,Colin Ware	2005	On the Optimization of Visualizations of Complex Phenomena.
IEEE Visualization	Design and Evaluation in Visualization Research.	Donald H. House,Victoria Interrante,David H. Laidlaw,Russell M. Taylor II,Colin Ware	2005	Design and Evaluation in Visualization Research.
IEEE Visualization	Teniae Coli Guided Navigation and Registration for Virtual Colonoscopy.	Adam Huang,Dave Roy,Marek Franaszek,Ronald M. Summers	2005	Teniae Coli Guided Navigation and Registration for Virtual Colonoscopy.
IEEE Visualization	Visualizing Data with Motion.	Daniel E. Huber,Christopher G. Healey	2005	Visualizing Data with Motion.
IEEE Visualization	HOT- Lines: Tracking Lines in Higher Order Tensor Fields.	Mario Hlawitschka,Gerik Scheuermann	2005	HOT- Lines: Tracking Lines in Higher Order Tensor Fields.
IEEE Visualization	Streaming Meshes.	Martin Isenburg,Peter Lindstrom	2005	Streaming Meshes.
IEEE Visualization	Illustration-inspired techniques for visualizing time-varying data.	Alark Joshi,Penny Rheingans	2005	Illustration-inspired techniques for visualizing time-varying data.
IEEE Visualization	Visualization of Time-Dependent Remote Adaptive Mesh Refinement Data.	Ralf Kähler,Steffen Prohaska,Andrei Hutanu,Hans-Christian Hege	2005	Visualization of Time-Dependent Remote Adaptive Mesh Refinement Data.
IEEE Visualization	Scale-Invariant Volume Rendering.	Martin Kraus	2005	Scale-Invariant Volume Rendering.
IEEE Visualization	Hardware-Accelerated Simulated Radiography.	Daniel E. Laney,Steven P. Callahan,Nelson L. Max,Cláudio T. Silva,Steven Langer,Randall Frank	2005	Hardware-Accelerated Simulated Radiography.
IEEE Visualization	Visualization and Mining of Temporal Data.	Eamonn J. Keogh	2005	Visualization and Mining of Temporal Data.
IEEE Visualization	Visual Analysis and Exploration of Fluid Flow in a Cooling Jacket.	Robert S. Laramee,Christoph Garth,Helmut Doleisch,Jürgen Schneider,Helwig Hauser,Hans Hagen	2005	Visual Analysis and Exploration of Fluid Flow in a Cooling Jacket.
IEEE Visualization	Scientific Visualization of time-Varying Oceanographic and Meteorological Data using VR.	Chang S. Kim,Jinah Kim,Hak Soo Lim,K. Parks,Jinah Park	2005	Scientific Visualization of time-Varying Oceanographic and Meteorological Data using VR.
IEEE Visualization	Exploiting Frame-to-Frame Coherence for Accelerating High-Quality Volume Raycasting on Graphics Hardware.	Thomas Klein,Magnus Strengert,Simon Stegmaier,Thomas Ertl	2005	Exploiting Frame-to-Frame Coherence for Accelerating High-Quality Volume Raycasting on Graphics Hardware.
IEEE Visualization	Statistically Quantitative Volume Visualization.	Joe Michael Kniss,Robert L. Van Uitert Jr.,Abraham Stephens,Guo-Shi Li,Tolga Tasdizen,Charles D. Hansen	2005	Statistically Quantitative Volume Visualization.
IEEE Visualization	General Purpose Computation on Graphics Hardware.	Aaron E. Lefohn,Ian Buck,Patrick S. McCormick,John D. Owens,Timothy J. Purcell,Robert Strzodka	2005	General Purpose Computation on Graphics Hardware.
IEEE Visualization	The Application of GPU Particle Tracing to Diffusion Tensor Field Visualization.	Polina Kondratieva,Jens Krüger,Rüdiger Westermann	2005	The Application of GPU Particle Tracing to Diffusion Tensor Field Visualization.
IEEE Visualization	A Handheld Flexible Display System.	"Jonathan Konieczny,Clement Shimizu,Gary W. Meyer,D'nardo Colucci"	2005	A Handheld Flexible Display System.
IEEE Visualization	Differential Protein Expression Analysis via Liquid-Chromatography/Mass-Spectrometry Data Visualization.	Lars Linsen,Julia Löcherbach,Matthias Berth,Jörg Bernhardt	2005	Differential Protein Expression Analysis via Liquid-Chromatography/Mass-Spectrometry Data Visualization.
IEEE Visualization	Enclosure Sphere Based Cell Visibility for Virtual Endoscopy.	Jianfei Liu,Xiaopeng Zhang	2005	Enclosure Sphere Based Cell Visibility for Virtual Endoscopy.
IEEE Visualization	Copyright.		2005	Copyright.
IEEE Visualization	Reflection Nebula Visualization.	Marcus A. Magnor,Kristian Hildebrand,Andrei Lintu,Andrew J. Hanson	2005	Reflection Nebula Visualization.
IEEE Visualization	Illuminated Lines Revisited.	Ovidio Mallo,Ronald Peikert,Christian Sigg,Filip Sadlo	2005	Illuminated Lines Revisited.
IEEE Visualization	Preface.		2005	Preface.
IEEE Visualization	The EventScope Remote Science Operations Tool: Using a Visualization of a Remote Location as an Interface for Robotic Exploration.	Eben Myers,Peter Coppin,Michael Wagner,Luisa Lu,W. Ronald McCloskey,David Seneker,Kevin McCloskey	2005	The EventScope Remote Science Operations Tool: Using a Visualization of a Remote Location as an Interface for Robotic Exploration.
IEEE Visualization	The Software Interface to the 3D-Force Microscope.	"David Marshburn,Christopher Weigle,Benjamin G. Wilde,Russell M. Taylor II,Kalpit Desai,J. K. Fisher,Jeremy Cribb,E. Timothy O'Brien,Richard Superfine"	2005	The Software Interface to the 3D-Force Microscope.
IEEE Visualization	Interactive Visual Analysis and Exploration of Injection Systems Simulations.	Kresimir Matkovic,Mario Jelovic,Josip Juric,Zoltan Konyha,Denis Gracanin	2005	Interactive Visual Analysis and Exploration of Injection Systems Simulations.
IEEE Visualization	Visualizing Tensor Fields in Geomechanics.	Alisa Neeman,Boris Jeremic,Alex Pang	2005	Visualizing Tensor Fields in Geomechanics.
IEEE Visualization	A web-based visualization solution for very large 2D datasets.	Rudolph L. Maxwell,Taras V. Gerya,David A. Yuen	2005	A web-based visualization solution for very large 2D datasets.
IEEE Visualization	Keynote.		2005	Keynote.
IEEE Visualization	Farthest Point Seeding for Efficient Placement of Streamlines.	Abdelkrim Mebarki,Pierre Alliez,Olivier Devillers	2005	Farthest Point Seeding for Efficient Placement of Streamlines.
IEEE Visualization	Exploring Defects in Nematic Liquid Crystals.	Ketan Mehta,Matthew Lee,T. J. Jankun-Kelly	2005	Exploring Defects in Nematic Liquid Crystals.
IEEE Visualization	Capstone Address.		2005	Capstone Address.
IEEE Visualization	"End Users' Perspectives on Volume Rendering in Medical Imaging: A job well done or not over yet?"	Michael Meissner,Karel J. Zuiderveld,Gordon Harris,John R. Lesser,Michael W. Vannier	2005	"End Users' Perspectives on Volume Rendering in Medical Imaging: A job well done or not over yet?"
IEEE Visualization	Surface Reconstruction Via Contour Metamorphosis: An Eulerian Approach With Lagrangian Particle Tracking.	Ola Nilsson,David E. Breen,Ken Museth	2005	Surface Reconstruction Via Contour Metamorphosis: An Eulerian Approach With Lagrangian Particle Tracking.
IEEE Visualization	Committees.		2005	Committees.
IEEE Visualization	Interactive CSG Trees Inside Complex Scenes.	Jan Ohlenburg,Jan Muller	2005	Interactive CSG Trees Inside Complex Scenes.
IEEE Visualization	Awards.		2005	Awards.
IEEE Visualization	Profile Flags: a Novel Metaphor for Probing of T2 Maps.	Matej Mlejnek,Pierre Ermes,Anna Vilanova,Rob van der Rijt,Harrie van den Bosch,Frans A. Gerritsen,M. Eduard Gröller	2005	Profile Flags: a Novel Metaphor for Probing of T2 Maps.
IEEE Visualization	Evaluation of Fiber Clustering Methods for Diffusion Tensor Imaging.	Bart Moberts,Anna Vilanova,Jarke J. van Wijk	2005	Evaluation of Fiber Clustering Methods for Diffusion Tensor Imaging.
IEEE Visualization	Title Page.		2005	Title Page.
IEEE Visualization	Stream-Processing Points.	Renato Pajarola	2005	Stream-Processing Points.
IEEE Visualization	Evaluation of Areal Touch Feedback for Palpation Simulation.	Jinah Park,Sang-Youn Kim,Ki-Uk Kyung,Dong-Soo Kwon	2005	Evaluation of Areal Touch Feedback for Palpation Simulation.
IEEE Visualization	VolQD: Direct Volume Rendering of Multi-million Atom Quantum Dot Simulations.	Wei Qiao,David S. Ebert,Alireza Entezari,Marek Korkusinski,Gerhard Klimeck	2005	VolQD: Direct Volume Rendering of Multi-million Atom Quantum Dot Simulations.
IEEE Visualization	PSALM: A Data Model for Pervasive Visualization.	Huamin Qu,Hong Zhou,Yingcai Wu	2005	PSALM: A Data Model for Pervasive Visualization.
IEEE Visualization	Effectively Visualizing Large Networks Through Sampling.	Davood Rafiei,Stephen Curial	2005	Effectively Visualizing Large Networks Through Sampling.
IEEE Visualization	Rendering Tetrahedral Meshes with Higher-Order Attenuation Functions for Digital Radiograph Reconstruction.	Ofri Sadowsky,Jonathan D. Cohen,Russell H. Taylor	2005	Rendering Tetrahedral Meshes with Higher-Order Attenuation Functions for Digital Radiograph Reconstruction.
IEEE Visualization	Particle Generation from User-specified Transfer Function for Point-based Volume Rendering.	Naohisa Sakamoto,Koji Koyamada	2005	Particle Generation from User-specified Transfer Function for Point-based Volume Rendering.
IEEE Visualization	Interactive Poster: Using CoMIRVA for Visualizing Similarities Between Music Artists.	Markus Schedl,Peter Knees,Gerhard Widmer	2005	Interactive Poster: Using CoMIRVA for Visualizing Similarities Between Music Artists.
IEEE Visualization	Fast Visualization by Shear-Warp on Quadratic Super-Spline Models Using Wavelet Data Decompositions.	Gregor Schlosser,Jürgen Hesser,Frank Zeilfelder,Christian Rössl,Reinhard Männer,Günther Nürnberger,Hans-Peter Seidel	2005	Fast Visualization by Shear-Warp on Quadratic Super-Spline Models Using Wavelet Data Decompositions.
IEEE Visualization	Framework for Visualizing Higher-Order Basis Functions.	"William J. Schroeder,François Bertel,Mathieu Malaterre,David Thompson,Philippe P. Pébay,Robert M. O'Bara,Saurabh Tendulkar"	2005	Framework for Visualizing Higher-Order Basis Functions.
IEEE Visualization	Color in Information Display.		2005	Color in Information Display.
IEEE Visualization	"NASA's Scientific Visualization Studio Image Server."	Eric Sokolowsky,Horace Mitchell,Jeff de La Beaujardière	2005	"NASA's Scientific Visualization Studio Image Server."
IEEE Visualization	View-dependent Meshing And Rendering of Tetrahedral Meshes.	Ralf Sondershaus,Wolfgang Straßer	2005	View-dependent Meshing And Rendering of Tetrahedral Meshes.
IEEE Visualization	Volume rendering of smoke propagation CFD data.	Oliver Staubli,Christian Sigg,Ronald Peikert,Markus H. Gross,Daniel Gubler	2005	Volume rendering of smoke propagation CFD data.
IEEE Visualization	Opening the Can of Worms: An Exploration Tool for Vortical Flows.	Simon Stegmaier,Ulrich Rist,Thomas Ertl	2005	Opening the Can of Worms: An Exploration Tool for Vortical Flows.
IEEE Visualization	Query-Driven Visualization of Large Data Sets.	Kurt Stockinger,John Shalf,Kesheng Wu,E. Wes Bethel	2005	Query-Driven Visualization of Large Data Sets.
IEEE Visualization	Visualization with stylized line primitives.	Carsten Stoll,Stefan Gumhold,Hans-Peter Seidel	2005	Visualization with stylized line primitives.
IEEE Visualization	Illustration and Photography Inspired Visualization of Flows and Volumes.	Nikolai A. Svakhine,Yun Jang,David S. Ebert,Kelly P. Gaither	2005	Illustration and Photography Inspired Visualization of Flows and Volumes.
IEEE Visualization	A Feature-Driven Approach to Locating Optimal Viewpoints for Volume Visualization.	Shigeo Takahashi,Issei Fujishiro,Yuriko Takeshima,Tomoyuki Nishita	2005	A Feature-Driven Approach to Locating Optimal Viewpoints for Volume Visualization.
IEEE Visualization	Extraction of Parallel Vector Surfaces in 3D Time-Dependent Fields and Application to Vortex Core Line Tracking.	Holger Theisel,Jan Sahner,Tino Weinkauf,Hans-Christian Hege,Hans-Peter Seidel	2005	Extraction of Parallel Vector Surfaces in 3D Time-Dependent Fields and Application to Vortex Core Line Tracking.
IEEE Visualization	Eyegaze Analysis of Displays With Combined 2D and 3D Views.	Melanie Tory,M. Stella Atkins,Arthur E. Kirkpatrick,Marios Nicolaou,Guang-Zhong Yang	2005	Eyegaze Analysis of Displays With Combined 2D and 3D Views.
IEEE Visualization	Neuroanatomical image alignment.	Issac J. Trotts,Bruno A. Olshausen,Edward Jones	2005	Neuroanatomical image alignment.
IEEE Visualization	Opening the Black Box - Data Driven Visualization of Neural Network.	Fan-Yin Tzeng,Kwan-Liu Ma	2005	Opening the Black Box - Data Driven Visualization of Neural Network.
IEEE Visualization	Interpolation And Visualization For Advected Scalar Fields.	Shyh-Kuang Ueng,Sheng-Chuan Wang	2005	Interpolation And Visualization For Advected Scalar Fields.
IEEE Visualization	Volume Exploration Using Spatially Linked Transfer Functions.	Vivek Vaidya,Rakesh Mullick,Navneeth Subramanian	2005	Volume Exploration Using Spatially Linked Transfer Functions.
IEEE Visualization	High Performance Volume Splatting for Visualization of Neurovascular Data.	Fernando Vega Higuera,Peter Hastreiter,Rudolf Fahlbusch,Günther Greiner	2005	High Performance Volume Splatting for Visualization of Neurovascular Data.
IEEE Visualization	Understanding Visualization through Spatial Ability Differences.	Maria C. Velez,Deborah Silver,Marilyn Tremaine	2005	Understanding Visualization through Spatial Ability Differences.
IEEE Visualization	Illustrative Visualization.	Ivan Viola,M. Eduard Gröller,Markus Hadwiger,Katja Bühler,Bernhard Preim,Mario Costa Sousa,David S. Ebert,Don Stredney	2005	Illustrative Visualization.
IEEE Visualization	Visualization in the Earth Sciences: A Discussion on Various Visualization Methods using amira.	S. M. Wang,M. Damon,David A. Yuen	2005	Visualization in the Earth Sciences: A Discussion on Various Visualization Methods using amira.
IEEE Visualization	Reconstructing Manifold and Non-Manifold Surfaces from Point Clouds.	Jianning Wang,Manuel M. Oliveira,Arie E. Kaufman	2005	Reconstructing Manifold and Non-Manifold Surfaces from Point Clouds.
IEEE Visualization	The Magic Volume Lens: An Interactive Focus+Context Technique for Volume Rendering.	Lujin Wang,Ye Zhao,Klaus Mueller,Arie E. Kaufman	2005	The Magic Volume Lens: An Interactive Focus+Context Technique for Volume Rendering.
IEEE Visualization	Visualizing intersecting surfaces with nested-surface techniques.	Chris Weigle,Russell M. Taylor II	2005	Visualizing intersecting surfaces with nested-surface techniques.
IEEE Visualization	Extracting Higher Order Critical Points and Topological Simplification of 3D Vector Fields.	Tino Weinkauf,Holger Theisel,Kuangyu Shi,Hans-Christian Hege,Hans-Peter Seidel	2005	Extracting Higher Order Critical Points and Topological Simplification of 3D Vector Fields.
IEEE Visualization	Visualization in the Einstein Year 2005: A Case Study on Explanatory and Illustrative Visualization of Relativity and Astrophysics.	Daniel Weiskopf,Marc Borchers,Thomas Ertl,Martin Falk,Oliver Fechtig,Regine Frank,Frank Grave,Andreas King,Ute Kraus,Thomas Müller,Hans-Peter Nollert,Isabel Rica Mendez,Hanns Ruder,Corvin Zahn,Michael Zatloukal,Tobias Schafhitzel,Sonja Schär	2005	Visualization in the Einstein Year 2005: A Case Study on Explanatory and Illustrative Visualization of Relativity and Astrophysics.
IEEE Visualization	Particle and Texture Based Spatiotemporal Visualization of Time-Dependent Vector Fields.	Daniel Weiskopf,Frederik Schramm,Gordon Erlebacher,Thomas Ertl	2005	Particle and Texture Based Spatiotemporal Visualization of Time-Dependent Vector Fields.
IEEE Visualization	Eyelet Particle Tracing - Steady Visualization of Unsteady Flow.	Alexander Wiebel,Gerik Scheuermann	2005	Eyelet Particle Tracing - Steady Visualization of Unsteady Flow.
IEEE Visualization	The Value of Visualization.	Jarke J. van Wijk	2005	The Value of Visualization.
IEEE Visualization	Visualization of the Genus of Knots.	Jarke J. van Wijk,Arjeh M. Cohen	2005	Visualization of the Genus of Knots.
IEEE Visualization	Evolutionary Morphing.	David F. Wiley,Nina Amenta,Dan A. Alcantara,Deboshmita Ghosh,Yong Joo Kil,Eric Delson,Will Harcourt-Smith,Katherine St. John,F. James Rohlf,Bernd Hamann	2005	Evolutionary Morphing.
IEEE Visualization	Sort-Middle Multi-Projector Immediate-Mode Rendering in Chromium.	Jorge Luis Williams,Robert E. Hiromoto	2005	Sort-Middle Multi-Projector Immediate-Mode Rendering in Chromium.
IEEE Visualization	Exploring 2D Tensor Fields Using Stress Nets.	Andrew Wilson,Rebecca M. Brannon	2005	Exploring 2D Tensor Fields Using Stress Nets.
IEEE Visualization	Perceptually-Based Comparisons of Direct Volume Rendered Images.	Hon-Cheng Wong,Huamin Qu,Un-Hong Wong,Zesheng Tang	2005	Perceptually-Based Comparisons of Direct Volume Rendered Images.
IEEE Visualization	A System for Interactive Volume Visualization on the PowerWall.	Paul R. Woodward,David H. Porter,Michael R. Knox,Steven T. Andringa,Alex J. Larson,Aaron Stender	2005	A System for Interactive Volume Visualization on the PowerWall.
IEEE Visualization	Anisotropic Ellipsoidal Smoothing of Volume Data.	Bai Xue,Nelson Max,Cristina Siegerist	2005	Anisotropic Ellipsoidal Smoothing of Volume Data.
IEEE Visualization	Strategy for Seeding 3D Streamlines.	Xiaohong Ye,David Kao,Alex Pang	2005	Strategy for Seeding 3D Streamlines.
IEEE Visualization	High Dynamic Range Volume Visualization.	Xiaoru Yuan,Minh X. Nguyen,Baoquan Chen,David H. Porter	2005	High Dynamic Range Volume Visualization.
IEEE Visualization	Volume Rendering of Real-Time 3D Echocardiographic Data.	Vladimir Zagrodsky,Raj Shekhar	2005	Volume Rendering of Real-Time 3D Echocardiographic Data.
IEEE Visualization	2D Asymmetric Tensor Analysis.	Xiaoqiang Zheng,Alex Pang	2005	2D Asymmetric Tensor Analysis.
IEEE Visualization	Topological Structures of 3D Tensor Fields.	Xiaoqiang Zheng,Beresford N. Parlett,Alex Pang	2005	Topological Structures of 3D Tensor Fields.
IEEE Visualization	16th IEEE Visualization Conference (VIS 2005), 23-28 October 2005, Minneapolis, MN, USA		2005	16th IEEE Visualization Conference (VIS 2005), 23-28 October 2005, Minneapolis, MN, USA
Information Visualization	A rank-by-feature framework for interactive exploration of multidimensional data.	Jinwook Seo,Ben Shneiderman	2005	Interactive exploration of multidimensional data sets is challenging because: (1) it is difficult to comprehend patterns in more than three dimensions, and (2) current systems often are a patchwork of graphical and statistical methods leaving many researchers uncertain about how to explore their data in an orderly manner. We offer a set of principles and a novel rank-by-feature framework that could enable users to better understand distributions in one (1D) or two dimensions (2D), and then discover relationships, clusters, gaps, outliers, and other features. Users of our framework can view graphical presentations (histograms, boxplots, and scatterplots), and then choose a feature detection criterion to rank 1D or 2D axis-parallel projections. By combining information visualization techniques (overview, coordination, and dynamic query) with summaries and statistical methods users can systematically examine the most important 1D and 2D axis-parallel projections. We summarize our Graphics, Ranking, and Interaction for Discovery (GRID) principles as: (1) study 1D, study 2D, then find features (2) ranking guides insight, statistics confirm. We implemented the rank-by-feature framework in the Hierarchical Clustering Explorer, but the same data exploration principles could enable users to organize their discovery process so as to produce more thorough analyses and extract deeper insights in any multidimensional data application, such as spreadsheets, statistical packages, or information visualization tools.
Information Visualization	Bioinformatics Visualization: Introduction to the Special Issue.	Chris North,Theresa-Marie Rhyne,Karen Duca	2005	Bioinformatics Visualization: Introduction to the Special Issue.
Information Visualization	Constructing and reconstructing the reorderable matrix.	Harri Siirtola,Erkki Mäkinen	2005	We consider the backgrounds, applications, implementations, and user interfaces of the reorderable matrix originally introduced by Jacques Bertin. As a new tool for handling the matrix, we propose a new kind of interface for interactive cluster analysis. As the main tool to order the rows and columns, we use the well-known barycenter heuristic. Two user tests are performed to verify the usefulness of the automatic tools.
Information Visualization	Self-organizing map learning nonlinearly embedded manifolds.	Timo Similä	2005	One of the main tasks in exploratory data analysis is to create an appropriate representation for complex data. In this paper, the problem of creating a representation for observations lying on a low-dimensional manifold embedded in high-dimensional coordinates is considered. We propose a modification of the Self-organizing map (SOM) algorithm that is able to learn the manifold structure in the high-dimensional observation coordinates. Any manifold learning algorithm may be incorporated to the proposed training strategy to guide the map onto the manifold surface instead of becoming trapped in local minima. In this paper, the Locally linear embedding algorithm is adopted. We use the proposed method successfully on several data sets with manifold geometry including an illustrative example of a surface as well as image data. We also show with other experiments that the advantage of the method over the basic SOM is restricted to this specific type of data.
Information Visualization	A system for visualizing and analyzing near-optimal protein sequence alignments.	Michael E. Smoot,Ellen J. Bass,Stephanie A. Guerlain,William R. Pearson	2005	System modeling and analysis is an imperfect art and as a consequence there are problems where alternative solutions may exist. Information visualization is often used to enhance system modeling and analysis efforts by taking advantage of human perceptual capabilities. Our research uses visual representations of alternative solutions as a way to aid understanding and foster insight into problems, their solutions, and the underlying algorithms. Specifically, our system visualizes large sets of near-optimal protein sequence alignments. Prior efforts focused on the visualization of near-optimal alignments, using an animation metaphor for providing context to detailed displays. This paper discusses novel extensions: (1) a new overview display, the enhanced path graph; (2) the integration of the detail-oriented (animation) and overview-oriented (path graph) displays; and (3) additional human-computer interaction capabilities for filtering, highlighting, and mixed-initiative interaction that allow researchers to combine their expertise with algorithmic output to gain insight into the sequence alignment algorithms and the problem domain. Two case studies illustrate the effectiveness of this system.
Information Visualization	A new pixel-oriented visualization technique through color image.	Frédéric Blanchard,Michel Herbin,Laurent Lucas	2005	Color image is often considered as a fundamental perceptual unit of visualization. In this paper, we suggest using this medium (color image) to summarize multidimensional data and thus to turn a data set into a meaningful insight. The methodology we use is based on the theory of Keim for designing pixel-oriented visualization techniques. The technique we propose consists in a three-step pipeline. The first one is devoted to dimensionality reduction by projecting multidimensional data into a three-dimensional space. In this work, we use the classical principal component analysis (PCA) to reduce the dimension to three. The second step, called color mapping, is based on the reverse color information transformation defined by Ohta et al. This stage is the main novelty of this work in addition to the pipeline itself. The third step consists in a pixel-oriented method to display large data sets with an image using space-filling curve techniques. The combination of these steps (first, dimensionality reduction with PCA, second, color mapping with color information of Ohta et al., and third, space-filling curve with Peano-Hilbert curve) allows us to obtain a new unsupervised visualization technique through color images. This blind (i.e. unsupervised) technique using a color image gives a previsualization that can be used before exploring the data set or choosing more effective colors. Some applications are proposed in the field of multicomponent image visualization.
Information Visualization	Visualizing Argumentation: Software tools for collaborative and educational sense-making.	Ramon Prudencio S. Toledo	2005	Visualizing Argumentation: Software tools for collaborative and educational sense-making.
Information Visualization	Supporting visual queries on medium-sized node-link diagrams.	Colin Ware,Robert J. Bobrow	2005	For reasons of clarity, a typical node-link diagram statically displayed on paper or a computer screen contains fewer than 30 nodes. However, many problems would benefit if far more complex information could be diagrammed. Following Munzner et al., we suggest that with interactive diagrams this may be possible. We describe an interactive technique whereby a subset of a larger network diagram is highlighted by being set into oscillatory motion when a node is selected with a mouse. The subset is determined by a breadth first search of the underlying graph starting from the selected node. This technique is designed to support visual queries on moderately large node-link diagrams containing up to a few thousand nodes. An experimental evaluation was carried out with networks having 32, 100, 320, 1000, and 3200 nodes respectively, and with four highlighting techniques: static highlighting, motion highlighting, static + motion highlighting, and none. The results show that the interactive highlighting methods support rapid visual queries of nodes in close topological proximity to one another, even for the largest diagrams tested. Without highlighting, error rates were high even for the smallest network that was evaluated. Motion highlighting and static highlighting were equally effective. A second experiment was carried out to evaluate methods for showing two subsets of a larger network simultaneously in such a way that both are clearly distinct. The specific task was to determine if the two subsets had nodes in common. The results showed that this task could be performed rapidly and with few errors if one subset was highlighted using motion and the other was highlighted using a static technique. We discuss the implications for information visualization.
Information Visualization	Visualizing large-scale data in educational, behavioral, psychometrical and social sciences: utilities and design patterns of the SEER computational and graphical statistics.	Christopher W. T. Chiu,Peter Pashley,Marilyn Seastrom,Peggy Carr	2005	"This paper introduces a graphical method SEE Repeated-measure data (SEER) to visually analyze data commonly collected in large-scale surveys, market research, biostatistics, and educational and psychological measurement. Many researchers in these disciplines encounter large amounts of data. Examples include the Law School Admission Test (LSAT) repeater scores, career paths of students graduated from college, essays scores in the writing assessments of the National Assessment of Educational Progress (NAEP), and scores derived from different test equating methods in the discipline of psychometrics. Efficiency, ease-of-interpretations, applicability, user interactions are challenges due to the graphical complexity in visualizing large-scale data sets. To overcome these challenges, the author expands a systematic data-visualization technique, called SEER. The SEER technique was originally designed to depict career paths and occupational stability for professionals in the science and engineering discipline. In this paper, the author summarizes this example and highlights its applications in legal education, psychometrics, and other related areas. The author also, (a) expands this technique to examine repeat test takers' scores, (b) illustrates how to monitor inter-rater consistency for essay scoring and for depicting multi-faceted data that involve human judgments, and (c) demonstrates how to investigate differences of test equating and scaling methodology using the SEER method. The broader impacts and design patterns of the SEER method are discussed."
Information Visualization	Animated interval scatter-plot views for the exploratory analysis of large-scale microarray time-course data.	Paul Craig,Jessie B. Kennedy,Andrew Cumming	2005	"Microarray technologies are a relatively new development that allow biologists to monitor the activity of thousands of genes (normally around 8,000) in parallel across multiple stages of a biological process. While this new perspective on biological functioning is recognised as having the potential to have a significant impact on the diagnosis, treatment, and prevention of diseases, it is only through effective analysis of the data produced that biologists can begin to unlock this potential. A significant obstacle to achieving effective analysis of microarray time-course is the combined scale and complexity of the data. This inevitably makes it difficult to reveal certain significant patterns in the data. In particular, it is less dominant patterns and, specifically, patterns that occur over smaller intervals of an experiment's overall time-frame that are more difficult to find. While existing techniques are capable of finding either unexpected patterns of activity over the majority of an experiment's time-frame or expected patterns of activity over smaller intervals of the time-frame, there are no techniques, or combination of techniques, that are suitable for finding unsuspected patterns of activity over smaller intervals. In order to overcome this limitation we have developed the Time-series Explorer, which specifically supports biologists in their attempts to reveal these types of pattern by allowing them to control an animated interval scatter-plot view of their data. This paper discusses aspects of the technique that make such an animated overview viable and describes the results of a user evaluation assessing the practical utility of the technique within the wider context of microarray time-series analysis as a whole."
Information Visualization	Dust & Magnet: multivariate information visualization using a magnet metaphor.	Ji Soo Yi,Rachel Melton,John T. Stasko,Julie A. Jacko	2005	The use of multivariate information visualization techniques is intrinsically difficult because the multidimensional nature of data cannot be effectively presented and understood on real-world displays, which have limited dimensionalities. However, the necessity to use these techniques in daily life is increasing as the amount and complexity of data grows explosively in the information age. Thus, multivariate information visualization techniques that are easier to understand and more accessible are needed for the general population. In order to meet this need, the present paper proposes Dust & Magnet, a multivariate information visualization technique using a magnet metaphor and various interactive techniques. The intuitive magnet metaphor and subsequent interactions facilitate the ease of learning this multivariate information visualization technique. A visualization tool such as Dust & Magnet has the potential to increase the acceptance of and utility for multivariate information by a broader population of users who are not necessarily knowledgeable about multivariate information visualization techniques.
Information Visualization	Visualizing biological pathways: requirements analysis, systems evaluation and research agenda.	Purvi Saraiya,Chris North,Karen Duca	2005	"Pathway diagrams are used by life scientists to represent complex interactions at the molecular level in living cells. The recent shift towards data-intensive bioinformatics and systems-level science has created a strong need for advanced pathway visualizations that support exploratory analysis. This paper presents a comprehensive list of requirements for pathway visualization systems, based on interviews conducted to understand life scientists' needs for pathway analysis. A variety of existing pathway visualization systems are examined, to analyze common approaches by which the contemporary systems address these requirements. A heuristic evaluation, by biology domain experts, of five popular pathway visualization systems is conducted to analyze the end-user perception of these systems. Based on these studies, a research agenda is presented concerning five critical requirements for pathway visualization systems. If addressed effectively, these requirements can prove to be most helpful in supporting exploratory pathway analysis. These include: (1) automated construction and updating of pathways by searching literature databases, (2) overlaying information on pathways in a biologically relevant format, (3) linking pathways to multi-dimensional data from high-throughput experiments such as microarrays, (4) overviewing multiple pathways simultaneously with interconnections between them, (5) scaling pathways to higher levels of abstraction to analyze effects of complex molecular interactions at higher levels of biological organization."
Information Visualization	A Framework for Visualization of Microarray Data and Integrated Meta Information.	Nils Gehlenborg,Janko Dietzsch,Kay Nieselt	2005	We have developed a methodology that allows integration of microarray data and meta information within a visualization in order to guide the investigator during data exploration and analysis. A simple mathematical framework is introduced that uses scoring functions to map meta information to relevance ratings of genes. To explore the potential of this framework we extended the traditional heatmap with new features to graphically represent the relevance ratings. These ratings are visualized by an additional color gradient, by scaling the vertical height of matrix rows, by rearranging rows or by inserting new columns into the heatmap. This visualization is called an enhanced heatmap. We have applied our approach to microarray data of the Saccharomyces cerevisiae cell cycle, complemented with supplemental data that we both derived from the microarray data itself and retrieved from public databases. Using these data we demonstrate how this visualization concept can be efficiently used to identify certain features of genes and to detect inconsistencies in the data. Thus, the investigator has the possibility to get an overview of data from various sources and at the same time can gain a deeper insight into the structure of the combined data. The concept is not restricted to heatmaps, and can be used to extend further visualization techniques, such as profile plots. We found that our method is a powerful tool to integrate supplemental data into microarray visualizations and that it increases the efficiency of visual data exploration, which is a fundamental part of microarray data analyses.
Information Visualization	From the lab to the field - Steve Roth: a Memoriam.	Nahum D. Gershon,Jake Kolojejchick	2005	From the lab to the field - Steve Roth: a Memoriam.
Information Visualization	On the readability of graphs using node-link and matrix-based representations: a controlled experiment and statistical analysis.	Mohammad Ghoniem,Jean-Daniel Fekete,Philippe Castagliola	2005	In this article, we describe a taxonomy of generic graph related tasks along with a computer-based evaluation designed to assess the readability of two representations of graphs: matrix-based representations and node-link diagrams. This evaluation encompasses seven generic tasks and leads to insightful recommendations for the representation of graphs according to their size and density. Typically, we show that when graphs are bigger than twenty vertices, the matrix-based visualization outperforms node-link diagrams on most tasks. Only path finding is consistently in favor of node-link diagrams throughout the evaluation.
Information Visualization	Extending taxonomic visualisation to incorporate synonymy and structural markers.	Martin Graham,Jessie B. Kennedy	2005	The visualisation of taxonomic hierarchies has evolved from indented lists of names to techniques that can display thousands of nodes and onto hundreds of thousands of nodes over multiple taxonomies. However, challenges remain within multiple hierarchy visualisation, and for taxonomic hierarchy visualisation in particular. Firstly, at present, there is no support for handling specific taxonomic information such as synonymy, with current visualisations matching solely on names. Synonymy is extremely important as it reflects expert opinion on the compatibility of data held in separate taxonomies, and is needed to produce an accurate picture of taxonomic overlap. Also, current techniques for exploring large hierarchies find it difficult to convey internal reorganisations between hierarchies, with most systems showing only addition, removal or wide-ranging fragmentation of information between taxonomies. Finding the source of changes that have occurred within an existing structure is currently only achievable through exhaustive drill-down exploration. This paper describes work that tackles these problems, incorporating synonymy information into a model for multiple hierarchy visualisation of large taxonomies, and also detailing techniques that aid navigation for discovering structural re-organisations between hierarchies and for revealing information about nodes that lie below the effective display resolution of the hierarchy layout. Two examples on real taxonomic data sets are annotated to show the effectiveness of these techniques in operation.
Information Visualization	GeoTime information visualization.	Thomas Kapler,William Wright	2005	GeoTime information visualization.
Information Visualization	Exploratory visualization of array-based comparative genomic hybridization.	Robert Kincaid,Amir Ben-Dor,Zohar Yakhini	2005	Recent developments in DNA microarray technology have enabled a new and highly effective platform for performing comparative genomic hybridization (CGH) measurements. CGH measures anomalies in DNA copy number. Such copy number changes are now thought to play an important role in a number of diseases, particularly cancer and developmental disorders, and may also lead to important insights relevant to personalized treatments of such diseases. In contrast to gene expression measurements, the genomic positions of probe targets and their correlation to genomic aberrations lead to a natural ordering of the data. This ordering can be leveraged in data visualizations of CGH measurements. This paper describes a research prototype we have named VistaChrom, which provides a highly interactive, exploratory visualization scheme for analyzing array-based CGH data. For efficient navigation, VistaChrom is based on tiled, multi-level coordinated views organized by genome, chromosome, gene and microarray probe. Raw probe data, moving averages, and statistical measures can all be displayed individually or simultaneously to aid visual discovery of significant genomic aberrations. Visual analysis can be performed with single arrays or multi-array studies. The result is a novel and effective environment for visually analyzing CGH data. Example visualization results are shown for two different data sets derived from tumor cell lines. The application also provides a framework for further exploring advanced computational methods for aberration analysis.
Information Visualization	Visualizing and discovering non-trivial patterns in large time series databases.	Jessica Lin,Eamonn J. Keogh,Stefano Lonardi	2005	Data visualization techniques are very important for data analysis, since the human eye has been frequently advocated as the ultimate data-mining tool. However, there has been surprisingly little work on visualizing massive time series data sets. To this end, we developed VizTree, a time series pattern discovery and visualization system based on augmenting suffix trees. VizTree visually summarizes both the global and local structures of time series data at the same time. In addition, it provides novel interactive solutions to many pattern discovery problems, including the discovery of frequently occurring patterns (motif discovery), surprising patterns (anomaly detection), and query by content. VizTree works by transforming the time series into a symbolic representation, and encoding the data in a modified suffix tree in which the frequency and other properties of patterns are mapped onto colors and other visual properties. We demonstrate the utility of our system by comparing it with state-of-the-art batch algorithms on several real and synthetic data sets. Based on the tree structure, we further device a coefficient which measures the dissimilarity between any two time series. This coefficient is shown to be competitive with the well-known Euclidean distance.
Information Visualization	The agent visualization system: a graphical and textual representation for multi-agent systems.	Hala Mostafa,Reem Bahgat	2005	As scientists from various domains increasingly resort to agent-based simulation for a more thorough understanding of real-world phenomena, the need for a simulation environment that facilitates rapid development of multi-agent systems is growing. Such a platform should provide means of visualizing the simulated scenario. In this paper we present the agent visualization system, the first system of its kind to specifically focus on catering to the visualization needs of agent-based simulation. The proposed system is a generic add-on that equips a simulation environment with a rich set of visualization facilities offering a variety of textual and graphical browsers that allow the modeler to detect trends and relationships in the simulation scenario. Some techniques from the field of information visualization were adapted and added to the system, while others were devised especially to be used in it. Regardless of their origin, all visualization techniques were thoroughly revised to make them generic enough to fit in our generic system. Agent visualization is more challenging than traditional information visualization in more than one respect. One of them is that the data to be visualized is not static; the simulation system is constantly producing data with every time step. Moreover, the sheer amount of data, together with its diversity, call for special adaptations to ensure that the system remains responsive and generic. To illustrate the various features of the proposed agent visualization system, we present a visualization of MicroTerra; a simulation scenario involving a group of beings trying to maximize their food intake.
Information Visualization	Introduction to the Special Section.	Tamara Munzner,Matthew O. Ward	2005	Introduction to the Special Section.
Information Visualization	EncCon: an approach to constructing interactive visualization of large hierarchical data.	Quang Vinh Nguyen,Mao Lin Huang	2005	"This paper describes a new technique called EncCon for visualizing and navigating large hierarchical information. This technique consists of two components: visualization and navigation. Visualization uses a fast enclosure+ connection method to calculate the geometrical layout for the display of large hierarchies in a two-dimensional space. Our technique uses a rectangular division algorithm for recursively positioning the graph. This visualization aims to maximize the utilization of display space while retaining a good geometrical layout as well as a clear (explicit) presentation of the hierarchical structure of graphs. This paper also presents an experimental evaluation of EncCon's layout algorithm. Besides the layout algorithm, EncCon uses a new focus + context viewing technique for the navigation of large hierarchies. We use the zooming + layering concept to achieve the focus + context viewing, rather than the traditional enlarge + embedded concept, which is used by most of the available focus + context techniques. Technically, it employs semitransparency to achieve the display of two layers of information in z-coordination at the same visualization. Both context view and detail view are drawn at two separate layers. These layers are then displayed in an overlapped manner at the same physical screen space."
Information Visualization	Image fusion based on topographic mappings using the hyperbolic space.	Axel Saalbach,Jörg Ontrup,Helge Ritter,Tim W. Nattkemper	2005	The analysis of multivariate image data is a field of research that is becoming increasingly important in a broad range of applications from remote sensing to medical imaging. While traditional scientific visualization techniques are often not suitable for the analysis of this kind of data, methods of image fusion have evolved as a promising approach for synergistic data integration. In this paper, a new approach for the analysis of multivariate image data by means of image fusion is presented, which employs topographic mapping techniques based on non-Euclidean geometry. The hyperbolic self-organizing map (HSOM) facilitates the exploration of high-dimensional data and provides an interface in the tradition of distortion-oriented presentation techniques. For the analysis of hidden patterns and spatial relationships, the HSOM gives rise to an intuitive and efficient framework for the dynamic visualization of multivariate image data by means of color. In an application, the hyperbolic data explorer (HyDE) is employed for the visualization of image data from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). Using 12 image sequences from breast cancer research, the method is introduced by different visual representations of the data and is also quantitatively analyzed. The HSOM is compared to different standard classifiers and evaluated with respect to topology preservation.
IEEE Trans. Vis. Comput. Graph.	A Sampling Framework for Accurate Curvature Estimation in Discrete Surfaces.	Gady Agam,Xiaojing Tang	2005	Accurate curvature estimation in discrete surfaces is an important problem with numerous applications. Curvature is an indicator of ridges and can be used in applications such as shape analysis and recognition, object segmentation, adaptive smoothing, anisotropic fairing of irregular meshes, and anisotropic texture mapping. In this paper, a new framework is proposed for accurate curvature estimation in discrete surfaces. The proposed framework is based on a local directional curve sampling of the surface where the sampling frequency can be controlled. This local model has a large number of degrees of freedoms compared with known techniques and, so, can better represent the local geometry. The proposed framework is quantitatively evaluated and compared with common techniques for surface curvature estimation. In order to perform an unbiased evaluation in which smoothing effects are factored out, we use a set of randomly generated Bezier surface patches for which the curvature values can be analytically computed. It is demonstrated that, through the establishment of sampling conditions, the error in estimations obtained by the proposed framework is smaller and that the proposed framework is less sensitive to low sampling density, sampling irregularities, and sampling noise.
IEEE Trans. Vis. Comput. Graph.	Knowledge Precepts for Design and Evaluation of Information Visualizations.	Robert A. Amar,John T. Stasko	2005	"The design and evaluation of most current information visualization systems descend from an emphasis on a user's ability to ""unpack¿ the representations of data of interest and operate on them independently. Too often, successful decision-making and analysis are more a matter of serendipity and user experience than of intentional design and specific support for such tasks; although humans have considerable abilities in analyzing relationships from data, the utility of visualizations remains relatively variable across users, data sets, and domains. In this paper, we discuss the notion of analytic gaps, which represent obstacles faced by visualizations in facilitating higher-level analytic tasks, such as decision-making and learning. We discuss support for bridging these gaps, propose a framework for the design and evaluation of information visualization systems, and demonstrate its use."
IEEE Trans. Vis. Comput. Graph.	Sharpen&Bend: Recovering Curved Sharp Edges in Triangle Meshes Produced by Feature-Insensitive Sampling.	Marco Attene,Bianca Falcidieno,Jarek Rossignac,Michela Spagnuolo	2005	Sharpen&Bend: Recovering Curved Sharp Edges in Triangle Meshes Produced by Feature-Insensitive Sampling.
IEEE Trans. Vis. Comput. Graph.	Signed Distance Computation Using the Angle Weighted Pseudonormal.	J. Andreas Baerentzen,Henrik Aanaes	2005	The normals of closed, smooth surfaces have long been used to determine whether a point is inside or outside such a surface. It is tempting to also use this method for polyhedra represented as triangle meshes. Unfortunately, this is not possible since, at the vertices and edges of a triangle mesh, the surface is not C^1 continuous, hence, the normal is undefined at these loci. In this paper, we undertake to show that the angle weighted pseudonormal (originally proposed by Thürmer and Wüthrich and independently by Séquin) has the important property that it allows us to discriminate between points that are inside and points that are outside a mesh, regardless of whether a mesh vertex, edge, or face is the closest feature. This inside-outside information is usually represented as the sign in the signed distance to the mesh. In effect, our result shows that this sign can be computed as an integral part of the distance computation. Moreover, it provides an additional argument in favor of the angle weighted pseudonormals being the natural extension of the face normals. Apart from the theoretical results, we also propose a simple and efficient algorithm for computing the signed distance to a closed C^0 mesh. Experiments indicate that the sign computation overhead when running this algorithm is almost negligible.
IEEE Trans. Vis. Comput. Graph.	"Editor's Note."	David S. Ebert	2005	"Editor's Note."
IEEE Trans. Vis. Comput. Graph.	Clifford Fourier Transform on Vector Fields.	Julia Ebling,Gerik Scheuermann	2005	Image processing and computer vision have robust methods for feature extraction and the computation of derivatives of scalar fields. Furthermore, interpolation and the effects of applying a filter can be analyzed in detail and can be advantages when applying these methods to vector fields to obtain a solid theoretical basis for feature extraction. We recently introduced the Clifford convolution, which is an extension of the classical convolution on scalar fields and provides a unified notation for the convolution of scalar and vector fields. It has attractive geometric properties that allow pattern matching on vector fields. In image processing, the convolution and the Fourier transform operators are closely related by the convolution theorem and, in this paper, we extend the Fourier transform to include general elements of Clifford Algebra, called multivectors, including scalars and vectors. The resulting convolution and derivative theorems are extensions of those for convolution and the Fourier transform on scalar fields. The Clifford Fourier transform allows a frequency analysis of vector fields and the behavior of vector-valued filters. In frequency space, vectors are transformed into general multivectors of the Clifford Algebra. Many basic vector-valued patterns, such as source, sink, saddle points, and potential vortices, can be described by a few multivectors in frequency space.
IEEE Trans. Vis. Comput. Graph.	In Memoriam: Jane Wilhelms.	Brian A. Barsky	2005	In Memoriam: Jane Wilhelms.
IEEE Trans. Vis. Comput. Graph.	A Practical Approach to Spectral Volume Rendering.	Steven Bergner,Torsten Möller,Melanie Tory,Mark S. Drew	2005	"To make a spectral representation of color practicable for volume rendering, a new low-dimensional subspace method is used to act as the carrier of spectral information. With that model, spectral light material interaction can be integrated into existing volume rendering methods at almost no penalty. In addition, slow rendering methods can profit from the new technique of postillumination¿generating spectral images in real-time for arbitrary light spectra under a fixed viewpoint. Thus, the capability of spectral rendering to create distinct impressions of a scene under different lighting conditions is established as a method of real-time interaction. Although we use an achromatic opacity in our rendering, we show how spectral rendering permits different data set features to be emphasized or hidden as long as they have not been entirely obscured. The use of postillumination is an order of magnitude faster than changing the transfer function and repeating the projection step. To put the user in control of the spectral visualization, we devise a new widget, a ""light-dial,¿ for interactively changing the illumination and include a usability study of this new light space exploration tool. Applied to spectral transfer functions, different lights bring out or hide specific qualities of the data. In conjunction with postillumination, this provides a new means for preparing data for visualization and forms a new degree of freedom for guided exploration of volumetric data sets."
IEEE Trans. Vis. Comput. Graph.	Boolean Operations with Implicit and Parametric Representation of Primitives Using R-Functions.	Yohan D. Fougerolle,Andrei V. Gribok,Sebti Foufou,Frédéric Truchetet,Mongi A. Abidi	2005	"We present a new and efficient algorithm to accurately polygonize an implicit surface generated by multiple Boolean operations with globally deformed primitives. Our algorithm is special in the sense that it can be applied to objects with both an implicit and a parametric representation, such as superquadrics, supershapes, and Dupin cyclides. The input is a Constructive Solid Geometry tree (CSG tree) that contains the Boolean operations, the parameters of the primitives, and the global deformations. At each node of the CSG tree, the implicit formulations of the subtrees are used to quickly determine the parts to be transmitted to the parent node, while the primitives' parametric definition are used to refine an intermediary mesh around the intersection curves. The output is both an implicit equation and a mesh representing its solution. For the resulting object, an implicit equation with guaranteed differential properties is obtained by simple combinations of the primitives' implicit equations using R-functions. Depending on the chosen R--function, this equation is continuous and can be differentiable everywhere. The primitives' parametric representations are used to directly polygonize the resulting surface by generating vertices that belong exactly to the zero-set of the resulting implicit equation. The proposed approach has many potential applications, ranging from mechanical engineering to shape recognition and data compression. Examples of complex objects are presented and commented on to show the potential of our approach for shape modeling."
IEEE Trans. Vis. Comput. Graph.	Reflectance from Images: A Model-Based Approach for Human Faces.	Martin Fuchs,Volker Blanz,Hendrik P. A. Lensch,Hans-Peter Seidel	2005	"In this paper, we present an image-based framework that acquires the reflectance properties of a human face. A range scan of the face is not required. Based on a morphable face model, the system estimates the 3D shape and establishes point-to-point correspondence across images taken from different viewpoints and across different individuals' faces. This provides a common parameterization of all reconstructed surfaces that can be used to compare and transfer BRDF data between different faces. Shape estimation from images compensates deformations of the face during the measurement process, such as facial expressions. In the common parameterization, regions of homogeneous materials on the face surface can be defined a priori. We apply analytical BRDF models to express the reflectance properties of each region and we estimate their parameters in a least-squares fit from the image data. For each of the surface points, the diffuse component of the BRDF is locally refined, which provides high detail. We present results for multiple analytical BRDF models, rendered at novel orientations and lighting conditions."
IEEE Trans. Vis. Comput. Graph.	An Infrastructure for Realizing Custom-Tailored Augmented Reality User Interfaces.	Wolfgang Broll,Irma Lindt,Jan Ohlenburg,Iris Herbst,Michael Wittkämper,Thomas Novotny	2005	Augmented Reality (AR) technologies are rapidly expanding into new application areas. However, the development of AR user interfaces and appropriate interaction techniques remains a complex and time-consuming task. Starting from scratch is more common than building upon existing solutions. Furthermore, adaptation is difficult, often resulting in poor quality and limited flexibility with regard to user requirements. In order to overcome these problems, we introduce an infrastructure for supporting the development of specific AR interaction techniques and their adaptation to individual user needs. Our approach is threefold: a flexible AR framework providing independence from particular input devices and rendering platforms, an interaction prototyping mechanism allowing for fast prototyping of new interaction techniques, and a high-level user interface description, extending user interface descriptions into the domain of AR. The general usability and applicability of the approach is demonstrated by means of three example AR projects.
IEEE Trans. Vis. Comput. Graph.	Warp Sculpting.	James E. Gain,Patrick Marais	2005	"The task of computer-based free-form shape design is fraught with practical and conceptual difficulties. Incorporating elements of traditional clay sculpting has long been recognized as a means of shielding the user from these complexities. We present warp sculpting, a variant of spatial deformation, which allows deformations to be initiated by the rigid body transformation or uniform scaling of volumetric tools. This is reminiscent of a tool imprinting, flexing, and molding clay. Unlike previous approaches, the deformation is truly interactive. Tools, encoded in a distance field, can have arbitrarily complex shapes. Although individual tools have a static shape, several tools can be applied simultaneously. We enhance the basic formulation of warp sculpting in two ways. First, deformation is toggled to automatically overcome the problem of ""sticky¿ tools, where the object's surface clings to parts of a tool that are moving away. Second, unlike many other spatial deformations, we ensure that warp sculpting remains foldover-free and, hence, prevent self-intersecting objects."
IEEE Trans. Vis. Comput. Graph.	Camera-Based Calibration Techniques for Seamless Multiprojector Displays.	Michael S. Brown,Aditi Majumder,Ruigang Yang	2005	"Multiprojector, large-scale displays are used in scientific visualization, virtual reality, and other visually intensive applications. In recent years, a number of camera-based computer vision techniques have been proposed to register the geometry and color of tiled projection-based display. These automated techniques use cameras to ""calibrate¿ display geometry and photometry, computing per-projector corrective warps and intensity corrections that are necessary to produce seamless imagery across projector mosaics. These techniques replace the traditional labor-intensive manual alignment and maintenance steps, making such displays cost-effective, flexible, and accessible. In this paper, we present a survey of different camera-based geometric and photometric registration techniques reported in the literature to date. We discuss several techniques that have been proposed and demonstrated, each addressing particular display configurations and modes of operation. We overview each of these approaches and discuss their advantages and disadvantages. We examine techniques that address registration on both planar (video walls) and arbitrary display surfaces and photometric correction for different kinds of display surfaces. We conclude with a discussion of the remaining challenges and research opportunities for multiprojector displays."
IEEE Trans. Vis. Comput. Graph.	Topological Fisheye Views for Visualizing Large Graphs.	Emden R. Gansner,Yehuda Koren,Stephen C. North	2005	Topological Fisheye Views for Visualizing Large Graphs.
IEEE Trans. Vis. Comput. Graph.	Guest Editorial: Special Issue on Haptics, Virtual, and Augmented Reality.	Grigore C. Burdea,Ming C. Lin,William Ribarsky,Benjamin Watson	2005	Guest Editorial: Special Issue on Haptics, Virtual, and Augmented Reality
IEEE Trans. Vis. Comput. Graph.	Hardware-Assisted Visibility Sorting for Unstructured Volume Rendering.	Steven P. Callahan,Milan Ikits,João Luiz Dihl Comba,Cláudio T. Silva	2005	Harvesting the power of modern graphics hardware to solve the complex problem of real-time rendering of large unstructured meshes is a major research goal in the volume visualization community. While, for regular grids, texture-based techniques are well-suited for current GPUs, the steps necessary for rendering unstructured meshes are not so easily mapped to current hardware. We propose a novel volume rendering technique that simplifies the CPU-based processing and shifts much of the sorting burden to the GPU, where it can be performed more efficiently. Our hardware-assisted visibility sorting algorithm is a hybrid technique that operates in both object-space and image-space. In object-space, the algorithm performs a partial sort of the 3D primitives in preparation for rasterization. The goal of the partial sort is to create a list of primitives that generate fragments in nearly sorted order. In image-space, the fragment stream is incrementally sorted using a fixed-depth sorting network. In our algorithm, the object-space work is performed by the CPU and the fragment-level sorting is done completely on the GPU. A prototype implementation of the algorithm demonstrates that the fragment-level sorting achieves rendering rates of between one and six million tetrahedral cells per second on an ATI Radeon 9800.
IEEE Trans. Vis. Comput. Graph.	From the Lab to the Field: Steve Roth - A Memoriam.		2005	From the Lab to the Field: Steve Roth - A Memoriam.
IEEE Trans. Vis. Comput. Graph.	Spatial Domain Wavelet Design for Feature Preservation in Computational Data Sets.	Gheorghe Craciun,Ming Jiang,David Thompson,Raghu Machiraju	2005	High-fidelity wavelet transforms can facilitate visualization and analysis of large scientific data sets. However, it is important that salient characteristics of the original features be preserved under the transformation. We present a set of filter design axioms in the spatial domain which ensure that certain feature characteristics are preserved from scale to scale and that the resulting filters correspond to wavelet transforms admitting in-place implementation. We demonstrate how the axioms can be used to design linear feature-preserving filters that are optimal in the sense that they are closest in L^2 to the ideal low pass filter. We are particularly interested in linear wavelet transforms for large data sets generated by computational fluid dynamics simulations. Our effort is different from classical filter design approaches which focus solely on performance in the frequency domain. Results are included that demonstrate the feature-preservation characteristics of our filters.
IEEE Trans. Vis. Comput. Graph.	vLOD: High-Fidelity Walkthrough of Large Virtual Environments.	Jatin Chhugani,Budirijanto Purnomo,Shankar Krishnan,Jonathan D. Cohen,Suresh Venkatasubramanian,David S. Johnson,Subodh Kumar	2005	"We present visibility computation and data organization algorithms that enable high-fidelity walkthroughs of large 3D geometric data sets. A novel feature of our walkthrough system is that it performs work proportional only to the required detail in visible geometry at the rendering time. To accomplish this, we use a precomputation phase that efficiently generates per cell vLOD: the geometry visible from a view-region at the right level of detail. We encode changes between neighboring cells' vLODs, which are not required to be memory resident. At the rendering time, we incrementally construct the vLOD for the current view-cell and render it. We have a small CPU and memory requirement for rendering and are able to display models with tens of millions of polygons at interactive frame rates with less than one pixel screen-space deviation and accurate visibility."
IEEE Trans. Vis. Comput. Graph.	A Statistical Wisp Model and Pseudophysical Approaches for InteractiveHairstyle Generation.	Byoungwon Choe,Hyeong-Seok Ko	2005	A Statistical Wisp Model and Pseudophysical Approaches for InteractiveHairstyle Generation.
IEEE Trans. Vis. Comput. Graph.	Modal Warping: Real-Time Simulation of Large Rotational Deformation and Manipulation.	Min Gyu Choi,Hyeong-Seok Ko	2005	"This paper proposes a real-time simulation technique for large deformations. Green's nonlinear strain tensor accurately models large deformations; however, time stepping of the resulting nonlinear system can be computationally expensive. Modal analysis based on a linear strain tensor has been shown to be suitable for real-time simulation, but is accurate only for moderately small deformations. In the present work, we identify the rotational component of an infinitesimal deformation and extend traditional linear modal analysis to track that component. We then develop a procedure to integrate the small rotations occurring at the nodal points. An interesting feature of our formulation is that it can implement both position and orientation constraints in a straightforward manner. These constraints can be used to interactively manipulate the shape of a deformable solid by dragging/twisting a set of nodes. Experiments show that the proposed technique runs in real-time, even for a complex model, and that it can simulate large bending and/or twisting deformations with acceptable realism."
IEEE Trans. Vis. Comput. Graph.	"Guest Editors' Introduction: Special Section on IEEE Visualization Applications."	Eduard Gröller,Kwan-Liu Ma,Klaus Mueller	2005	"Guest Editors' Introduction: Special Section on IEEE Visualization Applications."
IEEE Trans. Vis. Comput. Graph.	Visualization of Geologic Stress Perturbations Using Mohr Diagrams.	Patricia Crossno,David H. Rogers,Rebecca M. Brannon,David Coblentz,Joanne T. Fredrich	2005	Huge salt formations, trapping large untapped oil and gas reservoirs, lie in the deepwater region of the Gulf of Mexico. Drilling in this region is high-risk and drilling failures have led to well abandonments, with each costing tens of millions of dollars. Salt tectonics plays a central role in these failures. To explore the geomechanical interactions between salt and the surrounding sand and shale formations, scientists have simulated the stresses in and around salt diapirs in the Gulf of Mexico using nonlinear finite element geomechanical modeling. In this paper, we describe novel techniques developed to visualize the simulated subsurface stress field. We present an adaptation of the Mohr diagram, a traditional paper-and-pencil graphical method long used by the material mechanics community for estimating coordinate transformations for stress tensors, as a new tensor glyph for dynamically exploring tensor variables within three-dimensional finite element models. This interactive glyph can be used as either a probe or a filter through brushing and linking.
IEEE Trans. Vis. Comput. Graph.	Stroke Surfaces: Temporally Coherent Artistic Animations from Video.	John P. Collomosse,David Rowntree,Peter M. Hall	2005	"The contribution of this paper is a novel framework for synthesizing nonphotorealistic animations from real video sequences. We demonstrate that, through automated mid-level analysis of the video sequence as a spatiotemporal volume¿a block of frames with time as the third dimension¿we are able to generate animations in a wide variety of artistic styles, exhibiting a uniquely high degree of temporal coherence. In addition to rotoscoping, matting, and novel temporal effects unique to our method, we demonstrate the extension of static nonphotorealistic rendering (NPR) styles to video, including painterly, sketchy, and cartoon shading. We demonstrate how this novel coherent shading framework may be combined with our earlier motion emphasis work to produce a comprehensive ""Video Paintbox¿ capable of rendering complete cartoon-styled animations from video clips."
IEEE Trans. Vis. Comput. Graph.	Non-Euclidean Spring Embedders.	Stephen G. Kobourov,Kevin Wampler	2005	We present a conceptually simple approach to generalizing force-directed methods for graph layout from Euclidean geometry to Riemannian geometries. Unlike previous work on non-Euclidean force-directed methods, ours is not limited to special classes of graphs, but can be applied to arbitrary graphs. The method relies on extending the Euclidean notions of distance, angle, and force-interactions to smooth non-Euclidean geometries via projections to and from appropriately chosen tangent spaces. In particular, we formally describe the calculations needed to extend such algorithms to hyperbolic and spherical geometries. We also study the theoretical and practical considerations that arise when working with non-Euclidean geometries.
IEEE Trans. Vis. Comput. Graph.	Radiance Caching for Efficient Global Illumination Computation.	Jaroslav Krivánek,Pascal Gautron,Sumanta N. Pattanaik,Kadi Bouatouch	2005	In this paper, we present a ray tracing-based method for accelerated global illumination computation in scenes with low-frequency glossy BRDFs. The method is based on sparse sampling, caching, and interpolating radiance on glossy surfaces. In particular, we extend the irradiance caching scheme proposed by Ward et al. [1] to cache and interpolate directional incoming radiance instead of irradiance. The incoming radiance at a point is represented by a vector of coefficients with respect to a hemispherical or spherical basis. The surfaces suitable for interpolation are selected automatically according to the roughness of their BRDF. We also propose a novel method for computing translational radiance gradient at a point.
IEEE Trans. Vis. Comput. Graph.	A Particle System for Interactive Visualization of 3D Flows.	Jens Krüger,Peter Kipfer,Polina Kondratieva,Rüdiger Westermann	2005	We present a particle system for interactive visualization of steady 3D flow fields on uniform grids. For the amount of particles we target, particle integration needs to be accelerated and the transfer of these sets for rendering must be avoided. To fulfill these requirements, we exploit features of recent graphics accelerators to advect particles in the graphics processing unit (GPU), saving particle positions in graphics memory, and then sending these positions through the GPU again to obtain images in the frame buffer. This approach allows for interactive streaming and rendering of millions of particles and it enables virtual exploration of high resolution fields in a way similar to real-world experiments. The ability to display the dynamics of large particle sets using visualization options like shaded points or oriented texture splats provides an effective means for visual flow analysis that is far beyond existing solutions. For each particle, flow quantities like vorticity magnitude and \lambda_2 are computed and displayed. Built upon a previously published GPU implementation of a sorting network, visibility sorting of transparent particles is implemented. To provide additional visual cues, the GPU constructs and displays visualization geometry like particle lines and stream ribbons.
IEEE Trans. Vis. Comput. Graph.	Comparing 2D Vector Field Visualization Methods: A User Study.	David H. Laidlaw,Robert M. Kirby,Cullen D. Jackson,J. Scott Davidson,Timothy S. Miller,Marco da Silva,William H. Warren,Michael J. Tarr	2005	We present results from a user study that compared six visualization methods for two-dimensional vector data. Users performed three simple but representative tasks using visualizations from each method: 1) locating all critical points in an image, 2) identifying critical point types, and 3) advecting a particle. Visualization methods included two that used different spatial distributions of short arrow icons, two that used different distributions of integral curves, one that used wedges located to suggest flow lines, and line-integral convolution (LIC). Results show different strengths and weaknesses for each method. We found that users performed these tasks better with methods that: 1) showed the sign of vectors within the vector field, 2) visually represented integral curves, and 3) visually represented the locations of critical points. Expert user performance was not statistically different from nonexpert user performance. We used several methods to analyze the data including omnibus analysis of variance, pairwise t-tests, and graphical analysis using inferential confidence intervals. We concluded that using the inferential confidence intervals for displaying the overall pattern of results for each task measure and for performing subsequent pairwise comparisons of the condition means was the best method for analyzing the data in this study. These results provide quantitative support for some of the anecdotal evidence concerning visualization methods. The tasks and testing framework also provide a basis for comparing other visualization methods, for creating more effective methods and for defining additional tasks to further understand the tradeoffs among the methods. In the future, we also envision extending this work to more ambitious comparisons, such as evaluating two-dimensional vectors on two-dimensional surfaces embedded in three-dimensional space and defining analogous tasks for three-dimensional visualization methods.
IEEE Trans. Vis. Comput. Graph.	Data Streaming in Telepresence Environments.	Edouard Lamboray,Stephan Würmlin,Markus H. Gross	2005	In this paper, we discuss data transmission in telepresence environments for collaborative virtual reality applications. We analyze data streams in the context of networked virtual environments and classify them according to their traffic characteristics. Special emphasis is put on geometry-enhanced (3D) video. We review architectures for real-time 3D video pipelines and derive theoretical bounds on the minimal system latency as a function of the transmission and processing delays. Furthermore, we discuss bandwidth issues of differential update coding for 3D video. In our telepresence system¿the blue-c¿we use a point-based 3D video technology which allows for differentially encoded 3D representations of human users. While we discuss the considerations which lead to the design of our three-stage 3D video pipeline, we also elucidate some critical implementation details regarding decoupling of acquisition, processing and rendering frame rates, and audio/video synchronization. Finally, we demonstrate the communication and networking features of the blue-c system in its full deployment. We show how the system can possibly be controlled to face processing or networking bottlenecks by adapting the multiple system components like audio, application data, and 3D video.
IEEE Trans. Vis. Comput. Graph.	Interactive Transparency Rendering for Large CAD Models.	Jianbing Huang,Michael B. Carter	2005	Transparency is an important graphics effect that can be used to significantly increase the realism of the rendered scene or to enable more effective visual inspection in engineering visualization. In this paper, we propose achieving interactive transparency rendering of a static scene by sorting the triangles in back-to-front order on CPU and supplying the sorted triangles to the graphics pipeline for rendering on GPU hardware. Our sorting method sorts the triangles in object space and is built upon the Binary Space Partition (BSP) and depth-sort methods with its behavior readily tunable to exploit the strengths of both methods. We propose novel techniques to optimize the BSP construction process with respect to multiple factors including tree construction time, tree size, and expected sorting cost. We also propose an improved depth-sort algorithm that can produce correct depth order without triangle split when no cyclic occlusion exists. We demonstrate that the proposed system results in a penalty factor of 4~6 for various types of parts, among which the largest one has nearly 1.2 million triangles. In addition, the penalty factor may be further improved if sorting in CPU and rendering in GPU are executed in parallel. Two approximation strategies are also studied to test the practicality of our system against large CAD assemblies. Experimental results on an assembly containing over 16 million triangles distributed in about 10,000 transparent parts show that the proposed system still results in a penalty factor of 4~6 while producing few artifacts.
IEEE Trans. Vis. Comput. Graph.	Real-Time Optimal Adaptation for Planetary Geometry and Texture: 4-8 Tile Hierarchies.	Lok M. Hwa,Mark A. Duchaineau,Kenneth I. Joy	2005	The real-time display of huge geometry and imagery databases involves view-dependent approximations, typically through the use of precomputed hierarchies that are selectively refined at runtime. A classic motivating problem is terrain visualization in which planetary databases involving billions of elevation and color values are displayed on PC graphics hardware at high frame rates. This paper introduces a new diamond data structure for the basic selective-refinement processing, which is a streamlined method of representing the well-known hierarchies of right triangles that have enjoyed much success in real-time, view-dependent terrain display. Regular-grid tiles are proposed as the payload data per diamond for both geometry and texture. The use of 4-8 grid refinement and coarsening schemes allows level-of-detail transitions that are twice as gradual as traditional quadtree-based hierarchies, as well as very high-quality low-pass filtering compared to subsampling-based hierarchies. An out-of-core storage organization is introduced based on Sierpinski indices per diamond, along with a tile preprocessing framework based on fine-to-coarse, same-level, and coarse-to-fine gathering operations. To attain optimal frame-to-frame coherence and processing-order priorities, dual split and merge queues are developed similar to the Realtime Optimally Adapting Meshes (ROAM) Algorithm, as well as an adaptation of the ROAM frustum culling technique. Example applications of lake-detection and procedural terrain generation demonstrate the flexibility of the tile processing framework.
IEEE Trans. Vis. Comput. Graph.	Metamorphosis of 3D Polyhedral Models Using Progressive Connectivity Transformations.	Chao-Hung Lin,Tong-Yee Lee	2005	"Three-dimensional metamorphosis is a powerful technique to produce a 3D shape transformation between two or more existing models. In this paper, we propose a novel 3D morphing technique that avoids creating a merged embedding that contains the faces, edges, and vertices of two given embeddings. This novel 3D morphing technique dynamically adds or removes vertices to gradually transform the connectivity of 3D polyhedrons from a source model into a target model and simultaneously creates the intermediate shapes. In addition, a priority control function provides the animators with control of arising or dissolving of input models' features in a morphing sequence. This is a useful tool to control a morphing sequence more easily and flexibly. Several examples of aesthetically pleasing morphs are demonstrated using the proposed method."
IEEE Trans. Vis. Comput. Graph.	A Method to Generate Soft Shadows Using a Layered Depth Image and Warping.	Yeon-Ho Im,Chang-Young Han,Lee-Sup Kim	2005	"We present an image-based method for propagating area light illumination through a Layered Depth Image (LDI) to generate soft shadows from opaque and nonrefractive transparent objects. In our approach, using the depth peeling technique, we render an LDI from a reference light sample on a planar light source. Light illumination of all pixels in an LDI is then determined for all the other sample points via warping, an image-based rendering technique, which approximates ray tracing in our method. We use an image-warping equation and McMillan's warp ordering algorithm to find the intersections between rays and polygons and to find the order of intersections. Experiments for opaque and nonrefractive transparent objects are presented. Results indicate our approach generates soft shadows fast and effectively. Advantages and disadvantages of the proposed method are also discussed."
IEEE Trans. Vis. Comput. Graph.	Accelerated Unsteady Flow Line Integral Convolution.	Zhanping Liu,Robert J. Moorhead	2005	Unsteady flow line integral convolution (UFLIC) is a texture synthesis technique for visualizing unsteady flows with high temporal-spatial coherence. Unfortunately, UFLIC requires considerable time to generate each frame due to the huge amount of pathline integration that is computed for particle value scattering. This paper presents Accelerated UFLIC (AUFLIC) for near interactive (1 frame/second) visualization with 160,000 particles per frame. AUFLIC reuses pathlines in the value scattering process to reduce computationally expensive pathline integration. A flow-driven seeding strategy is employed to distribute seeds such that only a few of them need pathline integration while most seeds are placed along the pathlines advected at earlier times by other seeds upstream and, therefore, the known pathlines can be reused for fast value scattering. To maintain a dense scattering coverage to convey high temporal-spatial coherence while keeping the expense of pathline integration low, a dynamic seeding controller is designed to decide whether to advect, copy, or reuse a pathline. At a negligible memory cost, AUFLIC is 9 times faster than UFLIC with comparable image quality.
IEEE Trans. Vis. Comput. Graph.	Six Degree-of-Freedom Haptic Rendering Using Spatialized Normal Cone Search.	David E. Johnson,Peter Willemsen,Elaine Cohen	2005	This paper describes a haptic rendering algorithm for arbitrary polygonal models using a six degree-of-freedom haptic interface. The algorithm supports activities such as virtual prototyping of complex polygonal models and adding haptic interaction to virtual environments. The underlying collision system computes local extrema in distance between the model controlled by the haptic device and the rest of the scene. The haptic rendering computes forces and torques on the moving model based on these local extrema. The system is demonstrated on models with tens of thousands of triangles and developed in an accessibility application for finding collision-free paths.
IEEE Trans. Vis. Comput. Graph.	Dynamic Interactions in Physically Realistic Collaborative Virtual Environments.	Pieter Jorissen,Maarten Wijnants,Wim Lamotte	2005	This work describes our efforts in creating a general object interaction framework for dynamic collaborative virtual environments. Furthermore, we increase the realism of the interactive world by using a rigid body simulator to calculate all actor and object movements. The main idea behind our interactive platform is to construct a virtual world using only objects that contain their own interaction information. As a result, the object interactions are application independent and only a single scheme is required to handle all interactions in the virtual world. In order to have more dynamic interactions, we also created a new and efficient way for human users to dynamically interact within virtual worlds through their avatar. In particular, we show how inverse kinematics can be used to increase the interaction possibilities and realism in collaborative virtual environments. This results in a higher feeling of presence for connected users and allows for easy, on-the-fly creation of new interactions. For the distribution of both the interactive objects and the dynamic avatar interactions, we keep the network load as low as possible. To demonstrate the effectiveness of our techniques, we incorporate them into an existing CVE framework.
IEEE Trans. Vis. Comput. Graph.	A New Technique for Rendering Complex Portals.	Nick Lowe,Amitava Datta	2005	A New Technique for Rendering Complex Portals.
IEEE Trans. Vis. Comput. Graph.	A Parallel Coordinates Style Interface for Exploratory Volume Visualization.	Melanie Tory,Simeon Potts,Amitava Datta,Torsten Möller	2005	A Parallel Coordinates Style Interface for Exploratory Volume Visualization.
IEEE Trans. Vis. Comput. Graph.	Reconstruction and Visualization of Planetary Nebulae.	Marcus A. Magnor,Gordon L. Kindlmann,Charles D. Hansen,Neb Duric	2005	"From our terrestrially confined viewpoint, the actual three-dimensional shape of distant astronomical objects is, in general, very challenging to determine. For one class of astronomical objects, however, spatial structure can be recovered from conventional 2D images alone. So-called planetary nebulae (PNe) exhibit pronounced symmetry characteristics that come about due to fundamental physical processes. Making use of this symmetry constraint, we present a technique to automatically recover the axisymmetric structure of many planetary nebulae from photographs. With GPU-based volume rendering driving a nonlinear optimization, we estimate the nebula's local emission density as a function of its radial and axial coordinates and we recover the orientation of the nebula relative to Earth. The optimization refines the nebula model and its orientation by minimizing the differences between the rendered image and the original astronomical image. The resulting model allows creating realistic 3D visualizations of these nebulae, for example, for planetarium shows and other educational purposes. In addition, the recovered spatial distribution of the emissive gas can help astrophysicists gain deeper insight into the formation processes of planetary nebulae."
IEEE Trans. Vis. Comput. Graph.	Creating Speech-Synchronized Animation.	Scott A. King,Richard E. Parent	2005	We present a facial model designed primarily to support animated speech. Our facial model takes facial geometry as input and transforms it into a parametric deformable model. The facial model uses a muscle-based parameterization, allowing for easier integration between speech synchrony and facial expressions. Our facial model has a highly deformable lip model that is grafted onto the input facial geometry to provide the necessary geometric complexity needed for creating lip shapes and high-quality renderings. Our facial model also includes a highly deformable tongue model that can represent the shapes the tongue undergoes during speech. We add teeth, gums, and upper palate geometry to complete the inner mouth. To decrease the processing time, we hierarchically deform the facial surface. We also present a method to animate the facial model over time to create animated speech using a model of coarticulation that blends visemes together using dominance functions. We treat visemes as a dynamic shaping of the vocal tract by describing visemes as curves instead of keyframes. We show the utility of the techniques described in this paper by implementing them in a text-to-audiovisual-speech system that creates animation of speech from unrestricted text. The facial and coarticulation models must first be interactively initialized. The system then automatically creates accurate real-time animated speech from the input text. It is capable of cheaply producing tremendous amounts of animated speech with very low resource requirements.
IEEE Trans. Vis. Comput. Graph.	Tetrahedralization of Point Sets Using Expanding Spheres.	Martin S. Maltz	2005	The expanding sphere algorithm computes an alpha shape tetrahedralization of a point set. Starting with a seed tetrahedron, the circumscribing sphere is squeezed through each face until it either touches another point or exceeds a preset radius. If no point is found, that face of the tetrahedron is part of the surface of an object. If a point is found, a new tetrahedron is constructed. This process is iterated until all the faces of the tetrahedra have been processed and no more connected points can be found. If there are points left over, the process is iterated, creating additional objects. The algorithm generates a list of objects, with an alpha shape tetrahedralization and a surface triangulation for each. Any points that cannot be made part of a valid tetrahedron are also returned in the extra points list. The algorithm is efficient for uniformly distributed point sets, with a running time that is linear in the number of points for such sets. Since the operations are local, it is also robust.
IEEE Trans. Vis. Comput. Graph.	Correcting Interperspective Aliasing in Autostereoscopic Displays.	Christian N. Moller,Adrian R. L. Travis	2005	An image presented on an autostereoscopic system should not contain discontinuities between adjacent views. A viewer should experience a continuous scene when moving from one view to the next. If corresponding points in two perspectives do not spatially abut, a viewer will experience jumps in the scene. This is known as interperspective aliasing. Interperspective aliasing is caused by object features far away from the stereoscopic screen being too small, which results in visual artifacts. By modeling a 3D point as a defocused image point, we can adapt Fourier analysis to devise a depth-dependent filter kernel that allows filtering of a stereoscopic 3D image. For synthetic 3D data, we use a simpler approach, which is to smear the data by a distance proportional to its depth.
IEEE Trans. Vis. Comput. Graph.	Advanced Virtual Endoscopic Pituitary Surgery.	André Neubauer,Stefan Wolfsberger,Marie-Thérèse Forster,Lukas Mroz,Rainer Wegenkittl,Katja Bühler	2005	Endoscopy has recently been introduced to endonasal transsphenoidal pituitary surgery as a minimally invasive procedure for the removal of various kinds of pituitary tumors. To reduce morbidity and mortality with this new technique, the surgeon must be well-trained and well-prepared. Virtual endoscopy can be beneficial as a tool for training, preoperative planning, and intraoperative support. This paper introduces STEPS, a virtual endoscopy system designed to aid surgeons in getting acquainted with the endoscopic view of the anatomy, the handling of instruments, the transsphenoidal approach, and challenges associated with the procedure. STEPS also assists experienced surgeons in planning a real endoscopic intervention by getting familiar with the individual patient anatomy, identifying landmarks, planning the approach, and deciding upon the ideal target position of the actual surgical activity. The application provides interactive visualization, navigation, and perception aids and the possibility of simulating the procedure, including haptic feedback and simulation of surgical instruments.
IEEE Trans. Vis. Comput. Graph.	Real-Time 3D Human Capture System for Mixed-Reality Art and Entertainment.	Ta Huynh Duy Nguyen,Tran Cong Thien Qui,Ke Xu,Adrian David Cheok,Sze Lee Teo,Zhiying Zhou,Asitha Mallawaarachchi,Shang Ping Lee,Wei Liu,Hui Siang Teo,Le Nam Thang,Yu Li,Hirokazu Kato	2005	A real-time system for capturing humans in 3D and placing them into a mixed reality environment is presented in this paper. The subject is captured by nine cameras surrounding her. Looking through a head-mounted-display with a camera in front pointing at a marker, the user can see the 3D image of this subject overlaid onto a mixed reality scene. The 3D images of the subject viewed from this viewpoint are constructed using a robust and fast shape-from-silhouette algorithm. The paper also presents several techniques to produce good quality and speed up the whole system. The frame rate of our system is around 25 fps using only standard Intel processor-based personal computers. Besides a remote live 3D conferencing and collaborating system, we also describe an application of the system in art and entertainment, named Magic Land, which is a mixed reality environment where captured avatars of human and 3D computer generated virtual animations can form an interactive story and play with each other. This system demonstrates many technologies in human computer interaction: mixed reality, tangible interaction, and 3D communication. The result of the user study not only emphasizes the benefits, but also addresses some issues of these technologies.
IEEE Trans. Vis. Comput. Graph.	"Achieving Dialogue with Children with Severe Autism in an Adaptive Multisensory Interaction: The 'MEDIATE' Project."	Narcís Parés,Paul Masri,Gerard van Wolferen,Chris Creed	2005	This paper presents an adaptive physical environment that allows children with severe autism to successfully interact with multimodal stimuli, giving them a sense of control of the interaction and, hence, providing them with a sense of agency. This has been an extremely important effort for two main reasons: 1) This user group cannot be typified, hence making the design of an interactive system to fit all the spectrum of individuals a very complex task; 2) each individual PAS (Person on the Autistic Spectrum) user must be able to develop himself within the environment according to his own capacities and potentiality. Qualitative evaluation by psychologists shows very good results and sketches an encouraging future for research on these environments.
IEEE Trans. Vis. Comput. Graph.	Dynamic Range Reduction Inspired by Photoreceptor Physiology.	Erik Reinhard,Kate Devlin	2005	A common task in computer graphics is the mapping of digital high dynamic range images to low dynamic range display devices such as monitors and printers. This task is similar to the adaptation processes which occur in the human visual system. Physiological evidence suggests that adaptation already occurs in the photoreceptors, leading to a straightforward model that can be easily adapted for tone reproduction. The result is a fast and practical algorithm for general use with intuitive user parameters that control intensity, contrast, and level of chromatic adaptation, respectively.
IEEE Trans. Vis. Comput. Graph.	"Guest Editors' Introduction: Special Section on IEEE Visualization."	Holly E. Rushmeier,Jarke J. van Wijk,Greg Turk	2005	"Guest Editors' Introduction: Special Section on IEEE Visualization."
IEEE Trans. Vis. Comput. Graph.	An Insight-Based Methodology for Evaluating Bioinformatics Visualizations.	Purvi Saraiya,Chris North,Karen Duca	2005	High-throughput experiments, such as gene expression microarrays in the life sciences, result in very large data sets. In response, a wide variety of visualization tools have been created to facilitate data analysis. A primary purpose of these tools is to provide biologically relevant insight into the data. Typically, visualizations are evaluated in controlled studies that measure user performance on predetermined tasks or using heuristics and expert reviews. To evaluate and rank bioinformatics visualizations based on real-world data analysis scenarios, we developed a more relevant evaluation method that focuses on data insight. This paper presents several characteristics of insight that enabled us to recognize and quantify it in open-ended user tests. Using these characteristics, we evaluated five microarray visualization tools on the amount and types of insight they provide and the time it takes to acquire it. The results of the study guide biologists in selecting a visualization tool based on the type of their microarray data, visualization designers on the key role of user interaction techniques, and evaluators on a new approach for evaluating the effectiveness of visualizations for providing insight. Though we used the method to analyze bioinformatics visualizations, it can be applied to other domains.
IEEE Trans. Vis. Comput. Graph.	A Multiresolution Representation for Massive Meshes.	Eric Shaffer,Michael Garland	2005	We present a new external memory multiresolution surface representation for massive polygonal meshes. Previous methods for building such data structures have relied on resampled surface data or employed memory intensive construction algorithms that do not scale well. Our proposed representation combines efficient access to sampled surface data with access to the original surface. The construction algorithm for the surface representation exhibits memory requirements that are insensitive to the size of the input mesh, allowing it to process meshes containing hundreds of millions of polygons. The multiresolution nature of the surface representation has allowed us to develop efficient algorithms for view-dependent rendering, approximate collision detection, and adaptive simplification of massive meshes. The empirical performance of these algorithms demonstrates that the underlying data structure is a powerful and flexible tool for operating on massive geometric data.
IEEE Trans. Vis. Comput. Graph.	"Exploring Connectivity of the Brain's White Matter with Dynamic Queries."	Anthony Sherbondy,David Akers,Rachel Mackenzie,Robert Dougherty,Brian A. Wandell	2005	Diffusion Tensor Imaging (DTI) is a magnetic resonance imaging method that can be used to measure local information about the structure of white matter within the human brain. Combining DTI data with the computational methods of MR tractography, neuroscientists can estimate the locations and sizes of nerve bundles (white matter pathways) that course through the human brain. Neuroscientists have used visualization techniques to better understand tractography data, but they often struggle with the abundance and complexity of the pathways. In this paper, we describe a novel set of interaction techniques that make it easier to explore and interpret such pathways. Specifically, our application allows neuroscientists to place and interactively manipulate box or ellipsoid-shaped regions to selectively display pathways that pass through specific anatomical areas. These regions can be used in coordination with a simple and flexible query language which allows for arbitrary combinations of these queries using Boolean logic operators. A representation of the cortical surface is provided for specifying queries of pathways that may be relevant to gray matter structures and for displaying activation information obtained from functional magnetic resonance imaging. By precomputing the pathways and their statistical properties, we obtain the speed necessary for interactive question-and-answer sessions with brain researchers. We survey some questions that researchers have been asking about tractography data and show how our system can be used to answer these questions efficiently.
IEEE Trans. Vis. Comput. Graph.	Geometry-Aware Bases for Shape Approximation.	Olga Sorkine,Daniel Cohen-Or	2005	Geometry-Aware Bases for Shape Approximation.
IEEE Trans. Vis. Comput. Graph.	Barycentric Parameterizations for Isotropic BRDFs.	Michael M. Stark,James Arvo,Brian E. Smits	2005	"A bidirectional reflectance distribution function (BRDF) is often expressed as a function of four real variables: two spherical coordinates in each of the the ""incoming¿ and ""outgoing¿ directions. However, many BRDFs reduce to functions of fewer variables. For example, isotropic reflection can be represented by a function of three variables. Some BRDF models can be reduced further. In this paper, we introduce new sets of coordinates which we use to reduce the dimensionality of several well-known analytic BRDFs as well as empirically measured BRDF data. The proposed coordinate systems are barycentric with respect to a triangular support with a direct physical interpretation. One coordinate set is based on the BRDF model proposed by Lafortune. Another set, based on a model of Ward, is associated with the ""halfway¿ vector common in analytical BRDF formulas. Through these coordinate sets we establish lower bounds on the approximation error inherent in the models on which they are based. We present a third set of coordinates, not based on any analytical model, that performs well in approximating measured data. Finally, our proposed variables suggest novel ways of constructing and visualizing BRDFs."
IEEE Trans. Vis. Comput. Graph.	Development of Anthropomorphic Multi-D.O.F. Master-Slave Arm for Mutual Telexistence.	Riichiro Tadakuma,Yoshiaki Asahara,Hiroyuki Kajimoto,Naoki Kawakami,Susumu Tachi	2005	"We developed a robotic arm for a master-slave system to support ""mutual telexistence,¿ which realizes remote dexterous manipulation tasks and close physical communication with other people using gestures. In this paper, we describe the specifications of the experimental setup of the master-slave arm to demonstrate the feasibility of the mutual telexistence concept. We developed the master arm of a telexistence robot for interpersonal communication. The last degree of the 7-degree-of-freedom slave arm is resolved by placing a small orientation sensor on the operator's arm. This master arm is made light and impedance control is applied in order to grant the operator as much freedom of movement as possible. For this development stage, we compared three control methods and confirmed that the impedance control method is the most appropriate to this system."
IEEE Trans. Vis. Comput. Graph.	TORNADO: Omnistereo Video Imaging with Rotating Optics.	Kenji Tanaka,Susumu Tachi	2005	One of the key techniques for vision-based communication is omnidirectional stereo (omnistereo) imaging, in which stereoscopic images for an arbitrary horizontal direction are captured and presented according to the viewing direction of the observer. Although omnistereo models have been surveyed in several studies, few omnistereo sensors have actually been implemented. In this paper, a practical method for capturing omnistereo video sequences using rotating optics is proposed and evaluated. The rotating optics system consists of prism sheets, circular or linear polarizing films, and a hyperboloidal mirror. This system has two different modes of operation with regard to the separation of images for the left and right eyes. In the high-speed shutter mode, images are separated using postimage processing, while, in the low-speed shutter mode, the image separation is completed by optics. By capturing actual images, we confirmed the effectiveness of the methods.
IEEE Trans. Vis. Comput. Graph.	Creating and Simulating Skeletal Muscle from the Visible Human Data Set.	Joseph Teran,Eftychios Sifakis,Silvia S. Blemker,Victor Ng-Thow-Hing,Cynthia Lau,Ronald Fedkiw	2005	Simulation of the musculoskeletal system has important applications in biomechanics, biomedical engineering, surgery simulation, and computer graphics. The accuracy of the muscle, bone, and tendon geometry as well as the accuracy of muscle and tendon dynamic deformation are of paramount importance in all these applications. We present a framework for extracting and simulating high resolution musculoskeletal geometry from the segmented visible human data set. We simulate 30 contact/collision coupled muscles in the upper limb and describe a computationally tractable implementation using an embedded mesh framework. Muscle geometry is embedded in a nonmanifold, connectivity preserving simulation mesh molded out of a lower resolution BCC lattice containing identical, well-shaped elements, leading to a relaxed time step restriction for stability and, thus, reduced computational cost. The muscles are endowed with a transversely isotropic, quasi-incompressible constitutive model that incorporates muscle fiber fields as well as passive and active components. The simulation takes advantage of a new robust finite element technique that handles both degenerate and inverted tetrahedra.
IEEE Trans. Vis. Comput. Graph.	Topological Methods for 2D Time-Dependent Vector Fields Based on Stream Lines and Path Lines.	Holger Theisel,Tino Weinkauf,Hans-Christian Hege,Hans-Peter Seidel	2005	This paper describes approaches to topologically segmenting 2D time-dependent vector fields. For this class of vector fields, two important classes of lines exist: stream lines and path lines. Because of this, two segmentations are possible: either concerning the behavior of stream lines or of path lines. While topological features based on stream lines are well established, we introduce path line oriented topology as a new visualization approach in this paper. As a contribution to stream line oriented topology, we introduce new methods to detect global bifurcations like saddle connections and cyclic fold bifurcations as well as a method of tracking all isolated closed stream lines. To get the path line oriented topology, we segment the vector field into areas of attracting, repelling, and saddle-like behavior of the path lines. We compare both kinds of topologies and apply them to a number of test data sets.
IEEE Trans. Vis. Comput. Graph.	An Intelligent System Approach to Higher-Dimensional Classification of Volume Data.	Fan-Yin Tzeng,Eric B. Lum,Kwan-Liu Ma	2005	In volume data visualization, the classification step is used to determine voxel visibility and is usually carried out through the interactive editing of a transfer function that defines a mapping between voxel value and color/opacity. This approach is limited by the difficulties in working effectively in the transfer function space beyond two dimensions. We present a new approach to the volume classification problem which couples machine learning and a painting metaphor to allow more sophisticated classification in an intuitive manner. The user works in the volume data space by directly painting on sample slices of the volume and the painted voxels are used in an iterative training process. The trained system can then classify the entire volume. Both classification and rendering can be hardware accelerated, providing immediate visual feedback as painting progresses. Such an intelligent system approach enables the user to perform classification in a much higher dimensional space without explicitly specifying the mapping for every dimension used. Furthermore, the trained system for one data set may be reused to classify other data sets with similar characteristics.
IEEE Trans. Vis. Comput. Graph.	Importance-Driven Feature Enhancement in Volume Visualization.	Ivan Viola,Armin Kanitsar,M. Eduard Gröller	2005	This paper presents importance-driven feature enhancement as a technique for the automatic generation of cut-away and ghosted views out of volumetric data. The presented focus+context approach removes or suppresses less important parts of a scene to reveal more important underlying information. However, less important parts are fully visible in those regions, where important visual information is not lost, i.e., more relevant features are not occluded. Features within the volumetric data are first classified according to a new dimension, denoted as object importance. This property determines which structures should be readily discernible and which structures are less important. Next, for each feature, various representations (levels of sparseness) from a dense to a sparse depiction are defined. Levels of sparseness define a spectrum of optical properties or rendering styles. The resulting image is generated by ray-casting and combining the intersected features proportional to their importance (importance compositing). The paper includes an extended discussion on several possible schemes for levels of sparseness specification. Furthermore, different approaches to importance compositing are treated.
IEEE Trans. Vis. Comput. Graph.	Faster Isosurface Ray Tracing Using Implicit KD-Trees.	Ingo Wald,Heiko Friedrich,Gerd Marmitt,Philipp Slusallek,Hans-Peter Seidel	2005	The visualization of high-quality isosurfaces at interactive rates is an important tool in many simulation and visualization applications. Today, isosurfaces are most often visualized by extracting a polygonal approximation that is then rendered via graphics hardware or by using a special variant of preintegrated volume rendering. However, these approaches have a number of limitations in terms of the quality of the isosurface, lack of performance for complex data sets, or supported shading models. An alternative isosurface rendering method that does not suffer from these limitations is to directly ray trace the isosurface. However, this approach has been much too slow for interactive applications unless massively parallel shared-memory supercomputers have been used. In this paper, we implement interactive isosurface ray tracing on commodity desktop PCs by building on recent advances in real-time ray tracing of polygonal scenes and using those to improve isosurface ray tracing performance as well. The high performance and scalability of our approach will be demonstrated with several practical examples, including the visualization of highly complex isosurface data sets, the interactive rendering of hybrid polygonal/isosurface scenes, including high-quality ray traced shading effects, and even interactive global illumination on isosurfaces.
IEEE Trans. Vis. Comput. Graph.	Light Field Morphing Using 2D Features.	Lifeng Wang,Stephen Lin,Seungyong Lee,Baining Guo,Heung-Yeung Shum	2005	We present a 2D feature-based technique for morphing 3D objects represented by light fields. Existing light field morphing methods require the user to specify corresponding 3D feature elements to guide morph computation. Since slight errors in 3D specification can lead to significant morphing artifacts, we propose a scheme based on 2D feature elements that is less sensitive to imprecise marking of features. First, 2D features are specified by the user in a number of key views in the source and target light fields. Then the two light fields are warped view by view as guided by the corresponding 2D features. Finally, the two warped light fields are blended together to yield the desired light field morph. Two key issues in light field morphing are feature specification and warping of light field rays. For feature specification, we introduce a user interface for delineating 2D features in key views of a light field, which are automatically interpolated to other views. For ray warping, we describe a 2D technique that accounts for visibility changes and present a comparison to the ideal morphing of light fields. Light field morphing based on 2D features makes it simple to incorporate previous image morphing techniques such as nonuniform blending, as well as to morph between an image and a light field.
IEEE Trans. Vis. Comput. Graph.	Cutting on Triangle Mesh: Local Model-Based Haptic Display for Dental Preparation Surgery Simulation.	Daniel Wang,Yuru Zhang,Yuhui Wang,Yuan-Shin Lee,Pei-Jun Lü,Yong Wang	2005	A new method to realize stable and realistic cutting simulation using an impedance display haptic device and microcomputer is presented in this paper. Material removal or cutting simulation is a critical task in dental preparation surgery simulation. In this paper, a piecewise contact force model is proposed to approximately describe the cutting process. Challenging issues of minimizing the difference between the cutting simulation and haptic contact simulation are analyzed. The proposed contact-based simulation method is developed for a one-dimensional cutting task and can be expanded to three-dimensional cases. Local model-based multirate simulation cutting architecture is proposed and force control of the haptic device is decoupled from the cutting simulation loop, which can both ensure high fidelity of dynamical simulation as well as maintain stability of the haptic device. The cutting operation is realized using spherical and cylindrical shaped tools. An experiment based on the Phantom desktop proves that fidelity in one-dimensional cutting can be realized and stability in three-dimensional cutting can be ensured using the force-filtering method.
IEEE Trans. Vis. Comput. Graph.	"Guest Editors' Introduction: Special Section on InfoVis."	Matthew O. Ward,Tamara Munzner	2005	"Guest Editors' Introduction: Special Section on InfoVis."
IEEE Trans. Vis. Comput. Graph.	Uniform Remeshing with an Adaptive Domain: A New Scheme for View-Dependent Level-of-Detail Rendering of Meshes.	Yuanchen Zhu	2005	We present a new algorithm for view-dependent level-of-detail rendering of meshes. Not only can it effectively resolve complex geometry features similar to edge collapse-based schemes, but it also produces meshes that modern graphics hardware can render efficiently. This is accomplished through a novel hybrid approach: For each frame, we view-dependently refine the progressive mesh (PM) representation of the original mesh and use the output as the base domain of uniform regular refinements. The algorithm exploits frame-to-frame coherence and only updates portions of the output mesh corresponding to modified domain triangles. The PM representation is built using a custom volume preservation-based error function. A simple k-d tree enhanced jump-and-walk scheme is used to quickly map from the dynamic base domain to the original mesh during regular refinements. In practice, the PM refinement provides a view-optimized base domain for later regular refinements. The regular refinements ensure almost-everywhere regularity of output meshes, allowing optimization for vertex cache coherence and caching of geometry data in high-performance graphics memory. Combined, they also have the effect of allowing our algorithm to operate on uniform clusters of triangles instead of individual ones, reducing CPU workload.
IEEE Trans. Vis. Comput. Graph.	Rendering Deformable Surface Reflectance Fields.	Tim Weyrich,Hanspeter Pfister,Markus H. Gross	2005	Animation of photorealistic computer graphics models is an important goal for many applications. Image-based modeling has emerged as a promising approach to capture and visualize real-world objects. Animating image-based models, however, is still a largely unsolved problem. In this paper, we extend a popular image-based representation called surface reflectance field to animate and render deformable real-world objects under arbitrary illumination. Deforming the surface reflectance field is achieved by modifying the underlying impostor geometry. We augment the impostor by a local parameterization that allows the correct evaluation of acquired reflectance images, preserving the original light model on the deformed surface. We present a deferred shading scheme to handle the increased amount of data involved in shading the deformable surface reflectance field. We show animations of various objects that were acquired with 3D photography.
IEEE Trans. Vis. Comput. Graph.	Dynamic Interaction between Deformable Surfaces and Nonsmooth Objects.	Wingo Sai-Keung Wong,George Baciu	2005	In this paper, we introduce new techniques that enhance the computational performance for the interactions between sharp objects and deformable surfaces. The new formulation is based on a time-domain predictor-corrector model. For this purpose, we define a new kind of (\pi, \beta, {\bf I}){\hbox{-}}{\rm surface}. The partitioning of a deformable surface into a finite set of (\pi, \beta, {\bf I}){\hbox{-}}{\rm surfaces} allows us to prune a large number of noncolliding feature pairs. This leads to a significant performance improvement in the collision detection process. The intrinsic collision detection is performed in the time domain. Although it is more expensive compared to the static interference test, it avoids portions of the surfaces passing through each other in a single time step. In order to resolve all the possible collision events at a given time, a penetration-free motion space is constructed for each colliding particle. By keeping the velocity of each particle inside the motion space, we guarantee that the current colliding feature pairs will not penetrate each other in the subsequent motion. A static analysis approach is adopted to handle friction by considering the forces acting on the particles and their velocities. In our formulation, we further reduce the computational complexity by eliminating the need to compute repulsive forces.
IEEE Trans. Vis. Comput. Graph.	Psychophysical Evaluation of In-Situ Ultrasound Visualization.	Bing Wu,Roberta L. Klatzky,Damion Shelton,George D. Stetten	2005	"We present a novel psychophysical method for evaluating ultrasonography based on Real-Time Tomographic Reflection (RTTR), in comparison to Conventional Ultrasound (CUS). The method measures the user's perception of the location of an ultrasound-imaged target independently from assessing the action employed to reach it. Three experiments were conducted with the Sonic Flashlight (SF), an RTTR device, and CUS. The first two experiments determined subjects' perception of target location with a triangulation-by-pointing task. Depth perception with the SF was comparable to direct vision, while CUS caused considerable underestimation of target depth. Binocular depth information in the SF was shown to significantly contribute to its superiority. The third experiment tested subjects in an ultrasound-guided needle insertion task. Because the SF provides visualization of the target at its actual location, subjects performed insertions faster and more accurately by using the SF rather than CUS. Furthermore, the trajectory analysis showed that insertions with the SF generally went directly to the target along the desired path, while CUS often led to a large deviation from the correct path consistent with the observed underestimation of target depth. These findings lend great promise to the use of RTTR-based imaging in clinical practice and provide precise means of assessing efficacy."
IEEE Trans. Vis. Comput. Graph.	Quick-VDR: Out-of-Core View-Dependent Rendering of Gigantic Models.	Sung-Eui Yoon,Brian Salomon,Russell Gayle,Dinesh Manocha	2005	We present a novel approach for interactive view-dependent rendering of massive models. Our algorithm combines view-dependent simplification, occlusion culling, and out-of-core rendering. We represent the model as a clustered hierarchy of progressive meshes (CHPM). We use the cluster hierarchy for coarse-grained selective refinement and progressive meshes for fine-grained local refinement. We present an out-of-core algorithm for computation of a CHPM that includes cluster decomposition, hierarchy generation, and simplification. We introduce novel cluster dependencies in the preprocess to generate crack-free, drastic simplifications at runtime. The clusters are used for LOD selection, occlusion culling, and out-of-core rendering. We add a frame of latency to the rendering pipeline to fetch newly visible clusters from the disk and avoid stalls. The CHPM reduces the refinement cost of view-dependent rendering by more than an order of magnitude as compared to a vertex hierarchy. We have implemented our algorithm on a desktop PC. We can render massive CAD, isosurface, and scanned models, consisting of tens or a few hundred million triangles at 15--35 frames per second with little loss in image quality.
IEEE Trans. Vis. Comput. Graph.	Registration Based on Projective Reconstruction Technique for Augmented Reality Systems.	M. L. Yuan,Soh-Khim Ong,Andrew Y. C. Nee	2005	In AR systems, registration is one of the most difficult problems currently limiting their application. In this paper, we propose a simple registration method using projective reconstruction. This method consists of two steps: embedding and tracking. Embedding involves specifying four points to build the world coordinate system on which a virtual object will be superimposed. In tracking, a projective reconstruction technique is used to track these four specified points to compute the model view transformation for augmentation. This method is simple, as only four points need to be specified at the embedding stage and the virtual object can then be easily augmented onto a real scene from a video sequence. In addition, it can be extended to a scenario using the projective matrix that has been obtained from previous registration results using the same AR system. The proposed method has three advantages: 1) It is fast because the linear least square method can be used to estimate the related matrix in the algorithm and it is not necessary to calculate the fundamental matrix in the extended case. 2) A virtual object can still be superimposed on a related area even if some parts of the specified area are occluded during the whole process. 3) This method is robust because it remains effective even when not all the reference points are detected during the whole process, as long as at least six pairs of related reference points correspondences can be found. Some experiments have been conducted to validate the performance of the proposed method.
IEEE Trans. Vis. Comput. Graph.	Blind Robust Watermarking Schemes for Copyright Protection of 3D Mesh Objects.	Stefanos Zafeiriou,Anastasios Tefas,Ioannis Pitas	2005	In this paper, two novel methods suitable for blind 3D mesh object watermarking applications are proposed. The first method is robust against 3D rotation, translation, and uniform scaling. The second one is robust against both geometric and mesh simplification attacks. A pseudorandom watermarking signal is cast in the 3D mesh object by deforming its vertices geometrically, without altering the vertex topology. Prior to watermark embedding and detection, the object is rotated and translated so that its center of mass and its principal component coincide with the origin and the z-axis of the Cartesian coordinate system. This geometrical transformation ensures watermark robustness to translation and rotation. Robustness to uniform scaling is achieved by restricting the vertex deformations to occur only along the r coordinate of the corresponding (r, \theta, \phi) spherical coordinate system. In the first method, a set of vertices that correspond to specific angles \theta is used for watermark embedding. In the second method, the samples of the watermark sequence are embedded in a set of vertices that correspond to a range of angles in the \theta domain in order to achieve robustness against mesh simplifications. Experimental results indicate the ability of the proposed method to deal with the aforementioned attacks.
IEEE Trans. Vis. Comput. Graph.	Topological Lines in 3D Tensor Fields and Discriminant Hessian Factorization.	Xiaoqiang Zheng,Beresford N. Parlett,Alex Pang	2005	This paper addresses several issues related to topological analysis of 3D second order symmetric tensor fields. First, we show that the degenerate features in such data sets form stable topological lines rather than points, as previously thought. Second, the paper presents two different methods for extracting these features by identifying the individual points on these lines and connecting them. Third, this paper proposes an analytical form of obtaining tangents at the degenerate points along these topological lines. The tangents are derived from a Hessian factorization technique on the tensor discriminant and leads to a fast and stable solution. Together, these three advances allow us to extract the backbone topological lines that form the basis for topological analysis of tensor fields.
IEEE Trans. Vis. Comput. Graph.	Decorating Surfaces with Bidirectional Texture Functions.	Kun Zhou,Peng Du,Lifeng Wang,Yasuyuki Matsushita,Jiaoying Shi,Baining Guo,Heung-Yeung Shum	2005	We present a system for decorating arbitrary surfaces with bidirectional texture functions (BTF). Our system generates BTFs in two steps. First, we automatically synthesize a BTF over the target surface from a given BTF sample. Then, we let the user interactively paint BTF patches onto the surface such that the painted patches seamlessly integrate with the background patterns. Our system is based on a patch-based texture synthesis approach known as quilting. We present a graphcut algorithm for BTF synthesis on surfaces and the algorithm works well for a wide variety of BTF samples, including those which present problems for existing algorithms. We also describe a graphcut texture painting algorithm for creating new surface imperfections (e.g., dirt, cracks, scratches) from existing imperfections found in input BTF samples. Using these algorithms, we can decorate surfaces with real-world textures that have spatially-variant reflectance, fine-scale geometry details, and surfaces imperfections. A particularly attractive feature of BTF painting is that it allows us to capture imperfections of real materials and paint them onto geometry models. We demonstrate the effectiveness of our system with examples.
IEEE Trans. Vis. Comput. Graph.	Comparison of Path Visualizations and Cognitive Measures Relative to Travel Technique in a Virtual Environment.	Catherine A. Zanbaka,Benjamin Lok,Sabarish Babu,Amy Catherine Ulinski,Larry F. Hodges	2005	"We describe a between-subjects experiment that compared four different methods of travel and their effect on cognition and paths taken in an immersive virtual environment (IVE). Participants answered a set of questions based on Crook's condensation of Bloom's taxonomy that assessed their cognition of the IVE with respect to knowledge, understanding and application, and higher mental processes. Participants also drew a sketch map of the IVE and the objects within it. The users' sense of presence was measured using the Steed-Usoh-Slater Presence Questionnaire. The participants' position and head orientation were automatically logged during their exposure to the virtual environment. These logs were later used to create visualizations of the paths taken. Path analysis, such as exploring the overlaid path visualizations and dwell data information, revealed further differences among the travel techniques. Our results suggest that, for applications where problem solving and evaluation of information is important or where opportunity to train is minimal, then having a large tracked space so that the participant can walk around the virtual environment provides benefits over common virtual travel techniques."