Computer Vision is the first book to take a full approach to the challenging issue of veridical 3D object representation. It introduces mathematical and conceptual advances that offer an unprecedented framework for analyzing the complex scene structure of the world. Leading theorists cover full 3D scene reconstruction, instead of the simplistic 2D planar algorithms employed in the past. They explore cutting-edge research on computational algorithms for scene analysis and present an integrated, complementary treatment of neural, behavioral, mathematical, and computational approaches. The text includes numerous graphics of complex processes, with many in color.

One of the grand challenges of artificial intelligence is to enable computers to interpret 3D scenes and objects from imagery. This book organizes and introduces major concepts in 3D scene and object representation and inference from still images, with a focus on recent efforts to fuse models of geometry and perspective with statistical machine learning. The book is organized into three sections: (1) Interpretation of Physical Space; (2) Recognition of 3D Objects; and (3) Integrated 3D Scene Interpretation. The first discusses representations of spatial layout and techniques to interpret physical scenes from images. The second section introduces representations for 3D object categories that account for the intrinsically 3D nature of objects and provide robustness to change in viewpoints. The third section discusses strategies to unite inference of scene geometry and object pose and identity into a coherent scene interpretation. Each section broadly surveys important ideas from cognitive science and artificial intelligence research, organizes and discusses key concepts and techniques from recent work in computer vision, and describes a few sample approaches in detail. Newcomers to computer vision will benefit from introductions to basic concepts, such as single-view geometry and image classification, while experts and novices alike may find inspiration from the book's organization and discussion of the most recent ideas in 3D scene understanding and 3D object recognition. Specific topics include: mathematics of perspective geometry; visual elements of the physical scene, structural 3D scene representations; techniques and features for image and region categorization; historical perspective, computational models, and datasets and machine learning techniques for 3D object recognition; inferences of geometrical attributes of objects, such as size and pose; and probabilistic and feature-passing approaches for contextual reasoning about 3D objects and scenes. Table of Contents: Background on 3D Scene Models / Single-view Geometry / Modeling the Physical Scene / Categorizing Images and Regions / Examples of 3D Scene Interpretation / Background on 3D Recognition / Modeling 3D Objects / Recognizing and Understanding 3D Objects / Examples of 2D 1/2 Layout Models / Reasoning about Objects and Scenes / Cascades of Classifiers / Conclusion and Future Directions

This book constitutes the strictly refereed post-workshop proceedings of the second International Workshop on Object Representation in Computer Vision, held in conjunction with ECCV '96 in Cambridge, UK, in April 1996. The 15 revised full papers contained in the book were selected from 45 submissions for presentation at the workshop. Also included are three invited contributions based on the talks by Takeo Kanade, Jan Koenderink, and Ram Nevatia as well as a workshop report by the volume editors summarizing several panel discussions and the general state of the art in the area.

This book documents the scientific outcome of the International NSF-ARPA Workshop on Object Representation in Computer Vision, held in New York City in December 1994 with invited participants chosen among the recognized experts in the field. The volume presents the complete set of papers in revised full-length versions. In addition, the first paper is a report on the workshop in which the panel discussions as well as the conclusions and recommendations reached by the workshop participants are summarized. Altogether the volume provides an excellent, in-depth view of the state of the art in this active area of research and applications.

One of the most natural representations for modelling spatial objects in computers is discrete representations in the form of a 2D square raster and a 3D cubic grid, since these are naturally obtained by segmenting sensor images. However, the main difficulty is that discrete representations are only approximations of the original objects, and can only be as accurate as the cell size allows. If digitisation is done by real sensor devices, then there is the additional difficulty of sensor distortion. To overcome this, digital shape features must be used that abstract from the inaccuracies of digital representation. In order to ensure the correspondence of continuous and digital features, it is necessary to relate shape features of the underlying continuous objects and to determine the necessary resolution of the digital representation. This volume gives an overview and a classification of the actual approaches to describe the relation between continuous and discrete shape features that are based on digital geometric concepts of discrete structures. Audience: This book will be of interest to researchers and graduate students whose work involves computer vision, image processing, knowledge representation or representation of spatial objects.

This book documents the scientific outcome of the International NSF-ARPA Workshop on Object Representation in Computer Vision, held in New York City in December 1994 with invited participants chosen among the recognized experts in the field. The volume presents the complete set of papers in revised full-length versions. In addition, the first paper is a report on the workshop in which the panel discussions as well as the conclusions and recommendations reached by the workshop participants are summarized. Altogether the volume provides an excellent, in-depth view of the state of the art in this active area of research and applications.

Over the past 40 years, neurobiology and computational neuroscience has proved that deeper understanding of visual processes in humans and non-human primates can lead to important advancements in computational perception theories and systems. One of the main difficulties that arises when designing automatic vision systems is developing a mechanism that can recognize - or simply find - an object when faced with all the possible variations that may occur in a natural scene, with the ease of the primate visual system. The area of the brain in primates that is dedicated at analyzing visual information is the visual cortex. The visual cortex performs a wide variety of complex tasks by means of simple operations. These seemingly simple operations are applied to several layers of neurons organized into a hierarchy, the layers representing increasingly complex, abstract intermediate processing stages. In this Research Topic we propose to bring together current efforts in neurophysiology and computer vision in order 1) To understand how the visual cortex encodes an object from a starting point where neurons respond to lines, bars or edges to the representation of an object at the top of the hierarchy that is invariant to illumination, size, location, viewpoint, rotation and robust to occlusions and clutter; and 2) How the design of automatic vision systems benefit from that knowledge to get closer to human accuracy, efficiency and robustness to variations.

This monograph by one of the world's leading vision researchers provides a thorough, mathematically rigorous exposition of a broad and vital area in computer vision: the problems and techniques related to three-dimensional (stereo) vision and motion. The emphasis is on using geometry to solve problems in stereo and motion, with examples from navigation and object recognition. Faugeras takes up such important problems in computer vision as projective geometry, camera calibration, edge detection, stereo vision (with many examples on real images), different kinds of representations and transformations (especially 3-D rotations), uncertainty and methods of addressing it, and object representation and recognition. His theoretical account is illustrated with the results of actual working programs.Three-Dimensional Computer Vision proposes solutions to problems arising from a specific robotics scenario in which a system must perceive and act. Moving about an unknown environment, the system has to avoid static and mobile obstacles, build models of objects and places in order to be able to recognize and locate them, and characterize its own motion and that of moving objects, by providing descriptions of the corresponding three-dimensional motions. The ideas generated, however, can be used indifferent settings, resulting in a general book on computer vision that reveals the fascinating relationship of three-dimensional geometry and the imaging process.