Synopses & Reviews
This book presents a unique collection of articles on shape, contour and grouping in computer vision. Besides revised versions of research papers originally presented at a workshop, the book contains expository articles introducing the area to a broader audience and surveying the state of the art. The 19 articles presented were carefully reviewed. They are organized in the following sections: introduction; shape; shading; grouping; representation and recognition; and statistics,learning and recognition.
Synopsis
Computer vision has been successful in several important applications recently. Vision techniques can now be used to build very good models of buildings from pictures quickly and easily, to overlay operation planning data on a neuros- geon's view of a patient, and to recognise some of the gestures a user makes to a computer. Object recognition remains a very di cult problem, however. The key questions to understand in recognition seem to be: (1) how objects should be represented and (2) how to manage the line of reasoning that stretches from image data to object identity. An important part of the process of recognition { perhaps, almost all of it { involves assembling bits of image information into helpful groups. There is a wide variety of possible criteria by which these groups could be established { a set of edge points that has a symmetry could be one useful group; others might be a collection of pixels shaded in a particular way, or a set of pixels with coherent colour or texture. Discussing this process of grouping requires a detailed understanding of the relationship between what is seen in the image and what is actually out there in the world.
Table of Contents
An empirical-statistical agenda for recognition / David Forsyth -- A formal-physical agenda for recognition / Joe Mundy -- Shape models and object recognition / Jean Ponce ... et al. -- Order structure, correspondence, and shape based categories / Stefan Carlsson -- Quasi-invariant parameterisations and their applications in computer vision / Jun Sato and Roberto Cipolla -- Representations for recognition under variable illumination / David J. Kriegman, Peter N. Belhumeur, and Athinodoros S. Georghiades -- Shadows, shading, and projecctive ambiguity / Peter N. Belhumeur, David J. Kriegman, and Alan L. Yuille -- Grouping in the normalized cut framework / Jitendra Malik ... et al. -- Geometric grouping of repeated elements within images / Frederik Schaffalitzky and Andrew Zisserman -- Constrained symmetry for change detection / Rupert W. Curwen and Joe L. Mundy -- Grouping based on coupled diffusion maps / Marc Proesmans and Luc Van Gool -- Integrating geometric and photometric information for image retrieval / Cordelia Schmid, Andrew Zisserman, and Roger Mohr -- Towards the integration of geometric and appearance-based object recognition / Joe Mundy and Tushar Saxena -- Recognizing objects using color-annotated adjacency graphs / Peter Tu, Tushar Saxena, and Richard Hartley -- A cooperating strategy for objects recognition / Antonio Chella ... et al. -- Model selection for two view geometry : a review / Philip H.S. Torr -- Finding objects by grouping primitives / David Forsyth, John Haddon, and Sergey Ioffe -- Object recognition with gradient-based learning / Yann LeCun ... et al..