Review by Choice Review
The field of "computer vision" involves a variety of technologies including image processing, artificial intelligence, three-dimensional geometry, and robotics. It also addresses a variety of problems including the automatic recognition of objects, determining the positions of objects in a scene (localization), building 3-D models of objects from their 2-D projections, and characterizing the reflective properties of surfaces and textural properties of patterns. This book addresses two of these problems: recognition and localization. Taking the view that recognition is the verification of hypotheses (through state-space search) guided by geometric constraints, Grimson, with the assistance of two other contributors, shows how a wide variety of techniques contribute to the solution of practical problems. He is a recognized leader in research on computer vision, and his book is unique in bringing together in a coherent fashion the results of recent research on this aspect of computer vision. Numerous diagrams, graphs, tables, and photographs illustrate the ideas. An extensive bibliography, author index, and subject index are included. The writing style and production quality of the book are excellent. Readers need a firm grasp of undergraduate linear algebra, some calculus, and some combinatorics. Thus, the book will appeal primarily to graduate students, faculty, other researchers, and professional engineers involved with computer vision.-S. L. Tanimoto, University of Washington
Copyright American Library Association, used with permission.
Review by Choice Review
