Seeing
The computational approach to biological vision
About this book
Seeing has puzzled scientists and philosophers for centuries and it continues to do so. This new edition of a classic text offers an accessible but rigorous introduction to the computational approach to understanding biological visual systems.
The authors of Seeing, taking as their premise David Marr's statement that "to understand vision by studying only neurons is like trying to understand bird flight by studying only feathers," make use of Marr's three different levels of analysis in the study of vision: the computational level, the algorithmic level, and the hardware implementation level.
Each chapter applies this approach to a different topic in vision by examining the problems the visual system encounters in interpreting retinal images and the constraints available to solve these problems; the algorithms that can realise the solution; and the implementation of these algorithms in neurons.
Seeing has been thoroughly updated for this edition and expanded to more than three times its original length. It is designed to lead the reader through the problems of vision, from the common (but mistaken) idea that seeing consists just of making pictures in the brain to the minutiae of how neurons collectively encode the visual features that underpin seeing. Although it assumes no prior knowledge of the field, some chapters present advanced material.
This makes it the only textbook suitable for both undergraduate and graduate students that takes a consistently computational perspective, offering a firm conceptual basis for tackling the vast literature on vision.
It covers a wide range of topics, including aftereffects, the retina, receptive fields, object recognition, brain maps, Bayesian perception, motion, colour, and stereopsis.
MatLab code is available on the book's website, which includes a simple demonstration of image convolution.
Published 2010 by MIT Press
Download cover and contents page (PDF, 1.8MB)
Available from:
Contents
Preface to second edition ix
Preface to first edition xii
Chapter 1. Seeing: what is it? 1
Chapter 2. Seeing shape from texture 29
Chapter 3. Seeing with receptive fields 55
Chapter 4. Seeing aftereffects 75
Chapter 5. Seeing edges 111
Chapter 6. Seeing and the retina 133
Chapter 7. Seeing figure from ground 155
Chapter 8. Seeing objects 173
Chapter 9. Seeing with brain cells 205
Chapter 10. Seeing with brain maps 229
Chapter 11. Seeing and complexity theory 255
Chapter 12. Seeing and psychophysics 281
Chapter 13. Seeing as inference 307
Chapter 14. Seeing motion, part I 325
Chapter 15. Seeing motion, part II 355
Chapter 16. Seeing black, gray, and white 373
Chapter 17. Seeing color 397
Chapter 18. Seeing with two eyes, part I 419
Chapter 19. Seeing with two eyes, part II 465
Chapter 20. Seeing by combining cues 497
Chapter 21. Seeing in the blocks world 511
Chapter 22. Seeing and consciousness 527
Chapter 23. Seeing summarized 539
Index 551
Reviews
"Seeing is not a new edition but a completely new book, and a unique book – a carefully written, beautifully illustrated text of the computational approach to human vision that will take the reader from first principles to cutting-edge ideas about all levels of the visual process."
Oliver Braddick, Department of Experimental Psychology, University of Oxford
"It's back! In its first incarnation, this was one of the treasured books of vision, launching a thousand seminars, workshops, and courses on vision. This second edition covers even more than the first but keeps the excitement of the computational and physiological research that was the strength of the original. It's accessible, advanced, great to read, and fabulous for upper-level undergraduate and graduate courses – an absolute winner."
Patrick Cavanagh, Professeur des universites, Universite Paris Descartes, and Research Professor of Psychology, Harvard University
This volume is an excellent example of how textbooks ought to be written and classes taught. Like its predecessor, the current edition employs an exquisite selection of visual illusions and illustrations to provoke a strong desire to understand how the visual system works. Having done so, it proceeds to demonstrate, by example, how this can be achieved by formulating visual phenomena as computational problems that have to be solved by the brain.
Many topics are addressed in this work, from lightness perception to familiar object recognition. In contrast to most textbooks on visual perception, this volume is not organised to reflect the functional hierarchy of the visual system. Instead, it is specifically designed to accustom readers to the computational approach. Thus, "high-level" topics such as 3D-shape perception can sometimes be presented before "low- level" topics such as lightness constancy; this is because the former is better suited as an introductory demonstration of how computational theories can be derived from direct observations, while the latter avails itself to the demonstration of more advanced image processing techniques (ie deconvolution).
Readers can be expected to gain a full appreciation for the diverse competencies of the visual system. This is because they are directly engaged in thinking about the problems that the visual system has to deal with and are instructed in how acceptable solutions can be derived. Clearly, this is a more enjoyable learning experience than being presented with a list of factoids about the visual brain. In fact, the experience of reading this book is not unlike the pleasure that some might gain from working through a crossword puzzle.
In conclusion, this volume is suitable for general readers as well as vision scientists, regardless of their topical interests. It dispels any misconception that the "[computational] approach has something uniquely to do with computers" (p. 53) and provides a compelling case that a process is best understood by first thinking about the computations that it is expected to perform.
Review of Seeing in The Quarterly Review of Biology, September 2011 by Heinrich H. Buelthoff and Lewis Leewui Chuang, Max Planck Institute for BioCybernetics, Tubingen, Germany