A Computational Perspective on Visual Attention

by Tsotsos, John K.

• ISBN13: 9780262015417
• ISBN10: 0262015412
• Edition: 1st
• Format: Hardcover
• Copyright: 2011-05-06
• Publisher: Mit Pr

Summary

Although William James declared in 1890, "Everyone knows what attention is," today there are many different and sometimes opposing views on the subject. This fragmented theoretical landscape may be because most of the theories and models of attention offer explanations in natural language or in a pictorial manner rather than providing a quantitative and unambiguous statement of the theory. They focus on the manifestations of attention instead of its rationale. In this book, John Tsotsos develops a formal model of visual attention with the goal of providing a theoretical explanation for why humans (and animals) must have the capacity to attend. He takes a unique approach to the theory, using the full breadth of the language of computation--rather than simply the language of mathematics--as the formal means of description. The result, the Selective Tuning model of vision and attention, explains attentive behavior in humans and provides a foundation for building computer systems that see with human-like characteristics. The overarching conclusion is that human vision is based on a general purpose processor that can be dynamically tuned to the task and the scene viewed on a moment-by-moment basis. Tsotsos offers a comprehensive, up-to-date overview of attention theories and models and a full description of the Selective Tuning model, confining the formal elements to two chapters and two appendixes. The text is accompanied by more than 100 illustrations in black and white and color; additional color illustrations and movies are available on the book's Web site

Author Biography

John K. Tsotsos is Professor of Computer Science and Engineering, Distinguished Research Professor of Vision Science, and Canada Research Chair in Computational Vision at York University and a Fellow of the Royal Society of Canada (FRSC).

Table of Contents

Preface	p. xi
Acknowledgments	p. xv
Attention-We All Know What It Is	p. 1
But Do We Really?	p. 1
Moving Toward a Computational Viewpoint	p. 7
What Is Attention?	p. 10
Computational Foundations	p. 11
Attempting to Understand Visual Processing Capacity	p. 11
The Language of Computation	p. 14
Capacity Limits and Computational Complexity	p. 16
Human Perception/Cognition and Computation	p. 18
The Computational Complexity of Vision	p. 21
Extending to Active Vision	p. 29
Extending to Cognition and Action	p. 32
Extending to Sensor Planning	p. 32
Complexity Constrains Visual Processing Architecture	p. 33
The Problems with Pyramids	p. 38
Attention Is. …	p. 51
Theories and Models of Visual Attention	p. 53
The Elements of Visual Attention	p. 54
A Taxonomy of Models	p. 59
Other Relevant Ideas	p. 75
Summary	p. 78
Selective Tuning: Overview	p. 81
The Basic Model	p. 82
Saliency and Its Role in ST	p. 86
Selective Tuning with Fixation Control	p. 88
Differences with Other Models	p. 93
Summary	p. 96
Selective Tuning: Formulation	p. 97
Objective	p. 97
Representations	p. 98
Neurons and Circuits for Selective Tuning	p. 106
Selection	p. 114
Competition to Represent a Stimulus	p. 121
More on Top-Down Tracing	p. 122
Inhibition of Return	p. 124
Peripheral Priority Map Computation	p. 124
Fixation History Map Maintenance	p. 125
Task Guidance	p. 126
Comparisons with Other Models	p. 127
Summary	p. 131
Attention, Recognition, and Binding	p. 133
What Is Recognition?	p. 134
What Is Visual Feature Binding?	p. 139
Four Binding Processes	p. 141
Binding Decision Process	p. 145
Putting It All Together	p. 146
Summary	p. 149
Selective Tuning: Examples and Performance	p. 151
P-Lattice Representation of Visual Motion Information	p. 151
Priming	p. 153
Results After a Single Feed-Forward Pass (Convergence Binding)	p. 160
Results from a Single Feed-Forward Pass Followed by a Single Recurrent Pass (Full Recurrence Binding)	p. 164
Attending to Multiple Stimuli (Type I Iterative Recurrence Binding)	p. 166
Empirical Performance of Recurrence Binding (Localization)	p. 168
Visual Search	p. 174
Type II Iterative Recurrence Binding	p. 186
Saliency and AIM	p. 187
Summary	p. 190
Explanations and Predictions	p. 193
Explanations	p. 195
Predictions with Experimental Support	p. 205
Some Supporting Experiments	p. 211
Summary	p. 231
Wrapping Up the Loose Ends	p. 233
The Loose Ends	p. 236
Vision as Dynamic Tuning of a General-Purpose Processor	p. 247
Final Words	p. 248
Appendixes	p. 251
A Few Notes on Some Relevant Aspects of Complexity Theory	p. 251
Proofs of the Complexity of Visual Match	p. 255
The Representation of Visual Motion Processes	p. 265
References	p. 275
Author Index	p. 297
Subject Index	p. 305
Table of Contents provided by Ingram. All Rights Reserved.