ECCV 2008 Tutorial
Date: Sunday October 12th - Morning Venue: Palais des Congres Parc Chanot, in Marseille.
Visual attention: biology, computational models, and applications
Instructors: John K. Tsotsos, Neil Bruce, Albert L. Rothenstein
Recent years have seen a renewed interest in the study of visual attention. This tutorial aims to present the state of the art in the computational modeling of visual attention. Starting from the biology of the primate visual system, we will show how various aspects of visual attention have been addressed in the computational modeling literature, and conclude with a more detailed discussion of a number of recent applications.
The half day tutorial covers a broad range of topics from the basics of primate visual function to applications of visual attention in computer vision. Most of the major paradigms and issues in selective attention research are discussed in a systematic way. The tutorial deals with the following topics:
- Introduction - The biology of visual attention
- What is visual attention - Biology and computational motivation
- The visual cortex
- Classic results - psychophysics and imaging establishing the role of attention
- Kinds of visual attention
- Computational reasons for visual attention
- Computational models of visual attention
- Attention in computer vision: The story so far - overview of existing attempts at attentive behavior in a computer vision context
- Saliency and related models - This includes a critical examination of the limitations of saliency based models and pushes for a more general interpretation of the notion of saliency in machine vision
- Top-down bias
- Attentional selection and filtering - For the above two topics, once again a critical examination of strengths and weaknesses of models falling in to these categories will be included with discussion pertinent to their role in machine vision
- "Attentional control" : Inhibition of return, covert and overt attention
- Attention and recognition
- Discussion of current application areas: i.e. Selective processing, Object recognition, ROI selection etc.
- A look towards the future: We identify a variety of areas that would benefit from having a closer connection with attentive behavior. Some portion of this is devoted to thinking of attention in machine vision in a context more general than selecting regions of interest, perhaps towards the more general goal of establishing a computational representation for which the ratio of signal to noise is improved by virtue of the role of attention
- Discuss the relevance of visual attention to computer vision systems
- John K. Tsotsos completed his education at the University of Toronto, an honours B.A.Sc. Engineering Science (1974), M.Sc.in Computer Science (1976) and Ph.D. in Computer Science (1980). He was on the faculty in Computer Science and in Medicine at the University of Toronto from 1980--1999, and in the Dept. of Computer Science & Engineering at York University in Toronto since 2000. He was Director of York's Centre for Vision Research, 2000--2006. He holds the NSERC Tier I Canada Research Chair in Computational Vision.
- Neil Bruce received a Double Honours B.Sc. degree from the University of Guelph, Canada in Mathematics and Computer Science, and a M.A.Sc. in System Design Engineering from the University of Waterloo. He is currently completing a Ph.D. in Computer Science at York University in Toronto, Canada.
- Albert L. Rothenstein received an Electronics and Telecommunications Engineering degree from the Technical University of Timisoara, Romania and a M.Sc. in Computer Science from the University of Toronto. He was a senior software engineer at the IBM Toronto Laboratories before returning to school, where he is currently a Ph.D. student in Computer Science at York University in Toronto.
Rothenstein, A.L. and Tsotsos, J.K., Attention Links Sensing to Recognition, Image and Vision Computing, 26(1), 2008, pp. 114-126
Bruce, N., Tsotsos, J.K., Saliency Based on Information Maximization. Advances in Neural Information Processing Systems (NIPS 2005), 18, pp. 155-162, June 2006.
Tsotsos, J.K., Rodriguez-Sanchez, A., Rothenstein, A., Simine, E., Different Binding
Strategies for the Different Stages of Visual Recognition, Brain Research , Available
online 23 May 2008
Tsotsos, J.K., What Roles can Attention Play in Recognition?, 7th International
Conference on Development and Learning, Monterey, California, August 9-12, 2008.
Tsotsos, J.K., Liu, Y., Martinez-Trujillo, J., Pomplun, M., Simine, E., Zhou, K.,
Attending to Motion, Computer Vision and Image Understanding, Vol 100, 1-2, p 3 - 40,
Tsotsos, J.K., Itti, L., Rees, G., A Brief and Selective history of Attention,
Neurobiology of Attention, ed. by Itti, Rees and Tsotsos, Elsevier/Academic Press, 2005.
Tsotsos, J.K., Complexity and Visual Attention, Neurobiology of Attention, ed. by Itti,
Rees and Tsotsos, Elsevier/Academic Press, 2005.
Tsotsos, J.K., The Selective Tuning Model of Attention, Neurobiology of Attention, ed. by
Itti, Rees and Tsotsos, Elsevier/Academic Press, 2005.
Tsotsos, J.K., Culhane, S., Cutzu, F., "From Theoretical Foundations to a Hierarchical
Circuit for Selective Attention", Visual Attention and Cortical Cicrcuits, p. 285 -- 306,
ed. by J. Braun, C. Koch, and J. Davis, MIT Press, 2001.
Tsotsos, J.K., Culhane, S., Wai,
W., Lai, Y., Davis, N., Nuflo, F., Modeling visual attention via selective tuning,
Artifical Intelligence 78(1-2), p 507 - 547, 1995.
Tsotsos, J.K., On the Relative Complexity of Passive vs Active Visual Search,
International Journal of Computer Vision 7-2, p 127 - 141, 1992.
Tsotsos, J.K., Analyzing Vision at the Complexity Level, Behavioral and Brain Sciences
13-3, p423 - 445, 1990.
Tsotsos, J., "The Complexity of Perceptual Search Tasks", Proc. International Joint
Conference on Artificial Intelligence, Detroit, August, 1989, pp1571 - 1577.
Tsotsos, J.K., "A `Complexity Level' Analysis of Vision", Proceedings of International
Conference on Computer Vision: Human and Machine Vision Workshop, London, England,June
Centre for Vision Research
Laboratory for Active and Attentive Vision