Attention mechanism and its role in invariant pattern recognition

Abstract In this paper, the emergence of attention, its quantitative description and its role in the invariant pattern recognition is investigated. The visual invariance perception is considered as the self-organization information processing and an “attention function”, which is determined by the internal functional state of this system, modulates the parameters of the system, as well as the strategy for information processing. Thus, attention is the intrinsic property of the information processing cortex, rather than a separated mechanism of an additional “attention module” as is assumed by many researchers. A neural network model based on this concept is tested.

[1]  Laurenz Wiskott,et al.  Face recognition by dynamic link matching , 1996 .

[2]  J. Duncan,et al.  Competitive brain activity in visual attention , 1997, Current Opinion in Neurobiology.

[3]  T. Wiesel.  Neural Mechanisms of Visual Perception , 1997 .

[4]  Kenneth H. Britten,et al.  Attention is everywhere , 1996, Nature.

[5]  J. Duncan,et al.  Visual search and stimulus similarity. , 1989, Psychological review.

[6]  Wolfgang Konen,et al.  Learning to Generalize from Single Examples in the Dynamic Link Architecture , 1993, Neural Computation.

[7]  A. Treisman,et al.  A feature-integration theory of attention , 1980, Cognitive Psychology.

[8]  Wolfgang Konen,et al.  A fast dynamic link matching algorithm for invariant pattern recognition , 1994, Neural Networks.

[9]  F. Crick.  Function of the thalamic reticular complex: the searchlight hypothesis. , 1984, Proceedings of the National Academy of Sciences of the United States of America.