Presentation 1997/1/24
A Mathematical Model of Visual Attention
Teruyoshi Washizawa,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper peasants a mathematical model for visual attention with saccades. Our model consists of three parts, i.e., early vision, visual memory, and saliency map. In the early vision, subspace of multi-resolution representation (SubMRR) is employed in order to approximate non-uniform sampling of human retina. A probabilistic cellular automaton (PCA) is used for representation of the visual memory. The saliency map is a tree composed of nodes configuration of which is same as those of cells of the PCA. Every node has an average mutual information (AMI) between a target mode and itself. The corresponding location of the node with the maximum AMI is selected as a FOA.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Visual attention / saccades / average mutual information / MRR / context dependency
Paper # HIP96-27
Date of Issue

Conference Information
Committee HIP
Conference Date 1997/1/24(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Vice Chair

Paper Information
Registration To Human Information Processing (HIP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A Mathematical Model of Visual Attention
Sub Title (in English)
Keyword(1) Visual attention
Keyword(2) saccades
Keyword(3) average mutual information
Keyword(4) MRR
Keyword(5) context dependency
1st Author's Name Teruyoshi Washizawa
1st Author's Affiliation Canon Inc.()
Date 1997/1/24
Paper # HIP96-27
Volume (vol) vol.96
Number (no) 499
Page pp.pp.-
#Pages 6
Date of Issue