A functional role of multiple spatial resolution maps in form perception along the ventral visual pathway

Authors:
Yoshiki Kashimori;Nobuyuki Suzuki;Kazuhisa Fujita;Meihong Zheng;Takeshi Kambara
Affiliations:
Department of Applied Physics and Chemistry, Division of Bioinformatics, University of Electro-Communications, Chofu, Tokyo 182-8585, Japan and Department of Information Network Sciences, Graduate ...;Department of Information Network Sciences, Graduate School of Information Systems, University of Electro-Communications, Chofu, Tokyo 182-8585, Japan;Department of Information Network Sciences, Graduate School of Information Systems, University of Electro-Communications, Chofu, Tokyo 182-8585, Japan;Department of Applied Physics, School of Science and Engineering, Waseda University, Shinjuku, Tokyo 169-8555, Japan;Department of Applied Physics and Chemistry, Division of Bioinformatics, University of Electro-Communications, Chofu, Tokyo 182-8585, Japan and Department of Information Network Sciences, Graduate ...
Venue:
Neurocomputing
Year:
2005

Citing 2
Cited 1

A massively parallel architecture for a self-organizing neural pattern recognition machine

Computer Vision, Graphics, and Image Processing
A Hierarchical Dynamical Map as a Basic Frame for Cortical Mapping and Its Application to Priming

Neural Computation

Neural Mechanism for Extracting Object Features Critical for Visual Categorization Task

Neural Information Processing

Quantified Score

Hi-index	0.01

Visualization

Abstract

We present a functional model of form pathway in visual cortex based on the predictive recognition scheme, in which feedback image signals predicted from the memory system are compared with the feedforward image signals from the retina. We propose the functional roles of three kinds of spatial resolution maps in predictive perception of object form. The broad resolution map is used to categorize visual image based on the outline of the object and to choose the prediction image in the memory system based on the categorization. The middle resolution map is used for more detailed categorizations and for focusing prediction range. The fine resolution map is used to compare the feedforward image of a part of the object with the image predicted.