A multi-layer sparse coding network learns contour coding from natural images

An important approach in visual neuroscience considers how the function of the early visual system relates to the statistics of its natural input. Previous studies have shown how many basic properties of the primary visual cortex, such as the receptive fields of simple and complex cells and the spat...

Full description

Saved in:
Bibliographic Details
Published in:Vision research (Oxford) Vol. 42; no. 12; pp. 1593 - 1605
Main Authors: Hoyer, Patrik O, Hyvärinen, Aapo
Format: Journal Article
Language:English
Published: Oxford Elsevier Ltd 01.06.2002
Elsevier Science
Subjects:
Biological and medical sciences
Cerebral Cortex - physiology
Contours
Cortex
Fundamental and applied biological sciences. Psychology
Humans
Independent component analysis
Models, Neurological
Natural images
Nerve Net
Neural networks
Neurophysiology
Perception
Psychology. Psychoanalysis. Psychiatry
Psychology. Psychophysiology
Vision
Vision, Ocular - physiology
Neural networks
Contours
Cortex
Independent component analysis
Natural images
Learning
Visual cortex
Coding
Contour
Vision
Central nervous system
Perception
Simulation model
Neural network
Mathematical model
Brain (vertebrata)
ISSN:0042-6989, 1878-5646
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:An important approach in visual neuroscience considers how the function of the early visual system relates to the statistics of its natural input. Previous studies have shown how many basic properties of the primary visual cortex, such as the receptive fields of simple and complex cells and the spatial organization (topography) of the cells, can be understood as efficient coding of natural images. Here we extend the framework by considering how the responses of complex cells could be sparsely represented by a higher-order neural layer. This leads to contour coding and end-stopped receptive fields. In addition, contour integration could be interpreted as top-down inference in the presented model.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0042-6989
1878-5646
DOI:10.1016/S0042-6989(02)00017-2