Structures of Surround Modulation for the Border-Ownership Selectivity of V2 Cells

被引:0
|
作者
Nakata, Yusuke [1 ]
Sakai, Ko [1 ]
机构
[1] Univ Tsukuba, Dept Comp Sci, Tsukuba, Ibaraki 3058573, Japan
关键词
Visual Cortex; Computational Neuroscience; Figure Ground Segregation; Surround Modulation; Population Coding; ORGANIZATION; SUPPRESSION;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The determination of Border-Ownership (BO) that defines the direction of figure along contour is a crucial step toward figure-ground segregation. Recent physiological and computational studies have reported that neurons in V2/V4 are selective to BO [1], and that surround modulation underlies the selectivity [2, 3]. Accurate determination of BO along complex contours in natural images appears difficult for a single BO-selective cell, leading to the hypothesis that multiple cells work simultaneously for BO perception. We studied computationally how many BO cells and what characteristics of the surround modulation are required to achieve accurate determination of BO in natural images. The results indicated that 10 model BO-cells achieve 82% correct while single model cells yield 65%. The results suggest that grouping a small number of BO-selective cells increases effectively the capability for the determination of BO.
引用
收藏
页码:383 / 391
页数:9
相关论文
共 50 条