首页 | 本学科首页   官方微博 | 高级检索  
检索        

视知觉表征相似性理论
引用本文:陈玉翠,张伟伟,沈政.视知觉表征相似性理论[J].山西医科大学学报,2000(Z1).
作者姓名:陈玉翠  张伟伟  沈政
作者单位:北京大学心理学系!北京100871
摘    要:前言视觉是一个古老的研究课题,许多世纪以来,视知觉问题一直吸引着科学家的好奇心。近年来在视觉神经生理学、视觉心理学和人工视觉(计算机视觉、机器人视觉或视觉代偿)等领域都得到了很显著的发展,提出了一些非常有意义的理论1,2]。目前的视觉理论都认为视觉对外部世界的表征要么是一种几何复制子3],或者是相对于环境的这些大范围不变量的表征4]。他们强调的都是物体与相应的内部表征之间的一级关系(firstorderrelation)问题,即通过物体与其内部表征之间的相似性进行,即通过相似性表征(representatio…

关 键 词:视知觉  人工智能  表征上相似性

Representation of similarity:a theory on visual cognition
CHEN Yu-cui,ZHANG Wei-wei,SHEN Zheng.Representation of similarity:a theory on visual cognition[J].Journal of Shanxi Medical University,2000(Z1).
Authors:CHEN Yu-cui  ZHANG Wei-wei  SHEN Zheng
Abstract:Assessments of similarity seem to be important for a variety of cognitive acts, ranging from problem solving to categorization to memory retrieval. Similarity is important properties to both human cognitive activities and computer. Therefore interdisciplinary research has already accumulated a number of facts and different theories were put forward. According to the proposed theory of representation of similarity, a shape is presented internally by the response of a small number of tuned modules, each broadly selective for some reference shape, whose similarity to the stimulus it measures. This shape space supports representations of distal shape similarity that are verdical as second-order isomorphisms (i.e. correspondence between distal and proximal similarities among shapes, rather than between distal shape and their proximal representations). An ensemble or a Chorus of n classifiers maps the distal shape to a proximal representation space R k. In the domain of shape perception, Multidimensional Scaling (MDS) has been applied in the analysis of perceived similarity among relatively simple two-dimensional figures. The column organization of the inferotemporal cortex (IT) seems to be equally compatible with the notion that entire objects are represented, as called by the Chorus model.
Keywords:Visual perception  artificial intelligence  representation of similarity  
本文献已被 CNKI 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号