PARTS OF RECOGNITION
PARTS OF RECOGNITION
复制标题
DOI：
10.1016/0010-0277(84)90022-2
复制
发表时间：
1984-01-01
期刊：
影响因子：
3.4
通讯作者：
RICHARDS, WA
中科院分区：
文献类型：
--
作者：
HOFFMAN, DD;RICHARDS, WA
为了完成物体识别任务，视觉系统将形状分解成多个部分。它使用定义零件边界而不是零件形状的规则来实现这一点。这条规则利用了自然的一致性，即横向性。零件及其描述和空间关系为形状记忆提供了第一个索引。这条规则可以解释几种视觉错觉。强调归纳推理在现有理论中的作用。本文总结了一些尚未解决的问题。
For the task of object recognition, the visual system decomposes shapes into parts. It does so using a rule defining part boundaries rather than part shapes. The rule explots a uniformity of nature, transversality. Parts with their descriptions and spatial relations provide a 1st index into a memory of shapes. This rule allows an explanation of several visual illusions. The role inductive inference in the present theory is stressed. A precis of unsolved problems concludes the present paper.