PSYCHOPHYSICAL AND COMPUTATINAL STUDY ON DETERMINING SEPARATION AND INTEGRATION MECHANISMS FOR THE OBJECT RECOGNITION PROCESSES AND THREE-DIMENSIONAL PERCEPTION PROCESSES

确定物体识别过程和三维感知过程的分离和整合机制的心理物理和计算研究

基本信息

批准号：
13680861
负责人：
SAKAI Ko
金额：
$ 1.98万
依托单位：
UNIVERSITY OF TSUKUBA
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (C)
财政年份：
2001
资助国家：
日本
起止时间：
2001 至 2002
项目状态：
已结题

来源：
https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-13680861/
关键词：
VISION STEREOPSIS OBJECT RECOGNITION PSYCHOPHYSICS INTEGRATION OF INFORMATION

项目摘要

The project studied psychophysically and computationally the relevance between shape perception and three-dimensional perception in order to understand the information integration processes relevant to higher-order visual functions.In psychophysical experiments, we generated stimuli whose shape perception, three dimensional perception and stereopsis are independent, and obtained the relevance between the shape perception annd 3D perception. We were able to distinguish shape perception and stereo perception which are mixed in the previous experiments. Specifically, we we generated stimuli whose shape perception, shape form shading, and stereo perception are independent, and determined the relation between the shape perception and 3d perception. The results suggest that 3D perception from shading has different characteristics from depth perception from binocular stereopsis.In computational study, we investigated if 3D information is relevant to object recognition, what would be the roles and mechanisms. Specifically, (1) dynamics of information integration, (2) grouping of regions due to shape and binocular disparity are studied. We constructed a model for the grouping and integration in early to intermediate-level visual processes. We further constructed a model for V2 and V4 that reproduces the specificity to surface perception, and examined the integration of attributes to the perceived surfaces.

该项目在心理物理和计算上研究了形状感知和三维感知之间的相关性，以了解与高阶视觉功能相关的信息整合过程。在心理物理实验中，我们产生了刺激，其形状感知，三维感知和刻板性是独立的，并获得了相关性，并获得了Shape感知的相关性。我们能够区分以前实验中混合的形状感知和立体感知。具体而言，我们产生了刺激，其形状感知，形状形式的阴影和立体感感知是独立的，并确定了形状感知和3D感知之间的关系。结果表明，阴影的3D感知与双眼立体的深度感知具有不同的特征。在计算研究中，我们研究了3D信息是否与对象识别相关，是角色和机制。具体而言，（1）信息整合的动力学，（2）由于形状和双眼差异而导致的区域分组。我们构建了一个在早期至中级视觉过程中进行分组和集成的模型。我们进一步构建了V2和V4的模型，该模型重现了表面感知的特异性，并检查了属性与感知表面的集成。