ISSN 1671-3710
CN 11-4766/R
主办：中国科学院心理研究所
出版：科学出版社

心理科学进展 ›› 2012, Vol. 20 ›› Issue (3): 354-364.

• 研究前沿 • 上一篇    下一篇

视觉运动追踪的加工过程

陈婷婷;蒋长好;丁锦红   

  1. (1北京市“学习与认知”重点实验室, 首都师范大学心理系, 北京 100048)
    (2北京市“运动机能评定和技术分析”重点实验室, 首都体育学院, 北京 100191)
  • 收稿日期:2011-09-07 修回日期:1900-01-01 出版日期:2012-03-15 发布日期:2012-03-15
  • 通讯作者: 丁锦红

The Analysis of Visual Motion Tacking

CHEN Ting-Ting;JIANG Chang-Hao;DING Jin-Hong   

  1. (1 Beijing Key Lab of Learning and Cognition, Department of Psychology, Capital Normal University, Beijing 100048, China)
    (2 Beijing Key Lab of Physical Fitness Evaluation and Tech Analysis, Capital University of Physical Education and Sports, Beijing 100191, China)
  • Received:2011-09-07 Revised:1900-01-01 Online:2012-03-15 Published:2012-03-15
  • Contact: DING Jin-Hong

PDF (PC)

2279

摘要： 视觉运动追踪是运动知觉研究的一个重要领域。通过构建视觉运动追踪的过程模型和分析每个阶段的认知加工任务, 可以帮助人们认识运动物体识别的本质。视觉运动追踪包括目标获取和运动追踪两个加工过程：目标获取阶段的主要任务是将目标与背景分离, 集中注意力加工追踪目标; 运动追踪阶段的主要任务是启动平滑追踪眼动和追赶性眼跳, 并发挥行为水平、眼动水平和神经活动水平的预测机制。目标获取同时受背景和目标的运动特征和身份特征影响; 运动追踪系统发挥预测机制的基础是客体表征连续性, 而客体表征连续性的建立同时依赖于目标时空属性和身份特征的编码加工。因此, 视觉运动追踪是视觉系统对客体运动信息和身份语义信息整合的结果。其中, 客体运动信息的加工特性已经获得了比较广泛的研究, 而语义信息加工机制还有待进一步加强。

关键词: 视觉运动追踪, 目标获取, 运动追踪, 预测机制

Abstract: The aim of visual motion tracking is to capture and track moving objects through a sequence of images. The ability is crucial for the survival of many species. Visual motion tracking proceeds in two main parts: object capture and object tracking. The first step is to discriminate target’s signals from other object signals in the visual context, and then concentrate on encoding target’s motion and identity. The second step is to initiate smooth pursuit eye movements and catch-up saccades, and then predict target trajectory by behavioral response, eye movement or cerebral activity. Visual tracking system has to encode target’s motion features and semantic identity for completing predictive mechanism. In recent years the amount and complexity of data about processing of object’s motion features have increased substantially, whereas data about processing of object’s semantic identity is significantly less.

Key words: visual motion tracking, moving object capture, moving object tracking, prediction