最新消息

2014年7月12日学术报告:Motion perception under uncertainty,美国加州大学洛杉矶分校Hongjing Lu教授


添加时间:2014-09-02 16:11:38



时间:2014712 下午3

地点:中心教学楼902会议室
题目: Motion perception under uncertainty


报告人:Hongjing Lu (吕宏静), Associate Professor, Departments of Psychology and  Statistics, University of California, Los Angeles (UCLA)

 

Abstract
Humans perceive motion in visual scenes quickly and reliably despite the notorious correspondence problem in motion stimuli. This talk will illustrate how humans make inferences from noisy visual data in the context of motion perception.  I will present various motion phenomena that motivate the development of computational models to account for human performance.
First, I will focus on simple motion tasks, such as perceiving global motion, to show how to use computational models to provide a benchmark to assess how well humans perform in a specific visual task. The quantitative discrepancy between human performance and model predictions requires further model development, by incorporating generic priors and hierarchical generative models. Second, I will present both empirical and computational evidence for motion adaptation, and address the questions of why the motion system needs to be adaptive and how the system adapts. Finally, I will review how the human visual system analyzes complex motion patterns, such as human actions.

 



作者:宋浩