博士论坛

博士论坛第七十六期学术报告信息

作者:admin 日期:2014-09-26 20:15:00 浏览:

报告题目:Salient Motion Detection using   Potential Surface

报告人:京都大学 梁雪峰副教授

 

报告简介:

A challenge   in motion segmentation is that different motions are often mixed up and   interdependent in the real data. Since the 2D representation of the dependent   motions is obscure, it leads the segmentation difficult. In this talk, we   will discuss a motion segmentation and recovery method for complex scenarios.   Using an invariant interpretation of the motion vector, we first transform   the 2D motion vector field into the 3D potential surface. In which, different   motions are placed onto different layers so that they can be segmented much   easier. By applying the surface fitting, the potential surfaces of global and   local motions are then estimated. Finally, the recovered motions are obtained   by projecting the segmented potential surfaces back to motion field. Using   potential surface makes our method able to deal with both independent and   dependent, rigid and non-rigid motions without prior knowledge. We will also   explore its application in video alignment and action recognition.

 

报告人简介:

Xuefeng Liang is an associate professor in the Department of intelligence   Science of Informatics, Kyoto University. He was awarded a Ph.D. degree in   information science from Japan Advanced Institute of Science and Technology,   Japan, in 2006. After that, he worked on the vision system in robotics as a   postdoc in Ubiquitous Functions Research Group, ISRI, National Institute of   Advanced Industrial Science and Technology, Tsukuba, Japan. In 2008, he moved   to London and worked on vision perception study at both Vision Lab, Queen   Mary University of London and Department of Psychology, University College   London UK. In 2010, he was appointed associate professor in Graduate School   of Informatics at Kyoto University. His interests include computer vision,   pattern recognition, image processing, and computational geometry.

 

时间:2014年9月26日(星期五)上午9:30

地点:武汉大学信息学部教学实验大楼4楼会议室