【学术报告】Real Time Detection of Moving Foreground Objects

发稿时间:2010-03-01浏览次数:1089

Title: Real Time Detection of Moving Foreground Objects
 
Speaker: Prof. Minglun Gong, Memorial University of Newfoundland
 
Duration: 2010年3月3日,3:00-4:00 pm
Location: 311 Room, MMW
 
Abstract:
This talk discusses the problem of detecting moving foreground objects from
live video sequences. In particular, we deal with the difficult scenarios
where the background texture might change spatially and temporally. Three
algorithms are presented, which incrementally improve the accuracy of the
detection. All three algorithms utilize support vector machine based
background models, which are trained online to adapt to temporal background
changes. The algorithms are also designed for efficient parallel execution
on the Graphics Processing Units and achieve real-time processing speed on
middle class graphics cards. Empirical experiments on a variety of datasets
demonstrate the competitiveness of the proposed approach.
 
Bio:
Minglun Gong obtained his Ph.D. from University of Alberta in 2003, his
M.Sc. from Tsinghua University in 1997, and his B.Engr. from Harbin
Engineering University in 1994. After graduation, he was a faculty member at
Laurentian University for four years before joined the Memorial University
of Newfoundland in 2007. Minglun’s research interests include a variety of
topics in computer graphics, computer vision, image processing, pattern
recognition, and optimization techniques. He has published over 50 technical
papers in refereed journals and conference proceedings and served as program
committee member and reviewer for international journals and conferences. He
has been a member of IEEE and ACM since 2004.