Speaker: Dr. Minglun Gong, Department of Computing Science, Memorial University.
This talk discusses the problem of foreground separation from the background modeling perspective. In particular, we deal with the difficult scenarios where the background texture might change spatially and temporally. A novel approach is proposed that incorporates a pixel-based online learning method to adapt to temporal background changes promptly, together with a graph cuts method to propagate per-pixel evaluation results over nearby pixels. Empirical experiments on a variety of datasets demonstrate the competitiveness of the proposed approach, which is also able to work in real-time on the Graphics Processing Unit (
GPU) of programmable graphics cards.
About the speaker: Minglun Gong obtained his Ph.D. from University of Alberta (Edmonton, Canada) in 2003, his M.Sc. from Tsinghua University (Beijing, China) in 1997, and his B.Engr. from Harbin Engineering University (Harbin, China) in 1994. After graduation, he was a faculty member at Laurentian University (Sudbury, Canada) for four years before joined the Memorial University of Newfoundland (St. John’s, Canada) in 2007. Dr. Gongâs research interests include static and dynamic image-based rendering, general-purpose computation on GPUs, image morphing, image segmentation and restoration, binocular and multi-view stereo vision, and motion estimation. He also contributes to research on different optimization techniques, such as dynamic programming and
genetic algorithms. He has published over 40 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.
If you are planning to attend this seminar, please contact a SHARCNET staff member at your local institution as AG rooms will only be opened for these seminars where someone has indicated they plan to attend.