Title: Human Tracking, Identification and Activity Detection from Multiple Wearable Camera Videos
Speaker: Professor Wang Song
Date &Time: May 16, 2016 (Monday) 3:00-4:30 PM
Location: Conference room in School of Information Engineering
Profile of Speaker:
Professor Wang Song received his bachelor's degree and master's degree from Tsinghua University in 1994 and 1998, and received hisPh.D. degree from the Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, in 2002. Since then, He joined the Department of Computer Science and Engineering, University of South Carolina, where he is currently working as a professor. His research is focused on image processing, computer vision and machine learning. He has been working in a laboratory which has undertaken many research projects for U.S. National Natural Science Foundation, the air force scientific research bureau, and the advanced defense research program. Professor Wang Song is the chairman of PAMI Technical Committee of the IEEE Computer Society, associate editor of Pattern Recognition Letters, guest editor of Computer Vision and Image Understanding, and member of IEEE and IEEE Computer Society.
LectureAbstract:
Video activity detection has wide applications in public security and our daily life. The videos collected by fixed cameras can only cover limited spatial areas from pre-specified view angles. In recent years, the rapid advancement of the wearable-camera technology, such as GoPro and Google Glass, provides a new perspective for wider-range video activity detection, as well as reflecting the wearers’ intelligence is capturing the activities of interest. However, the continuous motion and the possible sudden view-angle change of the wearable cameras also bring new challenges to video activity detection. In this talk, Prof. Wang will introduce new problems emerging from the use of multiple wearable cameras for video activity detection. More specifically, He will discuss about the new problems and possible solutions in crowded human tracking over temporal intervals, cross-video person identification, and the collaborative human activity detection from multiple wearable-camera videos.
