| Lecturer Profile |
Guoliang Fan received the B.S. degree in Automation Engineering from Xi'an University of Technology, Xi'an, China, in 1993, the M.S. degree in Computer Engineering from Xidian University, Xi'an, China, in 1996, and the Ph.D. degree in Electrical Engineering from the University of Delaware, Newark, DE, in 2001. Since 2001, Dr. Fan has been with the School of Electrical and Computer Engineering at Oklahoma State University (OSU), Stillwater, OK, where currently he is a Professor and the Cal & Marilyn Vogt Professor of Engineering. Dr. Fan is directing the Visual Computing and Image Processing Laboratory (VCIPL) at OSU. His research interests include signal processing, image processing, machine learning, pattern recognition, and computer vision. Dr. Fan is a recipient of the 2004 National Science Foundation (NSF) CAREER award. He received the Halliburton Excellent Young Teacher Award in 2004, the Halliburton Outstanding Young Faculty Award in 2006 and the Teaching Excellence Award in 2015 from the College of Engineering at OSU, and the Outstanding Professor Award from IEEE-OSU in 2008 and 2011. He received the Young Alumni Achievement Award from the Department of Electrical and Computer Engineering at University of Delaware in 2015. He is an associate editor of IEEE Transactions on Image Processing, IEEE Journal of Biomedical and Health Informatics and EURASIP Journal on Image and Video Processing. Dr. Fan is a senior member of IEEE. |
| Lecture Abstract |
Human pose estimation is an important research topic in the field of computer vision and pattern recognition, which has been actively studied for decades. In this talk, we propose a new generative framework for 3D human pose estimation that is able to operate on both individual point sets and sequential depth data. We formulate human pose estimation as a point set registration problem, where we propose three new approaches to address several major technical challenges in this research. First, we integrate two registration techniques that have a complementary nature to cope with non-rigid and articulated deformations of the human body under a variety of poses. This unique combination allows us to handle point sets of complex body shapes and large pose variation without any initial conditions, as required by most existing approaches. Second, we introduce an efficient pose tracking strategy to deal with sequential depth data, where the major challenge is the incomplete data due to self-occlusions and view changes. We introduce a visible point extraction method to initialize a new template for the current frame from the previous frame, which effectively reduces the ambiguity and uncertainty during registration. Third, to support robust and stable pose tracking, we develop a segment volume validation technique to detect tracking failures and to re-initialize pose registration if needed. The experimental results on both benchmark 3D laser scan and depth datasets demonstrate the effectiveness of the proposed framework when compared with state-of-the-art algorithms. |
