Seminar by Prof. Tong on the parametric reconstruction of the human body

Prof. Ruofeng Tong from the Computer Science Department of the Zhejiang University, China has given a presentation on the Parametric Human Body Reconstruction from Depth Image or dressed human photos


Human body reconstruction from depth image or dressed-human photos is an important but challenging problem in virtual fitting. We propose a novel automatic framework to efficiently estimate 3D body shapes from depth image or dressed-human photos. We construct a database of 3D naked and dressed body pairs, based on which we learn how to predict 3D positions of body landmarks (which further constrain a parametric human body model) automatically according to human depth image or photo silhouettes. Critical vertices are selected on 3D registered human bodies as landmarks to represent body shapes, so as to avoid the time-consuming vertices correspondences finding process for parametric body reconstruction. Our method can estimate 3D body shapes from dressed-human silhouettes within 4 seconds. Our estimation error is within the size tolerance for clothing industry.

Short bio:

Ruofeng Tong is a professor of Computer Science Department, Zhejiang University, China. He obtained his Ph.D. degree from the Applied Mathematics Department of Zhejiang University in 1996. He was a visiting researcher of Intelligent Systems and Modeling Laboratory, Faculty of Engineering, Hiroshima University, Japan from 1999 to 2001. He is currently the vice director of the State Key Lab of CAD&CG, Zhejiang University, and the vice director of Multimedia Branch of China Society of Image and Graphics. His research interests include: image and video processing, computer graphics, and virtual reality.