Computer Vision & Media Computing

Vision is a complex process that can understand the physical environment for different tasks in a fast and efficient way. Computer vision studies programmable algorithms that assign intelligence to computers such that computers can process and understand images as human vision does. While computer vision usually solves the problems of moving from 2D to 3D, computer graphics studies problems in an inverse way, i.e., moving from 3D to 2D. In between computer vision and computer graphics, media encoding and decoding plays an important role in machine vision, since natural scenes usually contain millions of bits of information and neuron encoding is inevitable for a fast response in human vision system. The area of computer vision and media computing in the Department of Computer Science and Technology consists of nine faculty members who cover all the areas of computer vision, computer graphics, image/video coding and transmission, and image/video processing.

This area has been funded by the National Basic Research Program of China, the National High Technology Research and Development Program of China, the Natural Science Foundation of China (NSFC), and international cooperation research programs. In particular, we are leading a project of the National Basic Research Program entitled “Intelligent Processing of Visual Media” which is funded to about RMB 24M (the first term from 2006 to 2010) and RMB 30M (the second term from 2010 to 2015), and a key project of the NSFC entitled “Theories and Methodologies for Web Social Network Analysis” which is co-funded by NSFC and Microsoft.

We have published a number of high-quality papers in foremost peer-reviewed international journals and the top international conferences, including ACM SIGGRAPH (4), CVPR oral (2), ICCV oral (1), IEEE T-PAMI (4), IEEE T-VCG (8), IEEE T-CSVT (1). Among them one paper was awarded IEEE T-CSVT Best Paper in 2010, and one was awarded CVPR Best Student Paper in 2007. We also have more than 20 patents authorized in China.

In this area, many outstanding postgraduates have been trained and a number of high-level awards have been achieved, including one National Excellent Doctoral Dissertation Award, two China Computer Federation Excellent Doctoral Dissertation Awards, and one Beijing Excellent Doctoral Dissertation Award. One female master student received the CVPR Best Student Paper Award in 2007, and one female PhD student received the IEEE T-CSVT Best Paper Award in 2010.

Faculty members also devote academic services to the research community, for example by serving as chair or co-chair, or editor-in-chief or associate editor of various high-level international conferences, journals, and organizations. In particular, Professor Shi-Min Hu is on the editorial board of Computer Aided Design and was elected as the chair of the Executive Committee of the ACM Solid Modeling Association in 2011-2012.