Huang, Tiejun


Huang, Tiejun


Research Interests: Visual information processing, neuromorphic computing

Office Phone: 86-10-6275 6541


Huang, Tiejun is the Chair of the Department of Computing Intelligence and the Director of the Institute for Visual Technology. His research areas include video coding, image recognition, and neuromorphic computing. He joined PKU on 1 June 2007; before this he was for eight years a post-doctoral fellow, then associate Professor at the Institute for Computing Technology and the Graduated School of the Chinese Academy of Sciences. Professor Huang received Ph.D. degree in pattern recognition and intelligent system from the Huazhong (Central China) University of Science and Technology in 1998, and master and bachelor degrees in computer science from Wuhan University of Technology in 1995 and 1992, respectively.

Professor Huang received the National Science Fund for Distinguished Young Scholars of China in 2014, and was awarded the Distinguished Professor of the Chang Jiang Scholars Program by the Ministry of Education of China in 2015. He is a member of the Board of the Chinese Institute of Electronics, the distinguished member of the China Computer Federation, and the Head of Delegation of China for the multimedia standardization organization MPEG (ISO/IEC JTC1 SC29/WG11).

Professor Huang is famous for his contribution to video coding and visual information processing, especially in surveillance video compression and analysis. He has gained his international recognition through highly-cited publications, standards, high profile conference organizations and editorial board services. He has published about 200 papers in various journals and conferences and drafted four ISO/IEC international standards, four IEEE standards and five national standards of China. He holds 40+ granted patents, including 3 granted in USA. Professor Huang was awarded the National Science and Technology Progress Level 2 Prize for two times. He has served the advisory board of Computing Now of the IEEE Computer Society for six years, chaired the first IEEE conference on Multimedia Big Data, and has been actively involved in organizing ICME, ICMI and ISCAS as well as other international conferences over the years.

Surveillance video big data introduces many technological challenges, including compression, storage, transmission, analysis, and recognition. Among these, the two most critical challenges are how to efficiently transmit and store the huge amount of data, and how to intelligently analyze and understand the visual information inside. Professor Huang contributed to high efficient surveillance video coding algorithm, so called background modeling based predication, which doubles the performance of the generic video coding standards on surveillance video, and visual saliency estimation and objects detection algorithms which outperform state-of-the-arts on public benchmark datasets.

As the general secretary of the AVS (Audio and Video coding Standard) working group of China from 2002 and the task force chair of the IEEE 1857 WG from 2012, Professor Huang contributed the above algorithm to the IEEE std 1857. The new standard, featured by the algorithm proposed by the nominee, outperforms the in-use AVC/ H.264 with doubling compression ratio. For the latest HEVC/H.265, the algorithm can also save ~40% bitrate and ~40% encoding complexity. Even more, featured with the background modeling algorithm, the AVS2 national standard and the upcoming IEEE 1857.4 standard double the compression ratio of the ISO/IEC standard HEVC/ITU H.265 again.

Surveillance videos contain a great deal of information, people, vehicles, and other moving objects appearing in millions of cameras are rich sources for machine analysis to understand the complicated society and world. To address the challenge of video automatic analysis and understanding, Professor Huang and his team contributed a series of algorithms on visual saliency estimation, visual object detection and recognition.All these contributions reinforced the algorithms library for surveillance video analysis and visual information processing.

The research on surveillance video is very active in the past decade. Among them, Professor Huang’s contribution to surveillance video compression and analysis is significant, which is influencing the surveillance industry more and more deeply.