MA Siwei

Name:MA Siwei


Research interests:Video Coding, Video Transmission

E-mail:swma AT pku DOT edu DOT cn

Siwei Ma


Peking University


Google Scholar:

Siwei Ma is a professor in the School of Computer Science. He obtained his B.Sc. from Shandong Normal University in 1999, and Ph.D. from Institute of Computing Technology, Chinese Academy of Sciences in 2005 respectively. His research interests include video coding, processing, and coding systems.

Prof. Siwei Ma has made significant breakthroughs in video technology. He has won one First Prize and one Second Prize of China's State Technological Invention Award, and one Second Prizes of China's State Science and Technology Award. He published over 300 papers in refereed journals and selected international conferences. These articles have been widely recognized with the citation for over 12,000 times from Google Scholar.


  • Boya distinguished professor, 2021 - present
    Shool of Computer Science, Peking University, Beijing, China

  • Full professor, 2014 - present
    Shool of EECS, Peking University, Beijing, China

  • Visiting professor, 2016 - 2017
    Department of ECE, University of Washington, Seattle, US

  • Associate professor, 2009 - 2014
    School of EECS, Peking University, Beijing, China

  • Lecturer, 2007 - 2009
    School of EECS, Peking University, Beijing, China

  • Postdoc researcher, 2005 - 2007
    University of Southern California, Los Angeles, U.S.

  • Ph.D in Computer Science, 2005
    Chinese Academy of Science (CAS), Beijing, China

  • B.S in Computer Science, 1999
    Shandong Normal University, Jinan, China

Representative Works

[1] Video Coding Theory and Models

a) Rate Distortion Model for Video Coding [IEEE TCSVT 2005, Google Scholar Citation 400+]

b) Perceptually Optimized Rate-Distortion Optimization [IEEE TCSVT 2012]

c) Entropy of Primitive for Visual Information Evaluation [IEEE TCSVT 2017]

[2] Advanced Video Coding Technology

a) Nonlocal in-loop filter for Video Coding [IEEE Multimedia 2018 Best Paper]

b) Light Field Image Compression Framework [PCM 2017 Best Paper]

c) Video Prediction Using Spatial-temporal LSTM [NeurIPS2021, CVPR2021]

[3] Intelligent Video Coding

a) Content-Aware In-loop Filter for Video Coding [IEEE TIP 2019]

b) Fast QTBT Partitioning Decision for Video Coding [ICIP 2018 Best Student Paper]

c) Scalable Auto Encoder for Learned Image Coding [MIPR 2019 Best Student Paper]

[4] Video Coding System

a) The First FPGA-Accelerated 4K UHD Neural Video Coding System [IEEE TCSVT 2022]

b) Hardware Friendly Interweaved Prediction [IEEE PCS 2021 Top10 Best Paper]


c) 8K Real-time AVS3 Video Encoding System


  • Transform/prediction based high efficiency video coding techniques

  • Perceptual quality metric and perceptual distortion optimization coding

  • 3D video coding and VR video processing


  • State Scientific and Technological Progress Award (Second Level)

  • China Standard Innovation Contribution Award (First Level)

  • IEEE 1857 International Standard Contribution Award

  • AVS Technology Award and the National Excellent Doctoral Thesis Award

  • Tencent Xplorer prize, 5 awardees in information technology in China, 2022

  • World Leading Internet Scientific and Technological Achievement, World Internet Conference Wuzhen Summit, 2021

  • NSFC Distinguished Young Scholars, 6 awardees in the area of intelligent media computing in China, 2020

  • First Prize of China's State Technological Invention, 2020

  • Grand Prize for Invention of the Institute of Electronics in China, 2019

  • NSFC Excellent Young Scientists Fund, 5 awardees in the area of intelligent media computing in China, 2013

  • Second Prize of China's State Science and Technology Progress, 2012

  • Second Prize of China's State Technological Invention, 2006

  • IEEE PCS Top 10 Paper, 2021

  • IEEE MIPR Best Student Paper, 2019

  • IEEE Multimedia Best Paper Award, 2018

  • IEEE ICIP Best Student Paper, 2018

  • VCIP top 10% best paper award, 2016

  • IEEE-SA Standard Acknowledgement (for contribution to the development of the IEEE 1857 standard), 2013

  • PCM Best Paper Award, 2017


  • Boya distinguished professor in the School of Computer Science at Peking University

  • Dean Assistant to the School of Computer Science at Peking University

  • Vice director of the National Engineering Lab on Video Technology at Peking University

  • Video group leader of Audio and Video Coding Standard (AVS) Workgroup

  • Associate editor of IEEE Transactions on Image Processing (TIP)

  • Associate editor of IEEE Transactions on Circuits and System for Video Technology (TCSVT)

  • Member of Visual Signal Processing and Communications (VSPC), IEEE Circuits and Systems Society

  • Member of Multimedia Communications Technical Committee (MMTC) in IEEE Communications Society

  • Vice Chair of IEEE Computer Society Data Compression Standards Committee

  • Chief editor of IEEE 1857.10 standard

  • Technical Program Committee member of VCIP 2017

  • Publication chair of ICIP 2017

  • Program chair of IEEE VCIP 2017

  • Area Chair of ISM 2015

  • Associate editor of Journal of Visual Communication and Representation (JVCIR)

  • Sponsorship chair of China Multimedia Conference 2022 and 2023

  • Deputy director of the Technical Committee on Multimedia of the China Society of Image and Graph (CSIG).

  • Council member of CSIG

  • Deputy director of Image, Video and Security Technical Committee of CSIG

  • Executive committee member of Multimedia Technology Technical Committee of China Computer Federation


  • He proposed series of high efficiency video coding tools, including rate-distortion optimization transform, adaptive motion vector resolution, rate distortion optimization mode decision, and non-local structure based loop filtering etc. Many of these works have been published on top journals, e.g. IEEE TCSVT and IEEE TIP, and serval techniques have been accepted by HEVC/H.265 and AVS video coding standards.

  • Perceptual coding is very promising for future video applications. But the quality metric is an important issue in perceptual coding, and it would affect the design of perceptual coding methods significantly. He has proposed a spatial-temporal structural information based video quality metric, which has lower computational complexity while approximating the human visual testing results accurately. Moreover, based on the proposed perceptual quality metrics, optimized perceptual coding was researched and the proposed perceptual coding method can achieve more 20% bits saving compared with the traditional coding methods.

  • 3D video and VR applications becomes more and popular. His research interests focus on high efficiency 3D video coding and processing, including multi-view prediction coding, texture plus depth joint optimization coding, and panorama video stitching and coding. He proposed a low complexity synthesis view distortion estimation model, which has been adopted by MPEG 3DV video coding standards. His team has developed an AVS2 based multiview broadcasting system, which can support 8 HD video streams coding simultaneously, and a 4K VR system with real time panorama video stitching and streaming.