PeopleofSuperintelligence

Researcher

Chao Xu

Chao Xu is a computer vision researcher at Peking University whose work spans image processing, multi-view learning and neural architecture search. He earned degrees in optical instruments, optics and image processing at Tsinghua University, the University of Science and Technology of China, and the Chinese Academy of Sciences. He taught at the University of Science and Technology of China’s Beijing Graduate School and began teaching at PKU’s Information Science Center in 1997. His papers explore student-teacher networks, multi-view learning and neural architecture search, and include image-processing transformers (2021) and polarized-image deblurring (2025). He has been a professor at Peking University’s Information Science Center since 2005.

Badges

Timeline

  1. 1988–1997

    Degrees in optics and image processing

    In 1988, he earns a bachelor's degree in optical instruments at Tsinghua University; in 1991, a master's in optics at the University of Science and Technology of China; in 1997, a PhD in image processing at the Chinese Academy of Sciences.

  2. 1991–1997

    Teaching at two universities

    From 1991 to 1994, he serves as a teaching assistant at the Beijing Graduate School of the University of Science and Technology of China; in 1997, he begins teaching at Peking University's Information Science Center.

  3. 2017

    Learning from multiple teachers

    At KDD 2017, he coauthors Learning from Multiple Teacher Networks, a method that trains a thin student network using several teacher networks.

  4. 2019

    Learning from multiple views

    In 2019, he posts an arXiv version of Multi-View Intact Space Learning with Chang Xu and Dacheng Tao; its MISL algorithm combines complementary information from multiple views to learn a latent representation.

  5. 2019

    Searching neural architectures

    In 2019, he coauthors CARS, a continuous evolutionary method that searches neural-network architectures by sharing parameters within a SuperNet.

  6. 2021

    A pre-trained image transformer

    In 2021, he coauthors Pre-Trained Image Processing Transformer (IPT), a pre-trained model for low-level computer-vision tasks including denoising, super-resolution and deraining.

  7. 2025

    Deblurring polarized images

    In 2025, he coauthors Learning to Deblur Polarized Images in the International Journal of Computer Vision; its pipeline uses a polarization-aware divide-and-conquer strategy.

  8. Since 2005

    Professor at Peking University

    He serves as a professor at Peking University's Information Science Center; he has taught there since 1997.

Career

  1. Since 1997Peking UniversityProfessor
  2. 1991–1994University of Science and Technology of ChinaTeaching Assistant

Education

  1. 1997Chinese Academy of SciencesPhD
  2. 1991University of Science and Technology of ChinaMaster's degree
  3. 1988Tsinghua UniversityBachelor's degree

From the biography.

  1. The median h-index is 48 at universities and 14 at frontier AI labsIncludes Peking University. The median h-index is 48 at universities and 14 at frontier AI labs. We compare publishing careers, recent citations, dual appointments and job moves across 2,420 university researchers and 308 frontier-lab researchers.
  2. The company share of AI researchers’ top papers was 31% in 2020–2024 and 8% in 1970–1974Includes Peking University. We estimate where the people shaping AI worked when their most-cited papers appeared: the company share was 31% in 2020–2024 and 8% in 1970–1974, and Google led all organizations in 2020–2024.
  3. Companies supply the most prior-work hires at 7 of the 9 frontier AI labsIncludes Peking University. Our estimates of the organizations where 550 people worked directly before joining 9 frontier AI labs across 620 career transitions, comparing the main feeder organizations per lab.
  4. More people shaping AI studied at Massachusetts Institute of Technology than at any other institutionIncludes Peking University. We estimate where the people shaping AI studied and where their careers led. Massachusetts Institute of Technology leads with 467 former students, including 291 recorded doctorates.

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Chao Xu yet. Send us a free-licensed photo.

  1. Chao Xu -北京大学人工智能研究院 ai.pku.edu.cn. Checked 21 October 2026.
  2. Learning from Multiple Teacher Networks kdd.org. Checked 21 October 2026.
  3. Multi-View Intact Space Learning arxiv.org. Checked 21 October 2026.
  4. CARS: Continuous Evolution for Efficient Neural Architecture Search arxiv.org. Checked 21 October 2026.
  5. Pre-Trained Image Processing Transformer openaccess.thecvf.com. Checked 21 October 2026.
  6. Learning to Deblur Polarized Images github.com. Checked 21 October 2026.