Currently, I am a visiting scholar in the CyLab, Carnegie Mellon University, advised by Prof. Marios Savvides.
I am also a fouth-year Ph.D. student in Vision and System Design Lab , Hong Kong University of Science and Technology, supervised by Prof. Kwang-Ting Tim CHENG .
I interned in Face++ in 2019, mentored by Xiangyu Zhang , Jian Sun , and also in Tencent AI lab in 2018, mentored by Wenhan Luo , Baoyuan Wu , Wei Liu.
Before starting my Ph.D. study in HKUST in Sept. 2016, I obtained my bachelor’s degree from Fudan University in June. 2016.


Email: zliubq[at]connect[dot]ust[dot]hk; zechunl[at]andrew[dot]cmu[dot]edu
[Google Scholar] |  [Github]

Research Interest

My research interests are computer vision, machine learning. Specifically, I am interested in using deep learning to solve the practical problems in the industry such as the limitation of insufficient resources and a trade-off between computation and accuracy. My research focus is mainly on:

  • Neural architecture design and search
  • Network binarization and quantization
  • Network channel pruning
  • Meta learnning
  • Image synthesizing
  • Knowledge distillation
  • FPGA implementation of compact network


  • [July 2020] Two papers accepted to ECCV 2020.
  • [March 2020] One paper accepted to CVPR 2020.
  • [Sept 2019] I come to Canegie Mellon University as a visiting scholar.

  • Publications

    Zechun Liu, Zhiqiang Shen, Marios Savvides, Kwang-Ting Cheng.
    ReActNet: Towards Precise Binary NeuralNetwork with Generalized Activation Functions
    European Conference on Computer Vision (ECCV), 2020.
    Code & Models  |  Paper
    We achieve 65.9% (ResNet-based) and 69.5% (MobileNet-based) top-1 accuracy on ImageNet (the new results are slightly higher than those in our original paper after we fix a small loading bug), for the first time, exceeding the benchmarking ResNet-level accuracy (69.3%) while achieving more than 22× reduction in computational complexity.

    Zichao Guo, Xiangyu Zhang, Haoyuan Mu, Wen Heng, Zechun Liu, Yichen Wei, Jian Sun.
    Single Path One-Shot Neural Architecture Search with Uniform Sampling
    European Conference on Computer Vision (ECCV), 2020.
    Code & Models  |  Paper

    Hai Phan*, Zechun Liu*, Dang Huynh, Marios Savvides, Kwang-Ting Cheng, Zhiqiang Shen.(*:Equal contribution)
    Binarizing MobileNet via Evolution-based Searching
    IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020.

    Zechun Liu, Haoyuan Mu, Xiangyu Zhang, Zichao Guo, Xin Yang, Kwang-Ting Cheng, Jian Sun.
    MetaPruning: Meta Learning for Automatic Neural Network Channel Pruning
    IEEE International Conference on Computer Vision (ICCV), 2019.
    Code & Models  |  Paper

    Zechun Liu, Wenhan Luo, Baoyuan Wu, Xin Yang, Wei Liu, Kwang-Ting Cheng.
    Bi-Real Net: Binarizing Deep Network Towards Real-Network Performance
    International Journal of Computer Vision (IJCV).
    Code & Models  |  Paper

    Zechun Liu, Baoyuan Wu, Wenhan Luo, Xin Yang, Wei Liu, Kwang-Ting Cheng.
    Bi-Real Net: Enhancing the Performance of 1-bit CNNs with Improved Representational Capability and Advanced Training Algorithm
    European Conference on Computer Vision (ECCV), 2018.
    Code & Models  |  Paper

    Koen Helwegen, James Widdicombe, Lukas Geiger, Zechun Liu, Kwang-Ting Cheng, Roeland Nusselder.
    Latent Weights Do Not Exist: Rethinking Binarized Neural Network Optimization
    Advances in Neural Information Processing Systems 32 pre-proceedings (NIPS), 2019.
    Code & Models  |  Paper

    Awards and Honors

    • Postgraduate Studentship, HKUST 2016-2020
    • Oversea Research Award, HKUST 2019
    • Travel Award NeurIPS 2019, ICCV 2019, ECCV 2018
    • Shanghai Outstanding Graduate, Fudan University 2016
    • Chinese National Scholarship, Fudan University 2012-2013 & 2013-2014

    Teaching Assistant

    • 2017.2 - 2016.6, HKUST, ELEC 2300 Computer Organization
    • 2017.9 - 2018.1, HKUST, ELEC 4010K Machine Learning and Information Processing for Robotic Perception
    • 2017.2 - 2016.6, HKUST, ELEC 2300 Computer Organization