My Portrait            

Yixiao Ge (葛艺潇)

Ph.D. Candidate,
Multimedia Lab (MMLab),
The Chinese University of Hong Kong (CUHK)

  yxge@link.cuhk.edu.hk
  geyixiao831@gmail.com
   (+852) 90611750

About Me

I am currently a Ph.D. student in Multimedia Lab (MMLab), the Chinese University of Hong Kong (CUHK). I am supervised by Prof. Hongsheng Li and Prof. Xiaogang Wang. Before that, I received bachelor's degree in Measurement and Control Technology and Instrument at Huazhong University of Science and Technology (HUST) in 2017. My research interests include computer vision, deep learning, with focus on unsupervised learning, disentangled learning, domain adaptation, image retrieval and image generation.

News

  • alt text One paper accepted by NeurIPS 2020
  • We achieved 2nd place in VisDA-2020 Challenge (ECCVW)
    One paper accepted by ECCV 2020 as a spotlight presentation
    The OpenUnReID codebase is online
    One paper accepted by ICLR 2020
    One paper accepted by NeurIPS 2018

Codebase

OpenUnReID: An Open-source Toolbox for Unsupervised or Domain Adaptive Object Re-ID

OpenUnReID Development Team

July 2020

Publications

Self-paced Contrastive Learning with Hybrid Memory for Domain Adaptive Object Re-ID

Yixiao Ge, Dapeng Chen, Feng Zhu, Rui Zhao, Hongsheng Li

Conference on Neural Information Processing Systems (NeurIPS), 2020

Self-supervising Fine-grained Region Similarities for Large-scale Image Localization

Yixiao Ge, Haibo Wang, Feng Zhu, Rui Zhao, Hongsheng Li

European Conference on Computer Vision (ECCV), 2020, Spotlight (top-5%)

Mutual Mean-Teaching: Pseudo Label Refinery for Unsupervised Domain Adaptation on Person Re-identification

Yixiao Ge, Dapeng Chen, Hongsheng Li

International Conference on Learning Representations (ICLR), 2020

FD-GAN: Pose-guided Feature Distilling GAN for Robust Person Re-identification

Yixiao Ge*, Zhuowan Li*, Haiyu Zhao, Guojun Yin, Shuai Yi, Xiaogang Wang, Hongsheng Li

Conference on Neural Information Processing Systems (NeurIPS), 2018

Priprints

Improved Mutual Mean-Teaching for Unsupervised Domain Adaptive Re-ID

Yixiao Ge, Shijie Yu, Dapeng Chen

Technique report of top-performing solution to VisDA Challenge (ECCVW), 2020

Structured Domain Adaptation with Online Relation Regularization for Unsupervised Person Re-ID

Yixiao Ge, Feng Zhu, Rui Zhao, Hongsheng Li

arXiv preprint, Mar 2020 (v1), June 2020 (v2), (In Submission)

Working Experience

Professional Activities

  • Journal Reviewer of IJCV, IEEE TIP, IEEE TCSVT, IEEE TMM, Neurocomputing.
  • Conference Reviewer of NeurIPS 2020, ICLR 2021, CVPR 2021.

Invited Talks

  • [2020-08-24] Spotlight presentation in ECCV 2020, see here for records.
  • [2020-08-23] Presentation in VisDA Challenge 2020 (ECCVW) as one of the winners, see here for records.

Teaching Experience

I have been a teaching assistant for the following courses at CUHK:
  • ENGG 2720, Complex Analysis, Fall 2020.
  • ENGG 2420B, Complex Analysis and Differential Equations for Engineers, Fall 2019.
  • ELEG 5491, Introduction to Deep Learning, Spring 2019.
  • ENGG 2420A, Complex Analysis and Differential Equations for Engineers, Fall 2018.