Yifan Sun

Computer Vision


I am currently a Senior Expert at Baidu Inc. (since 2021). Before joining Baidu, I was a senior researcher at MegVii Inc. and an optical remote sensing institute, respectively. I received the B.E. degree, M.S. (Optical Engineering) degree and Ph.D. (Electronics Engineering) degree from Tsinghua University. My research interests include person re-identification, deep metric learning, long-tailed visual recognition, domain adaptation and generalization. [Google Scholar]


  • [Feb 2023] Three papers are accepted to CVPR 2023.
  • [Dec 2022] We conduct a series of explorations on the visual prompting mechanism and share our findings through NeurIPS 2022 (1 paper), ICLR 2023 (2 papers) and CVPR 2023 (1 paper on few-shot classification).
  • [Oct 2022] Two papers on Deep Metric Learning applications (Image Edited Copy Detection and Person Re-ID) are accepted to AAAI 2023.
  • [Jul 2022] Our paper exploring a novel foundation model paradigm named UFO (Unified Feature Optimization) is accepted to ECCV 2022. [project]
  • [Jun 2022] We win the 1st prize in eBay eProduct Visual Search Challenge (FGVC9 on CVPR'22). [pdf] [code]
  • [Feb 2022] Two papers on person re-ID and weakly-supervised domain adaptation are accepted to IEEE CVPR, 2022.
  • [Jan 2022] One paper on few-shot learning is accepted to ICLR, 2022.
  • [Dec 2021] Two papers are accepted to AAAI, 2022.
  • [Nov 2021] We win the 1st prize (matching track) and the 3rd prize (descriptor track) in the Facebook AI Image Similarity Challenge in NeurIPS, 2021. [pdf] [code]
  • [Sep 2021] One paper on model smoothing mechanism is accepted to NeurIPS, 2021. [pdf] [code]
  • [Mar 2021] One paper on deep metric learning is accepted to IEEE CVPR, 2021.
  • [Mar 2020] Two papers (circle loss and feature cloud) are accepted to IEEE CVPR, 2020.
  • [Jun 2019] One paper is accepted to IEEE TPAMI, 2019.
  • [Mar 2019] Two papers are accepted to CVPR, 2019.
  • [Jul 2018] One paper is accepted to ECCV, 2018.
  • [Jul 2018] Our paper wins Best Scientific Paper Award in ICPR, 2018.
  • [Jul 2017] One paper is accepted to IEEE ICCV, 2017.
  • [Jul 2016] One paper is accepted to ECCV, 2016.

Selected Publications

  • Spatial Ensemble: A Novel Model Smoothing Mechanism for Student-Teacher Framework.
    Tengteng Huang, Yifan Sun, Xun Wang, Haotian Yao, Chi Zhang
    NeurIPS, 2021 [Paper]
  • Dynamic Metric Learning: Towards a Scalable Metric Space to Accommodate Multiple Semantic Scales
    Yifan Sun, Yuke Zhu, Yuhan Zhang, Pengkun Zheng, Xi Qiu, Chi Zhang, Yichen Wei
    CVPR, 2021 [Paper]
  • Circle loss: A unified perspective of pair similarity optimization
    Yifan Sun, Changmao Cheng, Yuhan Zhang, Chi Zhang, Liang Zheng, Zhongdao Wang, Yichen Wei
    CVPR, 2020 (Oral) [Paper]
  • Deep Representation Learning on Long-tailed Data: A Learnable Embedding Augmentation Perspective.
    Jialun Liu*, Yifan Sun*, Chuchu Han, Zhaopeng Dou, Wenhui Li (* denotes equal contribution)
    CVPR, 2020 [Paper]
  • Learning part-based convolutional features for person re-identification
    Yifan Sun, Liang Zheng, Yali Li, Yi Yang, Qi Tian, Shengjin Wang
    IEEE TPAMI, 2019 [Paper]
  • Perceive where to focus: Learning visibility-aware part-level features for partial person re-identification
    Yifan Sun, Qin Xu, Yali Li, Chi Zhang, Yikang Li, Shengjin Wang, Jian Sun
    CVPR, 2019 [Paper]
  • Beyond part models: Person retrieval with refined part pooling (and a strong convolutional baseline)
    Yifan Sun, Liang Zheng, Yi Yang, Qi Tian, Shengjin Wang
    ECCV, 2018 [Paper]
  • Attend and align: Improving deep representations with feature alignment layer for person retrieval
    Qin Xu, Yifan Sun, Yali Li, Shengjin Wang
    ICPR, 2018 (Best Scientific Paper Award) [Paper]
  • SVDNet for Pedestrian Retrieva
    Yifan Sun, Liang Zheng, Weijian Deng, Shengjin Wang
    ICCV, 2017 (spotlight) [Paper]
  • Read more on Google Scholar