I am currently a computer vision lead at ByteDance/TikTok, Singapore and also holding an appointment as an adjunct assistant professor at the National University of Singapore. Before that, I was a research fellow at the University of Oxford, and a member of Torr Vision Group (TVG) working with Prof. Philip H.S. Torr. I serve as an Associate Editor of Pattern Recognition and Guest Editor of IEEE Transactions on Pattern Analysis and Machine Intelligence.

[Hiring!] Actively recruiting scientists, engineers, research interns, and engineering interns in China and Singapore. Feel free to contact me!

[Contact Information]
For job and research related matters: songbai [dot] site [at] gmail [dot] com
For teaching related matters in NUS: song [dot] bai [at] nus [dot] edu [dot] sg

[Call for Papers!] IEEE TPAMI Special Issue on Large-Scale Multimodal Learning: Universality, Robustness, Efficiency, and Beyond. Deadline: March 1, 2023 AoE

Recent News

  • Mar 2023: I am invited as an Area Chair (AC) of NeurIPS 2023.
  • Feb 2023: two papers are accepted by CVPR 2023.
  • Feb 2023: check MOSE here: a challenging dataset for video object segmentation!
  • Jan 2023: three papers are accepted by ICLR 2023.
  • Dec 2022: one paper is accepted by TPAMI.
  • Sep 2022: I am invited as an Area Chair (AC) of CVPR 2023.
  • Sep 2022: I am appointed as an adjunct assistant professor at the National University of Singapore.
  • Jul 2022: I am co-organizing the ECCV 2022 Workshop on Multiple Object Tracking and Segmentation in Complex Environments. Welcome to participate in the 2nd Occluded Video Instance Segmentation Challenge!
  • Jul 2022: five papers are accepted by ECCV 2022.
  • Jun 2022: one paper is accepted by ACM MM 2022.
  • Jun 2022: our team ByteVIS won the 1st place on CVPR 2022 YouTube Video Instance Segmentation Challenge (check our winner solution here).
  • May 2022: “Occluded Video Instance Segmentation: A Benchmark” is accepted by IJCV.
  • Mar 2022: seven papers are accepted by CVPR 2022.
  • Oct 2021: technical report on OVIS Dataset and ICCV 2021 Challenge is accepted by NeurIPS 2021 Datasets and Benchmarks Track.
  • Sep 2021: one paper about Federated Learning for COVID-19 Diagnosis is accepted by Nature Machine Intelligence.
  • Jul 2021: I am invited as a Senior Program Committee (SPC) member of AAAI 2022.
  • Jul 2021: one paper about 3D plane recovery is accpeted by ICCV 2021.
  • Mar 2021: I am co-organizing the ICCV 2021 Workshop on Occluded Video Instance Segmentation.
  • Mar 2021: three papers are accepted by CVPR 2021.
  • Feb 2021: I am co-organizing the CVPR 2021 Workshop on Robust Video Scene Understanding: Tracking and Video Segmentation.
  • Oct 2020: one paper is accepted by TPAMI.
  • Sep 2020: one paper is accepted by Pattern Recognition.
  • Jul 2020: one paper is accepted by ACM MM 2020.
  • Jul 2020: three papers are accepted by ECCV 2020.
  • Apr 2020: I am co-organizing the ECCV 2020 Workshop on Adversarial Robustness in the Real World.
  • Feb 2020: two papers are accepted by CVPR 2020.
  • Feb 2020: I receive a gift from DeepMind to sponsor our workshop held in conjunction with CVPR 2020.
  • Feb 2020: our workshop proposal about adversarial robustness is accepted by ECCV 2020.
  • Jan 2020: I am co-organizing the CVPR 2020 Workshop on Adversarial Machine Learning in Computer Vision.
  • Dec 2019: one paper about line segment detection is accepted by TPAMI.
  • Nov 2019: one paper about transferable adversarial attack is accepted by AAAI 2020 (PDF).
  • Oct 2019: our workshop proposal about adversarial machine learning is accepted by CVPR 2020.
  • Jul 2019: seven papers are accepted by ICCV 2019.
  • May 2019: our team Oxford-CASIA achieves the 2nd place in the 2019 DAVIS Challenge on Unsupervised Video Object Segmentation.
  • Apr 2019: our special issue proposal about graph networks is accepted by TPAMI.
  • Apr 2019: one paper about object detection (CenterNet: AP 47.0 on MS-COCO test-dev) is available in arXiv.
  • Mar 2019: the CODE of “Learning Attraction Field Reprensentation for Robust Line Segment Detection” is available.
  • Feb 2019: three papers are accepted by CVPR 2019.
  • Feb 2019: the CODE of “Adversarial Metric Attack for Person Re-identification” is available.
  • Feb 2019: the CODE of “Learning Transferable Adversarial Examples via Ghost Networks” is available.
  • Jan 2019: I am appointed as an Associate Editor (AE) of Pattern Recognition.
  • Dec 2018: the CODE of “Learn to Interpret Atari Agents” is available.
  • Oct 2018: one paper about object detection is accepted by TPAMI (PDF | CODE).
  • Aug 2018: one paper about 3D multi-organ segmentation is accepted by WACV 2019 as an oral presentation (PDF).
  • Aug 2018: I join the University of Oxford as a research fellow.
  • Jul 2018: one paper about image and 3D shape retrieval is accepted by TIP (PDF).
  • Jul 2018: one paper about person re-identification is accepted by ECCV 2018 (PDF).
  • Jun 2018: one paper about object retrieval is accepted by PR (PDF).
  • May 2018: I join the Inception Institute of Artificial Intelligence as a research scientist.
  • Apr 2018: one paper about object retrieval is accepted by TPAMI (PDF | CODE).
  • Mar 2018: one paper about 3D shape recognition is accepted by CVPR 2018 (PDF).
  • Jul 2017: one paper about object retrieval is accepted by ICCV 2017 as an oral presentation (PDF).
  • Mar 2017: one paper about reranking-based person re-identification is accepted by CVPR 2017 as a spotlight presentation (PDF).
  • Feb 2017: one paper about object retrieval is accepted by AAAI 2017 as an oral presentation (PDF).
My research interests span computer vision and machine learning with a series of topics, such as graph neural networks and adversarial learning. such as 3D shape recognition, image retrieval and classification, person re-identification and medical image analysis. Before that, I received my B.E. degree and Ph.D. degree from [Huazhong University of Science and Technology](http://english.hust.edu.cn) (HUST), under the supervision of Prof. [Xiang Bai](http://122.205.5.5:8071/~xbai/). I was a research scholar at the [University of Texas at San Antonio](https://www.utsa.edu/) (UTSA) supervised by Prof. [Qi Tian](http://www.cs.utsa.edu/~qitian/), and with the [Computational Cognition, Vision, and Learning](https://ccvl.jhu.edu/) (CCVL) research group at the [Johns Hopkins University](https://www.jhu.edu/) (JHU) supervised by Prof. [Alan Yuille](http://www.cs.jhu.edu/~ayuille/). I serve as an Associate Editor of [Pattern Recognition](https://www.journals.elsevier.com/pattern-recognition). [Call for Papers] ECCV 2020 Workshop on [Adversarial Robustness in the Real World](https://eccv20-adv-workshop.github.io/). Deadline: ~~July 20, 2020 AoE~~. (Submission Closed) [Call for Papers] CVPR 2020 Workshop on [Adversarial Machine Learning in Computer Vision](https://adv-workshop-2020.github.io/). Deadline: ~~March 15, 2020 AoE~~. (Submission Closed) [Call for Papers] IEEE TPAMI Special Issue on [Graphs in Vision and Pattern Analysis](http://songbai.site/files/Call-for-Papers.pdf). Deadline: ~~October 15, 2019 AoE~~ (Submission Closed). - Do not worry if you do not recieve the decision on time. We are chasing the reviewers. [Preprint] A versatile loss function for location-sensitive recognition: [Cross-IOU Loss](https://arxiv.org/abs/2104.04899)! [Preprint] A benchmark for video instance segmentation in occluded scenes: [OVIS](https://arxiv.org/abs/2102.01558)! Its [challenge](https://competitions.codalab.org/competitions/32377) is being held in conjunction with ICCV 2021! [Call for Papers] CVPR 2021 [Workshop on Robust Video Scene Understanding: Tracking and Video Segmentation](https://eval.vision.rwth-aachen.de/rvsu-workshop21/). [Call for Papers] ICCV 2021 [Workshop on Occluded Video Instance Segmentation](https://ovis-workshop.github.io/).