I'm a Professor in the School of Computing at ANU. I also work part-time as a Staff Researcher at Canva. My research areas are computer vision and generative AI.
I'm best known for my contributions to object re-identification, where I published widely used datasets (Market-1501, MARS, and PRW) and baselines (IDE and PCB). My works extended to object tracking and general recognition, where my representative works are JDE and Random Erasing. Recently, my group contributed to well-known works in generative AI, such as REPA-E, iREPA, R2E-Gym, and SPO. REPA-E and iREPA have been proven useful for commercial-level models like Flux.2 and Krea 2, which provide very good representations for generation.
I received an ARC DECRA Fellowship (2020-2023), an ARC Future Fellowship (2025-2029), and the inaugural Brian Anderson Medal from the Australian Academy of Science. I was named one of the "AI's 10 to Watch" by IEEE Intelligent Systems.
I joined ANU in 2018 as a Lecturer. I was promoted to Senior Lecturer, Associate Professor, and Professor in 2021, 2024, and 2026, respectively. Before ANU, I was a Postdoc at UTS working with Prof. Yi Yang and a Postdoc at UTSA working with Prof. Qi Tian. I obtained my Bachelor degree from the School of Life Science at Tsinghua University in 2010, and my PhD degree from the Department of Electronic Engineering at Tsinghua University in 2015, supervised by Prof. Shengjin Wang.
Prospective PhD students, postdocs, interns, and visitors: If you are interested in joining my group, please email me with your CV and a research statement.
News
I will serve as a General Co-Chair for IEEE International Conference on Advanced Visual and Signal-Based Systems (AVSS 2027), Macau, China
We have released the PersonX engine! It has several subsets to evaluate standard re-id and domain adaptive re-id. But most importantly, it allows us to generate datasets freely!
[Call for paper] IEEE International Conference on Multimedia and Expo (ICME) 2019 Special Session on "Multimedia Technologies Empowering Retail Experiences" [Deadline: 17 December 2018] [URL]
We have released the code for our paper "Generalizing A Person Retrieval Model Hetero- and Homogeneously". Link
We have released the code for our paper "Beyond Part Models: Person Retrieval with Refined Part Pooling (and A Strong Convolutional Baseline)". Link
SIFT Meets CNN: A Decade Survey of Instance Retrieval. TPAMI, Accepted. [PDF][Bibtex]
Professional Service
General Co-Chair: AVSS 2027
Program Co-Chair: AVSS 2024
Program Co-Chair: ACM Multimedia 2024
Workshop Co-organizer: DataCV Workshop and challenge 2022-2025
Associate Editor: IEEE Transactions on Circuits and Systems for Video Technology (2024- )
Associate Editor: ACM Computing Survey (2024- )
Lead Area Chair: ICCV 2025, CVPR 2025
Area Chair: CVPR 2021, 2023, 2024, 2025, 2026
Area Chair: ACM Multimedia 2020, 2021
Area Chair: NeurIPS 2024, 2025
Area Chair: NeurIPS 2024, 2025
Program Chair: IEEE Machine Learning for Signal Processing Workshop, 2021
Area Chair: ECCV 2020, 2022, 2024
Area Chair: ICCV 2023, 2025
Associate Editor: IEEE Transactions on Circuits and Systems for Video Technology (2021-2023)
Senior PC: IJCAI 2019, 2020, AAAI 2020, 2022
Organizer: CVPR 2020-2023 workshop on "2020 AI CITY CHALLENGE"
CVPR 2019 tutorial: "Textures, Objects, Scenes: From Handcrafted Features to CNNs and Beyond"
Organizer: CVPR 2019 workshop on "Target Re-Identification and Multi-Target Multi-Camera Tracking"
Area Chair, ICMR 2019
Associate Editor, Visual Computer Journal
Area Chair, ICPR 2018
ECCV 2018 Tutorial: "Representation Learning in Pedestrian Re-identification".
ICPR 2018 Tutorial: "Person Re-identification: State of the Art and Future Trend".
