Kai Han

Kai Han

韩锴

Assistant Professor

Visual AI Lab
School of Computing and Data Science
The University of Hong Kong

   Room 220, Run Run Shaw Building
   Pokfulam Road, Hong Kong
   kaihanx at hku.hk

𝕏 WeChat QR Code

I am an Assistant Professor at The University of Hong Kong, where I direct the Visual AI Lab. My research interests lie in computer vision, machine learning, and artificial intelligence. My current research focuses on spatial intelligence, open-world learning, foundation models, generative AI, and their relevant fields. The overarching goal of my research is to achieve principled and comprehensive visual understanding, close the intelligence gap between machines and humans, and build reliable AI systems for open-world use. Previously, I was a Visiting Faculty Researcher at Google Research, an Assistant Professor in the Department of Computer Science at University of Bristol, and a Postdoctoral Researcher in VGG at the University of Oxford, working with Prof. Andrew Zisserman and Prof. Andrea Vedaldi. I received my Ph.D. from the Department of Computer Science at The University of Hong Kong, advised by Prof. Kenneth K.Y. Wong. During my Ph.D., I also worked with Prof. Jean Ponce and Prof. Minsu Cho at the WILLOW team of Inria and École Normale Supérieure (ENS).

Openings

Interested? Send me your resume and transcripts.

News

July 2026 Will serve as Associate Editor for IEEE Robotics and Automation Letters (RA-L).
June 2026 Will serve as Action Editor for Transactions on Machine Learning Research (TMLR).
June 2026 Four papers (Spatial Panoptic Captioning, JoVA, Physical Simulation, Category Discovery) are accepted to ECCV 2026.
Apr 2026 Four papers (PartCo, Geometric Reciprocity, iVGR, ZeroBench) are accepted to ICML 2026.
Apr 2026 ZeroBench is covered by Meta's Muse Spark.
Apr 2026 Two papers (Infinite Ladder; CodeBind to Findings) are accepted to ACL 2026.
Mar 2026 One paper (AI in Oral Health Surveillance) is accepted to Journal of Dental Research (JDR).
Feb 2026 Three papers (Sculpt4D; Speed3R and Scene-Level Heterogeneous Physics to Findings) are accepted to CVPR 2026.
Dec 2025 Invited to serve as Area Chair for ECCV 2026.
Nov 2025 One paper (Deepfake Detection with Graph Neural Network) is accepted to KDD 2026.
Nov 2025 Two papers (LooC, OnlineVPO) are accepted to WACV 2026.
Nov 2025 One paper on semantic correspondence is accepted to TPAMI.
Sept 2025 Seven papers (Panoptic Captioning, 3DRS, Fin3R, Wukong, VaMP, SEAL, GSPN-2) are accepted to NeurIPS 2025.
Sept 2025 Invited talks at University of Cambridge and University of Birmingham in the UK.
Aug 2025 Invited to serve as: Area Chair for CVPR 2026, Area Chair for ICLR 2026.
July 2025 ZeroBench is covered by Google's Gemini 2.5 Report.
June 2025 Two papers (Inpaint4Drag, GRAB) are accepted to ICCV 2025.
June 2025 Talks @ CVPR 2025 in Nashville: Lightning talk at CVPR Area Chair Workshop; Keynote talks at CVPR Workshops on Fine-Grained Visual Categorization, Visual Anomaly and Novelty Detection, and Domain Generalization.
June 2025 Invited to serve as Area Chair for AAAI 2026.
May 2025 Two papers (GAMEBot, PruneVid) are accepted to ACL 2025.
Mar 2025 Splat4D is accepted to SIGGRAPH 2025.
Feb 2025 Six papers (ICE, HypCD, Mr. DETR, v-CLR, GSPN, PASS) are accepted to CVPR 2025.
Jan 2025 Five papers (HiLo, DebGCD, BiGR, Needle Threading, AvatarGO) are accepted to ICLR 2025.
Sept 2024 SciFIBench is accepted to NeurIPS 2024.
Sept 2024 Invited to serve as: Area Chair for ICLR 2025, Area Chair for CVPR 2025.
Aug 2024 Our Dissecting OOD and OSR paper is accepted to IJCV.
Jul 2024 Three papers (RegionDrag, PromptCCD, and ConceptExpress) are accepted to ECCV 2024.
Feb 2024 Three papers (IBD-SLAM, DreamAvatar, and SD4Match) are accepted to CVPR 2024.
Feb 2024 CiPR is accepted to TMLR 2024.
Jan 2024 Two papers (SPTNet and FROSTER) are accepted to ICLR 2024.
Oct 2023 Invited to serve as an Area Chair for ECCV 2024.
Sept 2023 One paper on text-guided 3D head avatar generation and editing is accepted to NeurIPS 2023.
Aug 2023 One paper on visual correspondence is accepted to TPAMI.
July 2023 Two papers (on generalized category discovery/open-vocabulary semantic segmentation) are accepted to ICCV 2023.
June 2023 Invited to serve as an Area Chair for CVPR 2024.
Mar 2023 OOD-CV workshop @ ICCV 2023. Welcome participants from all over!
Feb 2023 Two papers (on compositional zero-shot learning/3D human digitization) are accepted to CVPR 2023.
Jul 2022 One paper on novel category discovery without forgetting is accepted to ECCV 2022.
Jun 2022 Best Paper Runner-Up Award at CVPR 2022 Workshop on Continual Learning in Computer Vision.
Mar 2022 Three papers (about generalized category discovery/3D human digitization/instance segmentation) are accepted to CVPR 2022.
Jan 2022 One paper about open-set recognition is accepted to ICLR 2022.
Oct 2021 One paper about visual correspondence is accepted to BMVC 2021.
Sept 2021 One paper about novel category discovery is accepted to NeurIPS 2021.
Sept 2021 Recognized as an Outstanding Reviewer for ICCV 2021, in the top 5% of experienced reviewers.
July 2021 One paper about single- and multi-modal novel category discovery is accepted to ICCV 2021.
June 2021 Our AutoNovel paper is accepted to TPAMI.
June 2021 Recognized as an Outstanding Reviewer for CVPR 2021.
May 2021 One paper about dynamic convolution for semantic scene completion is accepted to TPAMI.
Mar 2021 One paper about long-tailed recognition is accepted to CVPR 2021.
Dec 2020 One paper about mirror surface reconstruction is accepted to IEEE TIP.
Sept 2020 One paper about dense correspondence is accepted to NeurIPS 2020.
Jun 2020 One paper about deep photometric stereo is accepted to TPAMI.
June 2020 Recognized as an Outstanding Reviewer for CVPR 2020.
Feb 2020 Two papers (about semantic correspondence / 3D semantic scene completion) are accepted to CVPR 2020.
Dec 2019 One paper about novel category discovery is accepted to ICLR 2020.
Jul 2019 One paper about novel category discovery is accepted to ICCV 2019.
Jul 2019 One paper about transparent object matting is accepted to IJCV.
Mar 2019 Two papers (about unsupervised object discovery and matching / uncalibrated photometric stereo) are accepted to CVPR 2019.

Publications

Recent Preprints

Beyond Single Expert: Harmonizing Diverse Visual Priors in MLLMs for Spatial Understanding
Xiao Lin, Xiaohu Huang, Kai Han
arXiv preprint arXiv:2607.15054, 2026.
BibTeX PDF arXiv Project 
@article{Lin2026ViPS,
 author    = {Xiao Lin and Xiaohu Huang and Kai Han},
 title     = {Beyond Single Expert: Harmonizing Diverse Visual Priors in MLLMs for Spatial Understanding},
 journal   = {arXiv preprint arXiv:2607.15054},
 year      = {2026},
}
Surgical Post-Training: Proximal On-Policy Distillation for Reasoning with Knowledge Retention
Wenye Lin, Kai Han
arXiv preprint arXiv:2603.01683, 2026.
BibTeX PDF arXiv 
@article{Lin2026Surgical,
 author    = {Wenye Lin and Kai Han},
 title     = {Surgical Post-Training: Proximal On-Policy Distillation for Reasoning with Knowledge Retention},
 journal   = {arXiv preprint arXiv:2603.01683},
 year      = {2026},
}
JoVA: Unified Multimodal Learning for Joint Video-Audio Generation
Xiaohu Huang, Hao Zhou, Qiangpeng Yang, Shilei Wen, Kai Han
arXiv preprint arXiv:2512.13677, 2026.
BibTeX PDF arXiv Project Code 
@article{Huang2025JoVA,
 author    = {Xiaohu Huang and Hao Zhou and Qiangpeng Yang and Shilei Wen and Kai Han},
 title     = {JoVA: Unified Multimodal Learning for Joint Video-Audio Generation},
 journal   = {arXiv preprint arXiv:2512.13677},
 year      = {2026},
}
Generalized Category Discovery under Domain Shifts: From Vision to Vision-Language Models
Hongjun Wang, Po Hu, Kai Han
arXiv preprint arXiv:2605.00906, 2026.
BibTeX PDF arXiv Project 
@article{Wang2026GCDDomainShifts,
 author    = {Hongjun Wang and Po Hu and Kai Han},
 title     = {Generalized Category Discovery under Domain Shifts: From Vision to Vision-Language Models},
 journal   = {arXiv preprint arXiv:2605.00906},
 year      = {2026},
}
Effective Prompt Pool Learning for Continual Category Discovery
Fernando Julio Cendra, Xinghui Li, Kai Han
arXiv preprint arXiv:2407.19001, 2026.
BibTeX PDF arXiv 
@article{Cendra2026PromptPool,
 author    = {Fernando Julio Cendra and Xinghui Li and Kai Han},
 title     = {Effective Prompt Pool Learning for Continual Category Discovery},
 journal   = {arXiv preprint arXiv:2407.19001},
 year      = {2026},
}
MHPO: Modulated Hazard-aware Policy Optimization for Stable Reinforcement Learning
Hongjun Wang, Wei Liu, Weibo Gu, Xing Sun, Kai Han
arXiv preprint arXiv:2603.16929, 2026.
BibTeX PDF arXiv 
@article{Wang2026MHPO,
 author    = {Hongjun Wang and Wei Liu and Weibo Gu and Xing Sun and Kai Han},
 title     = {MHPO: Modulated Hazard-aware Policy Optimization for Stable Reinforcement Learning},
 journal   = {arXiv preprint arXiv:2603.16929},
 year      = {2026},
}
Category Discovery: An Open-World Perspective
Zhenqi He, Yuanpei Liu, Kai Han
arXiv preprint arXiv:2509.22542, 2026
BibTeX PDF arXiv Project 
@article{He2025Category,
 author    = {He, Zhenqi and Liu, Yuanpei and Han, Kai},
 title     = {Category Discovery: An Open-World Perspective},
 journal = {arXiv preprint arXiv:2509.22542},
 year      = {2026},
}

2026

PartCo: Part-Level Correspondence Priors Enhance Category Discovery
Fernando Julio Cendra, Kai Han
International Conference on Machine Learning (ICML), 2026.
BibTeX PDF arXiv Project 
@inproceedings{Cendra2026PartCo,
 author    = {Fernando Julio Cendra and Kai Han},
 title     = {PartCo: Part-Level Correspondence Priors Enhance Category Discovery},
 booktitle = {International Conference on Machine Learning (ICML)},
 year      = {2026},
}
Geometric Reciprocity: Unlocking Self-Supervision for Stereoscopic Video Generation
Jingyi Lu, Kai Han
International Conference on Machine Learning (ICML), 2026.
BibTeX PDF arXiv Project 
@inproceedings{Lu2026GeomRecip,
 author    = {Jingyi Lu and Kai Han},
 title     = {Geometric Reciprocity: Unlocking Self-Supervision for Stereoscopic Video Generation},
 booktitle = {International Conference on Machine Learning (ICML)},
 year      = {2026},
}
iVGR: Internalizing Visually Grounded Reasoning for MLLMs with Reinforcement Learning
Chang-Bin Zhang, Yujie Zhong, Qiang Zhang, Kai Han
International Conference on Machine Learning (ICML), 2026.
BibTeX PDF arXiv Project Code 
@inproceedings{Zhang2026iVGR,
 author    = {Chang-Bin Zhang and Yujie Zhong and Qiang Zhang and Kai Han},
 title     = {iVGR: Internalizing Visually Grounded Reasoning for MLLMs with Reinforcement Learning},
 booktitle = {International Conference on Machine Learning (ICML)},
 year      = {2026},
}
ZeroBench: An Impossible Visual Benchmark for Contemporary Large Multimodal Models
Jonathan Roberts, Mohammad Reza Taesiri, Ansh Sharma, Akash Gupta, et al., Kai Han†, Samuel Albanie†
International Conference on Machine Learning (ICML), 2026.
BibTeX PDF arXiv Project Code 
@inproceedings{Roberts2026ZeroBench,
 author    = {Jonathan Roberts and Mohammad Reza Taesiri and Ansh Sharma and Akash Gupta and others and Kai Han and Samuel Albanie},
 title     = {ZeroBench: An Impossible Visual Benchmark for Contemporary Large Multimodal Models},
 booktitle = {International Conference on Machine Learning (ICML)},
 year      = {2026},
}
How Long Is a Piece of String? A Brief Empirical Analysis of Tokenizers
Jonathan Roberts, Kai Han, Samuel Albanie
ICML 2026 Workshop on Combining Theory and Benchmarks (CTB).
BibTeX PDF arXiv 
@inproceedings{Roberts2026Tokenizers,
 author    = {Jonathan Roberts and Kai Han and Samuel Albanie},
 title     = {How Long Is a Piece of String? A Brief Empirical Analysis of Tokenizers},
 booktitle = {ICML Workshop on Combining Theory and Benchmarks (CTB)},
 year      = {2026},
}
Sculpt4D: Generating 4D Shapes via Sparse-Attention Diffusion Transformers
Minghao Yin, Wenbo Hu, Jiale Xu, Ying Shan, Kai Han
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2026.
BibTeX PDF arXiv Project Code 
@inproceedings{Yin2026Sculpt4D,
 author    = {Minghao Yin and Wenbo Hu and Jiale Xu and Ying Shan and Kai Han},
 title     = {Sculpt4D: Generating 4D Shapes via Sparse-Attention Diffusion Transformers},
 booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 year      = {2026},
}
Speed3R: Sparse Feed-forward 3D Reconstruction Models
Weining Ren, Xiao Tan, Kai Han
IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Findings, 2026.
BibTeX PDF arXiv Project Code 
@inproceedings{Ren2026Speed3R,
 author    = {Weining Ren and Xiao Tan and Kai Han},
 title     = {Speed3R: Sparse Feed-forward 3D Reconstruction Models},
 booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Findings},
 year      = {2026},
}
Scene-Level Heterogeneous Physics Simulation with 3D Gaussian Splats
Xiaoyang Liu, Shangzhe Wu, Kai Han
IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Findings, 2026.
BibTeX PDF arXiv Project Code 
@inproceedings{Liu2026HeteroPhys,
 author    = {Xiaoyang Liu and Shangzhe Wu and Kai Han},
 title     = {Scene-Level Heterogeneous Physics Simulation with 3D Gaussian Splats},
 booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Findings},
 year      = {2026},
}
When Deepfake Detection Meets Graph Neural Network: a Unified and Lightweight Learning Framework
Haoyu Liu, Chaoyu Gong, Mengke He, Jiate Li, Kai Han, Siqiang Luo
ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2026.
BibTeX PDF arXiv 
@inproceedings{Liu2026Deepfake,
 author    = {Haoyu Liu and Chaoyu Gong and Mengke He and Jiate Li and Kai Han and Siqiang Luo},
 title     = {When Deepfake Detection Meets Graph Neural Network: a Unified and Lightweight Learning Framework},
 booktitle = {ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD)},
 year      = {2026},
}
Ascending the Infinite Ladder: Benchmarking Spatial Deformation Reasoning in Vision-Language Models
Jiahuan Zhang, Shunwen Bai, Tianheng Wang, Kaiwen Guo, Zijia Song, Hanqing Wu, Guozheng Rao, Kai Han, Kaicheng Yu
Annual Meeting of the Association for Computational Linguistics (ACL), 2026.
BibTeX PDF arXiv 
@inproceedings{Zhang2026Ladder,
 author    = {Jiahuan Zhang and Shunwen Bai and Tianheng Wang and Kaiwen Guo and Zijia Song and Hanqing Wu and Guozheng Rao and Kai Han and Kaicheng Yu},
 title     = {Ascending the Infinite Ladder: Benchmarking Spatial Deformation Reasoning in Vision-Language Models},
 booktitle = {Annual Meeting of the Association for Computational Linguistics (ACL)},
 year      = {2026},
}
CodeBind: Decoupled Representation Learning for Multimodal Alignment with Unified Compositional Codebook
Zeyu Chen, Jie Li, Kai Han
Annual Meeting of the Association for Computational Linguistics (ACL) Findings, 2026.
BibTeX PDF arXiv Project Code 
@inproceedings{Chen2026CodeBind,
 author    = {Zeyu Chen and Jie Li and Kai Han},
 title     = {CodeBind: Decoupled Representation Learning for Multimodal Alignment with Unified Compositional Codebook},
 booktitle = {Annual Meeting of the Association for Computational Linguistics (ACL) Findings},
 year      = {2026},
}
AI in Oral Health Surveillance: Critical Review
Zeyu Chen, Pei Liu, Kai Han, Peixi Liao, Yanqi Yang, May C.M. Wong, Cynthia K.Y. Yiu, Edward C.M. Lo
Journal of Dental Research (JDR), 2026.
BibTeX DOI 
@article{Chen2026JDR,
 author  = {Zeyu Chen and Pei Liu and Kai Han and Peixi Liao and Yanqi Yang and May C.M. Wong and Cynthia K.Y. Yiu and Edward C.M. Lo},
 title   = {AI in Oral Health Surveillance: Critical Review},
 journal = {Journal of Dental Research (JDR)},
 year    = {2026},
}
LooC: Effective Low-Dimensional Codebook for Compositional Vector Quantization
Jie Li, Kwan-Yee K. Wong, Kai Han
IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2026.
BibTeX PDF arXiv Project Code 
@inproceedings{Li2026LooC,
 author    = {Jie Li and Kwan-Yee K. Wong and Kai Han},
 title     = {LooC: Effective Low-Dimensional Codebook for Compositional Vector Quantization},
 booktitle = {IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)},
 year      = {2026},
}
Align Video Diffusion Model with Online Video-Centric Preference Optimization
Jiacheng Zhang, Jie Wu, Weifeng Chen, Yatai Ji, Weilin Huang, Xuefeng Xiao, Kai Han
IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2026.
BibTeX PDF arXiv Project 
@inproceedings{Zhang2026AlignVD,
 author    = {Jiacheng Zhang and Jie Wu and Weifeng Chen and Yatai Ji and Weilin Huang and Xuefeng Xiao and Kai Han},
 title     = {Align Video Diffusion Model with Online Video-Centric Preference Optimization},
 booktitle = {IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)},
 year      = {2026},
}

2025

Semantic Correspondence: Unified Benchmarking and a Strong Baseline
Kaiyan Zhang, Xinghui Li, Jingyi Lu, Kai Han
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2025
BibTeX PDF arXiv Project Code 
@article{Zhang2025Semantic,
 author    = {Zhang, Kaiyan and Li, Xinghui and Lu, Jingyi and Han, Kai},
 title     = {Semantic Correspondence: Unified Benchmarking and a Strong Baseline},
 journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)},
 year      = {2025},
}
Panoptic Captioning: An Equivalence Bridge for Image and Text
Kun-Yu Lin, Hongjun Wang, Weining Ren, Kai Han
Conference on Neural Information Processing Systems (NeurIPS), 2025
BibTeX PDF arXiv Project Code 
@inproceedings{lin2025PanCap,
 author    = {Kun-Yu Lin and Hongjun Wang and Weining Ren and Kai Han},
 title     = {Panoptic Captioning: Seeking An Equivalency Bridge for Image and Text},
 booktitle = {Conference on Neural Information Processing Systems (NeurIPS)},
 year      = {2025},
}
3DRS: MLLMs Need 3D-Aware Representation Supervision for Scene Understanding
Xiaohu Huang, Jingjing Wu, Qunyi Xie, Kai Han
Conference on Neural Information Processing Systems (NeurIPS), 2025
BibTeX PDF arXiv Project Code 
@inproceedings{huang2025ThreeDRS,
 author    = {Xiaohu Huang and Jingjing Wu and Qunyi Xie and Kai Han},
 title     = {MLLMs Need 3D-Aware Representation Supervision for Scene Understanding},
 booktitle = {Conference on Neural Information Processing Systems (NeurIPS)},
 year      = {2025},
}
Wukong's 72 Transformations: High-fidelity 3D Morphing via Flow Models
Minghao Yin, Yukang Cao, Kai Han
Conference on Neural Information Processing Systems (NeurIPS), 2025
BibTeX PDF arXiv Project 
@inproceedings{Yin2025Wukong,
 author    = {Minghao Yin and Yukang Cao and Kai Han},
 title     = {Wukong's 72 Transformations: High-fidelity 3D Morphing via Flow Models},
 booktitle = {Conference on Neural Information Processing Systems (NeurIPS)},
 year      = {2025},
}
VaMP: Variational Multi-Modal Prompt Learning for Vision-Language Models
Silin Cheng, Kai Han
Conference on Neural Information Processing Systems (NeurIPS), 2025
BibTeX PDF arXiv Project 
@inproceedings{Cheng2025VaMP,
 author    = {Silin Cheng and Kai Han},
 title     = {VaMP: Variational Multi-Modal Prompt Learning},
 booktitle = {Conference on Neural Information Processing Systems (NeurIPS)},
 year      = {2025},
}
Fin3R: Fine-Tuning Feed-Foward 3D Reconstruction Models via Monocular Knowledge Distillation
Weining Ren, Hongjun Wang, Xiao Tan, Kai Han
Conference on Neural Information Processing Systems (NeurIPS), 2025
BibTeX PDF arXiv Project Code 
@inproceedings{Ren2025Fin3R,
 author    = {Weining Ren and Hongjun Wang and Xiao Tan and Kai Han},
 title     = {Fin3R: Fine-Tuning Feed-Foward 3D Reconstruction Models via Monocular Knowledge Distillation},
 booktitle = {Conference on Neural Information Processing Systems (NeurIPS)},
 year      = {2025},
}
SEAL: Semantic-Aware Hierarchical Learning for Generalized Category Discovery
Zhenqi He*, Yuanpei Liu*, Kai Han
Conference on Neural Information Processing Systems (NeurIPS), 2025
BibTeX PDF arXiv Project Code 
@inproceedings{He2025SEAL,
 author    = {Zhenqi He and Yuanpei Liu and Kai Han},
 title     = {SEAL: Semantic-Aware Hierarchical Learning for Generalized Category Discovery},
 booktitle = {Conference on Neural Information Processing Systems (NeurIPS)},
 year      = {2025},
}
GSPN-2: Efficient Parallel Sequence Modeling
Hongjun Wang, Yitong Jiang, Collin McCarthy, David Wehr, Hanrong Ye, Xinhao Li, Ka Chun Cheung, Wonmin Byeon, Jinwei Gu, Ke Chen, Kai Han, Hongxu Yin, Pavlo Molchanov, Jan Kautz, Sifei Liu
Conference on Neural Information Processing Systems (NeurIPS), 2025
BibTeX PDF arXiv 
@inproceedings{Wang2025GSPN2,
 author    = {Hongjun Wang and Yitong Jiang and Collin McCarthy and David Wehr and Hanrong Ye and Xinhao Li and Ka Chun Cheung and Wonmin Byeon and Jinwei Gu and Ke Chen and Kai Han and Hongxu Yin and Pavlo Molchanov and Jan Kautz and Sifei Liu},
 title     = {GSPN-2: Efficient Parallel Sequence Modeling},
 booktitle = {Conference on Neural Information Processing Systems (NeurIPS)},
 year      = {2025},
}
ELIP: Enhanced Visual-Language Foundation Models for Image Retrieval
Guanqi Zhan*, Yuanpei Liu*, Kai Han, Weidi Xie, Andrew Zisserman
IEEE International Conference on Content-Based Multimedia Indexing (CBMI), 2025.
BibTeX PDF arXiv Project Code 
@inproceedings{Zhan2025ELIP,
 author    = {Zhan, Guanqi and Liu, Yuanpei and Han, Kai and Xie, Weidi and Zisserman, Andrew},
 title     = {ELIP: Enhanced Visual-Language Foundation Models for Image Retrieval},
 booktitle = {International Conference on Content-Based Multimedia Indexing (CBMI)},
 year      = {2025},
}
Inpaint4Drag: Repurposing Inpainting Models for Drag-Based Image Editing via Bidirectional Warping
Jingyi Lu, Kai Han
International Conference on Computer Vision (ICCV), 2025.
BibTeX PDF arXiv Project Code Demo 
@inproceedings{Lu2025Inpaint4Drag,
 author    = {Jingyi Lu and Kai Han},
 title     = {Inpaint4Drag: Repurposing Inpainting Models for Drag-Based Image Editing via Bidirectional Warping},
 booktitle = {International Conference on Computer Vision (ICCV)},
 year      = {2025},
}
GRAB: A Challenging GRaph Analysis Benchmark for Large Multimodal Models
Jonathan Roberts, Kai Han, Samuel Albanie
International Conference on Computer Vision (ICCV), 2025.
BibTeX PDF arXiv Project Data Code 
@inproceedings{Roberts2025GRAB,
 author    = {Jonathan Roberts and Kai Han and Samuel Albanie},
 title     = {GRAB: A Challenging GRaph Analysis Benchmark for Large Multimodal Models},
 booktitle = {International Conference on Computer Vision (ICCV)},
 year      = {2025},
}
GAMEBoT: Transparent Assessment of LLM Reasoning in Games
Wenye Lin, Jonathan Roberts, Yunhan Yang, Samuel Albanie, Zongqing Lu, Kai Han
Annual Meeting of the Association for Computational Linguistics (ACL), 2025.
BibTeX PDF arXiv Project Code 
@inproceedings{Lin2025GAMEBot,
 author    = {Wenye Lin and Jonathan Roberts and Yunhan Yang and Samuel Albanie and Zongqing Lu and Kai Han},
 title     = {GAMEBoT: Transparent Assessment of LLM Reasoning in Games},
 booktitle = {Annual Meeting of the Association for Computational Linguistics (ACL)},
 year      = {2025},
}
PruneVid: Visual Token Pruning for Efficient Video Large Language Models
Xiaohu Huang, Hao Zhou, Kai Han
Annual Meeting of the Association for Computational Linguistics (ACL) Findings, 2025.
BibTeX PDF arXiv Project Code 
@inproceedings{Huang2025PruneVid,
 author    = {Xiaohu Huang and Hao Zhou and Kai Han},
 title     = {PruneVid: Visual Token Pruning for Efficient Video Large Language Models},
 booktitle = {Annual Meeting of the Association for Computational Linguistics (ACL) Findings},
 year      = {2025},
}
Splat4D: Diffusion-Enhanced 4D Gaussian Splatting for Temporally and Spatially Consistent Content Creation
Minghao Yin, Yukang Cao, Songyou Peng, Kai Han
SIGGRAPH, 2025.
BibTeX PDF arXiv DOI Project 
@inproceedings{Yin2025Splat4D,
 author    = {Minghao Yin and Yukang Cao and Songyou Peng and Kai Han},
 title     = {Splat4D: Diffusion-Enhanced 4D Gaussian Splatting for Temporally and Spatially Consistent Content Creation},
 booktitle = {SIGGRAPH},
 year      = {2025},
}
ICE: Intrinsic Concept Extraction from a Single Image via Diffusion Models
Fernando Julio Cendra, Kai Han
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2025.
Highlight presentation · 2.8% of submissions
BibTeX PDF arXiv Project Code 
@inproceedings{Cendra2025ICE,
 author    = {Fernando Julio Cendra and Kai Han},
 title     = {ICE: Intrinsic Concept Extraction from a Single Image via Diffusion Models},
 booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 year      = {2025},
}
Hyperbolic Category Discovery
Yuanpei Liu*, Zhenqi He*, Kai Han
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2025.
BibTeX PDF arXiv Project Code 
@inproceedings{Liu2025HypCD,
 author    = {Yuanpei Liu and Zhenqi He and Kai Han},
 title     = {Hyperbolic Category Discovery},
 booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 year      = {2025},
}
Mr. DETR: Instructive Multi-Route Training for Detection Transformers
Chang-Bin Zhang, Yujie Zhong, Kai Han
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2025.
BibTeX PDF arXiv Project Code 
@inproceedings{Zhang2025MrDETR,
 author    = {Chang-Bin Zhang and Jinhong Ni and Yujie Zhong and Kai Han},
 title     = {Mr. DETR: Instructive Multi-Route Training for Detection Transformers},
 booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 year      = {2025},
}
v-CLR: View-Consistent Learning for Open-World Instance Segmentation
Chang-Bin Zhang, Jinhong Ni, Yujie Zhong, Kai Han
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2025.
Highlight presentation · 2.8% of submissions
BibTeX PDF arXiv Project Code 
@inproceedings{Zhang2025vCLR,
 author    = {Chang-Bin Zhang and Jinhong Ni and Yujie Zhong and Kai Han},
 title     = {v-CLR: View-Consistent Learning for Open-World Instance Segmentation},
 booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 year      = {2025},
}
Parallel Sequence Modeling via Generalized Spatial Propagation Network
Hongjun Wang, Wonmin Byeon, Jiarui Xu, Jinwei Gu, Ka Chun Cheung, Xiaolong Wang, Kai Han, Jan Kautz, Sifei Liu
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2025.
BibTeX PDF arXiv Project Code 
@inproceedings{Wang2025GSPN,
 author    = {Hongjun Wang and Wonmin Byeon and Jiarui Xu and Jinwei Gu and Ka Chun Cheung and Xiaolong Wang and Kai Han and Jan Kautz and Sifei Liu},
 title     = {Parallel Sequence Modeling via Generalized Spatial Propagation Network},
 booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 year      = {2025},
}
Detecting Open World Objects via Partial Attribute Assignment
Muli Yang, Gabriel James Goenawan, Huaiyuan Qin, Kai Han, Xi Peng, Yanhua Yang, Hongyuan Zhu
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2025.
BibTeX PDF Code 
@inproceedings{Yang2025PASS,
 author    = {Muli Yang and Gabriel James Goenawan and Huaiyuan Qin and Kai Han and Xi Peng and Yanhua Yang and Hongyuan Zhu},
 title     = {Detecting Open World Objects via Partial Attribute Assignment},
 booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 year      = {2025},
}
DebGCD: Debiased Learning with Distribution Guidance for Generalized Category Discovery
Yuanpei Liu, Kai Han
International Conference on Learning Representations (ICLR), 2025.
BibTeX PDF arXiv Project Code 
@inproceedings{Liu2025DebGCD,
 author    = {Yuanpei Liu and Kai Han},
 title     = {DebGCD: Debiased Learning with Distribution Guidance for Generalized Category Discovery},
 booktitle = {International Conference on Learning Representations (ICLR)},
 year      = {2025},
}
HiLo: A Learning Framework for Generalized Category Discovery Robust to Domain Shifts
Hongjun Wang, Sagar Vaze, Kai Han
International Conference on Learning Representations (ICLR), 2025.
BibTeX PDF arXiv Project Code 
@inproceedings{wang2025hilo,
 author    = {Hongjun Wang and Sagar Vaze and Kai Han},
 title     = {HiLo: A Learning Framework for Generalized Category Discovery Robust to Domain Shifts},
 booktitle  = {International Conference on Learning Representations (ICLR)},
 year      = {2025},
}
Needle Threading: Can LLMs Follow Threads Through Near-Million-Scale Haystacks?
Jonathan Roberts, Kai Han, Samuel Albanie
International Conference on Learning Representations (ICLR), 2025.
BibTeX PDF arXiv Project Code Dataset 
@inproceedings{Roberts2025Needle,
 author    = {Jonathan Roberts and Kai Han and Samuel Albanie},
 title     = {Needle Threading: Can LLMs Follow Threads Through Near-Million-Scale Haystacks?},
 booktitle = {International Conference on Learning Representations (ICLR)},
 year      = {2025},
}
BiGR: Harnessing Binary Latent Codes for Image Generation and Improved Visual Representation Capabilities
Shaozhe Hao, Xuantong Liu, Xianbiao Qi, Shihao Zhao, Bojia Zi, Rong Xiao, Kai Han, Kwan-Yee K. Wong
International Conference on Learning Representations (ICLR), 2025.
BibTeX PDF arXiv Project Code 
@inproceedings{Hao2025BiGR,
 author     = {Shaozhe Hao and Xuantong Liu and Xianbiao Qi and Shihao Zhao and Bojia Zi and Rong Xiao and Kai Han and Kwan-Yee~K. Wong},
 title      = {Bi{GR}: Harnessing Binary Latent Codes for Image Generation and Improved Visual Representation Capabilities}, 
 booktitle  = {International Conference on Learning Representations (ICLR)},
 year       = {2025},
}
AvatarGO: Zero-shot 4D Human-Object Interaction Generation and Animation
Yukang Cao, Liang Pan, Kai Han, Kwan-Yee K. Wong, Ziwei Liu
International Conference on Learning Representations (ICLR), 2025.
BibTeX PDF arXiv Project Code 
@inproceedings{Cao2025AvatarGO,
 author    = {Yukang Cao and Liang Pan and Kai Han and Kwan-Yee K. Wong and Ziwei Liu},
 title     = {AvatarGO: Zero-shot 4D Human-Object Interaction Generation and Animation},
 booktitle = {International Conference on Learning Representations (ICLR)},
 year      = {2025},
}
VipDiff: Towards Coherent and Diverse Video Inpainting via Training-free Denoising Diffusion Models
Chaohao Xie, Kai Han, Kwan-Yee K. Wong
IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2025.
BibTeX PDF arXiv Project Code 
@inproceedings{Xie2025VipDiff,
 author    = {Chaohao Xie and Kai Han and Kwan-Yee K. Wong},
 title     = {VipDiff: Towards Coherent and Diverse Video Inpainting via Training-free Denoising Diffusion Models},
 booktitle = {IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)},
 year      = {2025},
}
CusConcept: Customized Visual Concept Decomposition with Diffusion Models
Zhi Xu, Shaozhe Hao, Kai Han
IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2025.
BibTeX PDF arXiv Code 
@inproceedings{Xu2025CusConcept,
 author    = {Zhi Xu and Shaozhe Hao and Kai Han},
 title     = {CusConcept: Customized Visual Concept Decomposition with Diffusion Models},
 booktitle = {IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)},
 year      = {2025},
}

2024

SciFIBench: Benchmarking Large Multimodal Models for Scientific Figure Interpretation
Jonathan Roberts, Kai Han, Neil Houlsby, Samuel Albanie
Conference on Neural Information Processing Systems (NeurIPS), 2024.
BibTeX PDF arXiv Data & Code 
@inproceedings{Roberts2024SciFIBench,
 author    = {Jonathan Roberts and Kai Han and Neil Houlsby and Samuel Albanie},
 title     = {SciFIBench: Benchmarking Large Multimodal Models for Scientific Figure Interpretation},
 booktitle = {Conference on Neural Information Processing Systems},
 year      = {2024},
}
Dissecting Out-of-Distribution Detection and Open-Set Recognition: A Critical Analysis of Methods and Benchmarks
Hongjun Wang, Sagar Vaze, Kai Han
International Journal of Computer Vision (IJCV), 2024.
BibTeX PDF arXiv Project Code 
@article{wang2024dissect,
    author    = {Wang, Hongjun and Vaze, Sagar and Han, Kai},
    title     = {Dissecting Out-of-Distribution Detection and Open-Set Recognition: A Critical Analysis of Methods and Benchmarks},
    journal   = {International Journal of Computer Vision (IJCV)},
    year      = {2024}
}
RegionDrag: Fast Region-Based Image Editing with Diffusion Models
Jingyi Lu, Xinghui Li, Kai Han
European Conference on Computer Vision (ECCV), 2024.
BibTeX PDF arXiv Project Code 
@inproceedings{lu2024regiondrag,
    author    = {Jingyi Lu and Xinghui Li and Kai Han},
    title     = {RegionDrag: Fast Region-Based Image Editing with Diffusion Models},
    booktitle = {European Conference on Computer Vision (ECCV)},
    year      = {2024},
}
PromptCCD: Learning Gaussian Mixture Prompt Pool for Continual Category Discovery
Fernando Julio Cendra, Bingchen Zhao, Kai Han
European Conference on Computer Vision (ECCV), 2024.
BibTeX PDF arXiv Project Code 
 @inproceedings{cendra2024promptccd,
  author    = {Fernando Julio Cendra and Bingchen Zhao and Kai Han},
  title     = {PromptCCD: Learning Gaussian Mixture Prompt Pool for Continual Category Discovery},
  booktitle = {European Conference on Computer Vision},
  year      = {2024}
}
ConceptExpress: Harnessing Diffusion Models for Single-image Unsupervised Concept Extraction
Shaozhe Hao, Kai Han, Zhengyao Lv, Shihao Zhao, Kwan-Yee K. Wong
European Conference on Computer Vision (ECCV), 2024.
Oral presentation · 2.3% of submissions
BibTeX PDF arXiv Project Code 
 @inproceedings{hao2024conceptexpress,
  author    = {Shaozhe Hao and Kai Han and Zhengyao Lv and Shihao Zhao and Kwan-Yee~K. Wong},
  title     = {Concept{E}xpress: Harnessing Diffusion Models for Single-image Unsupervised Concept Extraction},
  booktitle = {European Conference on Computer Vision},
  year      = {2024}
}
IBD-SLAM: Learning Image-Based Depth Fusion for Generalizable SLAM
Minghao Yin, Shangzhe Wu, Kai Han
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2024.
BibTeX PDF Project 
@inproceedings{yin2024ibdslam,
 author    = {Minghao Yin and Shangzhe Wu and Kai Han},
 title     = {IBD-SLAM: Learning Image-Based Depth Fusion for Generalizable SLAM},
 booktitle  = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 year      = {2024},
}
DreamAvatar: Text-and-Shape Guided 3D Human Avatar Generation via Diffusion Models
Yukang Cao*, Yan-Pei Cao*, Kai Han, Ying Shan, Kwan-Yee K. Wong
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2024.
BibTeX PDF arXiv Project Code 
@inproceedings{cao24dream,
 author    = {Yukang Cao and Yan-Pei Cao and Kai Han and Ying Shan and Kwan-Yee K. Wong},
 title     = {DreamAvatar: Text-and-Shape Guided 3D Human Avatar Generation via Diffusion Models},
 booktitle  = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 year      = {2024},
}
SD4Match: Learning to Prompt Stable Diffusion Model for Semantic Matching
Xinghui Li, Jingyi Lu, Kai Han, Victor Prisacariu
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2024.
BibTeX PDF arXiv Project Code 
@inproceedings{li2024sd4match,
 author    = {Xinghui Li and Jingyi Lu and Kai Han and Victor Prisacariu},
 title     = {SD4Match: Learning to Prompt Stable Diffusion Model for Semantic Matching},
 booktitle  = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 year      = {2024},
}
What’s in a Name? Beyond Class Indices for Image Recognition
Kai Han*, Xiaohu Huang*, Yandong Li*, Sagar Vaze*, Jie Li, Xuhui Jia
CVPR Workshop on Computer Vision in the Wild, 2024.
BibTeX PDF arXiv Code 
@inproceedings{han23scd,
 author    = {Kai Han and Yandong Li and Sagar Vaze and Jie Li and Xuhui Jia},
 title     = {What's in a Name? Beyond Class Indices for Image Recognition},
 booktitle = {CVPR Workshop on Computer Vision in the Wild},
 year      = {2024},
}
Charting New Territories: Exploring the Geographic and Geospatial Capabilities of Multimodal LLMs
Jonathan Roberts, Timo Lüddecke, Rehan Sheikh, Kai Han, Samuel Albanie
CVPR Workshop on EarthVision, 2024.
BibTeX PDF arXiv Dataset 
@inproceedings{Roberts2024NewTerri,
 author    = {Roberts, Jonathan and L\"uddecke, Timo and Sheikh, Rehan and Han, Kai and Albanie, Samuel},
 title     = {Charting New Territories: Exploring the Geographic and Geospatial Capabilities of Multimodal LLMs},
 booktitle = {CVPR Workshop on EarthVision},
 year      = {2024},
}
CiPR: An Efficient Framework with Cross-instance Positive Relations for Generalized Category Discovery
Shaozhe Hao, Kai Han, Kwan-Yee K. Wong
Transactions on Machine Learning Research (TMLR), 2024.
BibTeX PDF arXiv Code 
@article{hao24cipr,
  title    = {CiPR: An Efficient Framework with Cross-instance Positive Relations for Generalized Category Discovery},
  author   = {Shaozhe Hao and Kai Han and Kwan-Yee K. Wong},
  journal  = {Transactions on Machine Learning Research},
  year     = {2024},
}
SPTNet: An Efficient Alternative Framework for Generalized Category Discovery with Spatial Prompt Tuning
Hongjun Wang, Sagar Vaze, Kai Han
International Conference on Learning Representations (ICLR), 2024.
BibTeX PDF arXiv Project Code 
@inproceedings{wang2024sptnet,
 author    = {Hongjun Wang and Sagar Vaze and Kai Han},
 title     = {SPTNet: An Efficient Alternative Framework for Generalized Category Discovery with Spatial Prompt Tuning},
 booktitle  = {International Conference on Learning Representations (ICLR)},
 year      = {2024},
}
FROSTER: Frozen CLIP is a Strong Teacher for Open-Vocabulary Action Recognition
Xiaohu Huang, Hao Zhou, Kun Yao, Kai Han
International Conference on Learning Representations (ICLR), 2024.
BibTeX PDF arXiv Project Code 
@inproceedings{huang2024froster,
 author    = {Xiaohu Huang and Hao Zhou and Kun Yao and Kai Han},
 title     = {FROSTER: Frozen CLIP is a Strong Teacher for Open-Vocabulary Action Recognition},
 booktitle  = {International Conference on Learning Representations (ICLR)},
 year      = {2024},
}

2023

HeadSculpt: Crafting 3D Head Avatars with Text
Xiao Han*, Yukang Cao*, Kai Han, Xiatian Zhu, Jiankang Deng, Yi-Zhe Song, Tao Xiang, Kwan-Yee K. Wong
Conference on Neural Information Processing Systems (NeurIPS), 2023.
BibTeX PDF arXiv Project Code 
@inproceedings{han2023headsculpt,
 author    = {Xiao Han and Yukang Cao and Kai Han and Xiatian Zhu and Jiankang Deng and Yi-Zhe Song and Tao Xiang and Kwan-Yee K. Wong},
 title     = {HeadSculpt: Crafting 3D Head Avatars with Text},
 booktitle  = {Conference on Neural Information Processing Systems (NeurIPS)},
 year      = {2023},
}
GPT4GEO: How a Language Model Sees the World's Geography
Jonathan Roberts, Timo Lüddecke, Sowmen Das, Kai Han, Samuel Albanie
NeurIPS Workshop on Foundation Models for Decision Making, 2023.
BibTeX PDF arXiv Code 
@inproceedings{roberts2023GPT4GEO,
 author = {Roberts, Jonathan and L{\"u}ddecke, Timo and Das, Sowmen and Han, Kai and Albanie, Samuel},
 title     = {GPT4GEO: How a Language Model Sees the World's Geography},
 booktitle  = {NeurIPS Workshop on Foundation Models for Decision Making},
 year      = {2023},
}
DualRC: A Dual-Resolution Learning Framework with Neighbourhood Consensus for Visual Correspondences
Xinghui Li, Kai Han, Shuda Li, Victor Prisacariu
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2023.
BibTeX DOI Project Code 
@article{li23dualrc,
 author  = {Xinghui Li and Kai Han and Shuda Li and Victor Prisacariu},
 title   = {DualRC: A Dual-Resolution Learning Framework with Neighbourhood Consensus for Visual Correspondences},
 journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)},
 year    = {2023},
}
Learning Semi-supervised Gaussian Mixture Models for Generalized Category Discovery
Bingchen Zhao, Xin Wen, Kai Han
International Conference on Computer Vision (ICCV), 2023.
BibTeX PDF arXiv Code 
@inproceedings{zhao23learning,
 author    = {Bingchen Zhao and Xin Wen and Kai Han},
 title     = {Learning Semi-supervised Gaussian Mixture Models for Generalized Category Discovery},
 booktitle = {International Conference on Computer Vision (ICCV)},
 year      = {2023},
}
Open-Vocabulary Semantic Segmentation with Decoupled One-Pass Network
Cong Han*, Yujie Zhong*, Dengjie Li, Kai Han, Lin Ma
International Conference on Computer Vision (ICCV), 2023.
BibTeX PDF arXiv Code 
@inproceedings{han23open,
 author    = {Cong Han and Yujie Zhong and Dengjie Li and Kai Han and Lin Ma},
 title     = {Open-Vocabulary Semantic Segmentation with Decoupled One-Pass Network},
 booktitle = {International Conference on Computer Vision (ICCV)},
 year      = {2023},
}
SATIN: A Multi-Task Metadataset for Classifying Satellite Imagery using Vision-Language Models
Jonathan Roberts, Kai Han, Samuel Albanie
ICCV TNGCV Workshop, 2023.
BibTeX PDF arXiv Project 
@inproceedings{roberts2023satin,
  title    = {SATIN: A Multi-Task Metadataset for Classifying Satellite Imagery using Vision-Language Models},
  author   = {Jonathan Roberts and Kai Han and Samuel Albanie},
  booktitle  = {ICCV TNGCV Workshop},
  year     = {2023},
}
Learning Attention as Disentangler for Compositional Zero-shot Learning
Shaozhe Hao, Kai Han, Kwan-Yee K. Wong
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023.
BibTeX PDF arXiv Project Code 
@inproceedings{hao2023ade,
 author    = {Shaozhe Hao and Kai Han and Kwan-Yee K. Wong},
 title     = {Learning Attention as Disentangler for Compositional Zero-shot Learning},
 booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 year      = {2023},
}
SeSDF: Self-evolved Signed Distance Field for Implicit 3D Clothed Human Reconstruction
Yukang Cao, Kai Han, Kwan-Yee K. Wong
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023.
BibTeX PDF arXiv Project Code 
@inproceedings{cao2023sesdf,
author    = {Yukang Cao and Kai Han and Kwan-Yee K. Wong},
title     = {SeSDF: Self-evolved Signed Distance Field for Implicit 3D Clothed Human Reconstruction},
booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
year      = {2023},
}
ViCo: Detail-Preserving Visual Condition for Personalized Text-to-Image Generation
Shaozhe Hao, Kai Han, Shihao Zhao, Kwan-Yee K. Wong
arXiv preprint, 2023.
BibTeX PDF arXiv Code 
@article{hao2023ViCo,
 author    = {Shaozhe Hao and Kai Han and Shihao Zhao and Kwan-Yee K. Wong},
 title     = {ViCo: Detail-Preserving Visual Condition for Personalized Text-to-Image Generation},
 journal   = {arXiv preprint arXiv:2306.00971},
 year      = {2023},
}
SimSC: A Simple Framework for Semantic Correspondence with Temperature Learning
Xinghui Li, Kai Han, Xingchen Wan, Victor Adrian Prisacariu
arXiv preprint, 2023.
BibTeX PDF arXiv 
@article{li23SimSC,
  title    = {SimSC: A Simple Framework for Semantic Correspondence with Temperature Learning},
  author   = {Xinghui Li and Kai Han and Xingchen Wan and Victor Adrian Prisacariu},
  journal  = {arXiv preprint arXiv:2305.02385},
  year     = {2023},
}

2022

Novel Class Discovery without Forgetting
K J Joseph, Sujoy Paul, Gaurav Aggarwal, Soma Biswas, Piyush Rai, Kai Han, Vineeth N Balasubramanian
European Conference on Computer Vision (ECCV), 2022.
BibTeX PDF arXiv 
@inproceedings{joseph22ncdwf,
 author    = {K J Joseph and Sujoy Paul and Gaurav Aggarwal and Soma Biswas and Piyush Rai and Kai Han and Vineeth N Balasubramanian},
 title     = {Novel Class Discovery without Forgetting},
 booktitle = {European Conference on Computer Vision (ECCV)},
 year      = {2022},
}
Generalized Category Discovery
Sagar Vaze, Kai Han, Andrea Vedaldi, Andrew Zisserman
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2022.
BibTeX PDF arXiv Project Code 
@inproceedings{vaze22generalized,
 author    = {Sagar Vaze and Kai Han and Andrea Vedaldi and Andrew Zisserman},
 title     = {Generalized Category Discovery},
 booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 year      = {2022},
}
JIFF: Jointly-aligned Implicit Face Function for High Quality Single View Clothed Human Reconstruction
Yukang Cao, Guanying Chen, Kai Han, Wenqi Yang, Kwan-Yee K. Wong
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2022.
Oral presentation · 4.2% of submissions
BibTeX PDF arXiv Project Code 
@inproceedings{cao22jiff,
 author    = {Yukang Cao and Guanying Chen and Kai Han and Wenqi Yang and Kwan-Yee K. Wong},
 title     = {JIFF: Jointly-aligned Implicit Face Function for High Quality Single View Clothed Human Reconstruction},
 booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 year      = {2022},
}
SharpContour: A Contour-based Boundary Refinement Approach for Efficient and Accurate Instance Segmentation
Chenming Zhu, Xuanye Zhang, Yanran Li, Liangdong Qiu, Kai Han, Xiaoguang Han
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2022.
BibTeX PDF arXiv Project 
@inproceedings{zhu22sharpcontour,
 author    = {Chenming Zhu and Xuanye Zhang and Yanran Li and Liangdong Qiu and Kai Han and Xiaoguang Han},
 title     = {SharpContour: A Contour-based Boundary Refinement Approach for Efficient and Accurate Instance Segmentation},
 booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 year      = {2022},
}
Spacing Loss for Discovering Novel Categories
K J Joseph, Sujoy Paul, Gaurav Aggarwal, Soma Biswas, Piyush Rai, Kai Han, Vineeth N Balasubramanian
CVPR Workshop on Continual Learning in Computer Vision, 2022.
Best Paper Runner-Up Award BibTeX PDF arXiv 
@inproceedings{joseph22spacing,
 author    = {K J Joseph and Sujoy Paul and Gaurav Aggarwal and Soma Biswas and Piyush Rai and Kai Han and Vineeth N Balasubramanian},
 title     = {Spacing Loss for Discovering Novel Categories},
 booktitle = {CVPR Workshop on Continual Learning in Computer Vision},
 year      = {2022},
}
Open-Set Recognition: A Good Closed-Set Classifier is All You Need?
Sagar Vaze, Kai Han, Andrea Vedaldi, Andrew Zisserman
International Conference on Learning Representations (ICLR), 2022.
Oral presentation · 1.6% of submissions
BibTeX PDF arXiv OpenReview Project Code 
@inproceedings{vaze22openset,
 author    = {Sagar Vaze and Kai Han and Andrea Vedaldi and Andrew Zisserman},
 title     = {Open-Set Recognition: A Good Closed-Set Classifier is All You Need?},
 booktitle = {International Conference on Learning Representations (ICLR)},
 year      = {2022},
}

2021

Novel Visual Category Discovery with Dual Ranking Statistics and Mutual Knowledge Distillation
Bingchen Zhao, Kai Han
Conference on Neural Information Processing Systems (NeurIPS), 2021.
BibTeX PDF arXiv OpenReview Supp Project Code 

@inproceedings{zhao21novel,
 author     = {Bingchen Zhao and Kai Han},
 title      = {Novel Visual Category Discovery with Dual Ranking Statistics and Mutual Knowledge Distillation},
 booktitle  = {Conference on Neural Information Processing Systems (NeurIPS)},
 year       = {2021}
}
Joint Representation Learning and Novel Category Discovery on Single- and Multi-modal Data
Xuhui Jia, Kai Han, Yukun Zhu, Bradley Green
International Conference on Computer Vision (ICCV), 2021.
BibTeX PDF arXiv 

@inproceedings{jia21joint,
 author    = {Xuhui Jia and Kai Han and Yukun Zhu and Bradley Green},
 title     = {Joint Representation Learning and Novel Category Discovery on Single- and Multi-modal Data},
 booktitle = {International Conference on Computer Vision (ICCV)},
 year      = {2021}
}
𝕏Resolution Correspondence Networks
Georgi Tinchev, Shuda Li, Kai Han, David Mitchell, Rigas Kouskouridas
British Machine Vision Conference (BMVC), 2021.
BibTeX PDF arXiv Project Code 

@inproceedings{tinchev20xresolution,
 author    = {Georgi Tinchev and Shuda Li and Kai Han and David Mitchell and Rigas Kouskouridas},
 title     = {{$\mathbb{X}$}Resolution Correspondence Networks},
 booktitle = {British Machine Vision Conference (BMVC)},
 year      = {2021}
}
LSD-C: Linearly Separable Deep Clusters
Sylvestre-Alvise Rebuffi*, Sebastien Ehrhardt*, Kai Han*, Andrea Vedaldi, Andrew Zisserman
ICCV Workshop on Visual Inductive Priors for Data-Efficient Deep Learning, 2021. (* indicates equal contribution.)
BibTeX PDF arXiv Code 

@inproceedings{rebuffi21lsdc,
 author    = {Sylvestre-Alvise Rebuffi and Sebastien Ehrhardt and Kai Han and Andrea Vedaldi and Andrew Zisserman},
 title     = {LSD-C: Linearly Separable Deep Clusters},
 booktitle = {ICCV Workshop on Visual Inductive Priors for Data-Efficient Deep Learning},
 year      = {2021}
}
Contrastive Learning based Hybrid Networks for Long-Tailed Image Classification
Peng Wang, Kai Han, Xiu-Shen Wei, Lei Zhang, Lei Wang
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021.
BibTeX PDF arXiv 
@inproceedings{wang21contrastive,
 author     = {Peng Wang and Kai Han and Xiu-Shen Wei and Lei Zhang and Lei Wang},
 title      = {Contrastive Learning based Hybrid Networks for Long-Tailed Image Classification},
 booktitle  = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 year       = {2021},
}
AutoNovel: Automatically Discovering and Learning Novel Visual Categories
Kai Han, Sylvestre-Alvise Rebuffi, Sebastien Ehrhardt, Andrea Vedaldi, Andrew Zisserman
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2021.
BibTeX PDF arXiv DOI Project Code 
@article{han21autonovel,
 author  = {Kai Han and Sylvestre-Alvise Rebuffi and Sebastien Ehrhardt and Andrea Vedaldi and Andrew Zisserman},
 title   = {AutoNovel: Automatically Discovering and Learning Novel Visual Categories},
 journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)},
 year    = {2021},
}
Anisotropic Convolutional Neural Networks for RGB-D based Semantic Scene Completion
Jie Li, Peng Wang, Kai Han, Yu Liu
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2021.
BibTeX PDF arXiv DOI Project Code 
@article{li21anisotropic,
 author  = {Jie Li and Peng Wang and Kai Han and Yu Liu},
 title   = {Anisotropic Convolutional Neural Networks for RGB-D based Semantic Scene Completion},
 journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)},
 year    = {2021},
}
Fixed Viewpoint Mirror Surface Reconstruction under an Uncalibrated Camera
Kai Han, Miaomiao Liu, Dirk Schnieders, Kwan-Yee K. Wong
IEEE Transactions on Image Processing (TIP), 2021.
BibTeX PDF arXiv DOI Supp Project Code 
@article{han21fixed,
 title   = {Fixed Viewpoint Mirror Surface Reconstruction under an Uncalibrated Camera},
 author  = {Kai Han and Miaomiao Liu and Dirk Schnieders and Kwan-Yee K. Wong},
 journal = {IEEE Transactions on Image Processing (TIP)},
 year    = {2021}
}

2020

Dual-Resolution Correspondence Networks
Xinghui Li, Kai Han, Shuda Li, Victor Prisacariu
Conference on Neural Information Processing Systems (NeurIPS), 2020.
BibTeX PDF arXiv Supp Project Code 
@inproceedings{li20dualrc,
 author     = {Xinghui Li and Kai Han and Shuda Li and Victor Prisacariu},
 title      = {Dual-Resolution Correspondence Networks},
 booktitle  = {Conference on Neural Information Processing Systems (NeurIPS)},
 year       = {2020},
}
Automatically Discovering and Learning New Visual Categories with Ranking Statistics
Kai Han*, Sylvestre-Alvise Rebuffi*, Sebastien Ehrhardt*, Andrea Vedaldi, Andrew Zisserman
International Conference on Learning Representations (ICLR), 2020. (* indicates equal contribution.)
BibTeX PDF arXiv OpenReview Video Project Code 
@inproceedings{han20automatically,
 author    = {Kai Han and Sylvestre-Alvise Rebuffi and Sebastien Ehrhardt and Andrea Vedaldi and Andrew Zisserman},
 title     = {Automatically Discovering and Learning New Visual Categories with Ranking Statistics},
 booktitle = {International Conference on Learning Representations (ICLR)},
 year      = {2020},
}
Correspondence Networks with Adaptive Neighbourhood Consensus
Shuda Li*, Kai Han*, Theo W. Costain, Henry Howard-Jenkins, Victor Prisacariu
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020. (* indicates equal contribution.)
BibTeX PDF arXiv Project Code 
@inproceedings{li20correspondence,
 author    = {Shuda Li and Kai Han and Theo W. Costain and Henry Howard-Jenkins and Victor Prisacariu},
 title     = {Correspondence Networks with Adaptive Neighbourhood Consensus},
 booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 year      = {2020},
}
Anisotropic Convolutional Networks for 3D Semantic Scene Completion
Jie Li, Kai Han, Peng Wang, Yu Liu, Xia Yuan
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020.
BibTeX PDF arXiv Project Code 
@inproceedings{li20anisotropic,
 author    = {Jie Li and Kai Han and Peng Wang and Yu Liu and Xia Yuan},
 title     = {Anisotropic Convolutional Networks for 3D Semantic Scene Completion},
 booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 year      = {2020},
}
Semi-Supervised Learning with Scarce Annotations
Sylvestre-Alvise Rebuffi*, Sebastien Ehrhardt*, Kai Han*, Andrea Vedaldi, Andrew Zisserman
CVPR Deep Vision Workshop, 2020. (* indicates equal contribution.)
BibTeX PDF arXiv Project Code 
@inproceedings{rebuffi20SSL,
 title={Semi-Supervised Learning with Scarce Annotations},
 author={Sylvestre-Alvise Rebuffi and Sebastien Ehrhardt and Kai Han and  Andrea Vedaldi and Andrew Zisserman},
 booktitle={CVPR Deep Vision Workshop},
 year={2020}
}
Deep Photometric Stereo for Non-Lambertian Surfaces
Guanying Chen, Kai Han, Boxin Shi, Yasuyuki Matsushita, Kwan-Yee K. Wong
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2020.
BibTeX PDF arXiv DOI Supp Project Code 
@article{chen20deepps,
 title   = {Deep Photometric Stereo for Non-Lambertian Surfaces},
 author  = {Guanying Chen and Kai Han and Boxin Shi and Yasuyuki Matsushita and Kwan-Yee K. Wong},
 journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)},
 year    = {2020}
}

2019

Learning to Discover Novel Visual Categories via Deep Transfer Clustering
Kai Han, Andrea Vedaldi, Andrew Zisserman
International Conference on Computer Vision (ICCV), 2019.
BibTeX PDF arXiv Supp Project Code 
@inproceedings{han19DTC,
 author    = {Kai Han and Andrea Vedaldi and Andrew Zisserman},
 title     = {Learning to Discover Novel Visual Categories via Deep Transfer Clustering},
 booktitle = {International Conference on Computer Vision (ICCV)},
 year      = {2019}
}
Unsupervised Image Matching and Object Discovery as Optimization
Huy V. Vo, Francis Bach, Minsu Cho, Kai Han, Yann LeCun, Patrick Pérez, Jean Ponce
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019.
BibTeX PDF arXiv Code 
@inproceedings{vo19unsup,
 title     = {Unsupervised Image Matching and Object Discovery as Optimization},
 author    = {Huy V. Vo and Francis Bach and Minsu Cho and Kai Han and Yann LeCun and Patrick P\'{e}rez and Jean Ponce},
 booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 year      = {2019}
}
Self-calibrating Deep Photometric Stereo Networks
Guanying Chen, Kai Han, Boxin Shi, Yasuyuki Matsushita, Kwan-Yee K. Wong
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019.
Oral presentation · 5.6% of submissions
BibTeX PDF arXiv Video Poster Project Code 
@inproceedings{chen19SDPS_Net,
 title     = {Self-calibrating Deep Photometric Stereo Networks},
 author    = {Guanying Chen and Kai Han and Boxin Shi and Yasuyuki Matsushita and Kwan-Yee K. Wong},
 booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 year      = {2019}
}
Learning Transparent Object Matting
Guanying Chen*, Kai Han*, Kwan-Yee K. Wong
International Journal of Computer Vision (IJCV), 2019. (* indicates equal contribution.)
BibTeX PDF arXiv DOI Project Code 
@article{chen19LTOM,
 title   = {Learning Transparent Object Matting},
 author  = {Guanying Chen and Kai Han and Kwan-Yee K. Wong},
 journal = {International Journal of Computer Vision (IJCV)},
 year    = {2019}
}

2018

PS-FCN: A Flexible Learning Framework for Photometric Stereo
Guanying Chen, Kai Han, Kwan-Yee K. Wong
European Conference on Computer Vision (ECCV), 2018.
BibTeX PDF arXiv Video Poster Project Code 
@inproceedings{chen18ps_fcn,
 title     = {PS-FCN: A Flexible Learning Framework for Photometric Stereo},
 author    = {Guanying Chen and Kai Han and Kwan-Yee K. Wong},
 booktitle = {European Conference on Computer Vision (ECCV)},
 year      = {2018}
}
TOM-Net: Learning Transparent Object Matting from a Single Image
Guanying Chen*, Kai Han*, Kwan-Yee K. Wong
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018. (* indicates equal contribution.)
Spotlight presentation · 6.7% of submissions
BibTeX PDF arXiv Video Poster Project Code 
@inproceedings{chen18tom_net,
 title     = {TOM-Net: Learning Transparent Object Matting from a Single Image},
 author    = {Guanying Chen and Kai Han and Kwan-Yee K. Wong},
 booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 year      = {2018}
}
Dense Reconstruction of Transparent Objects by Altering Incident Light Paths Through Refraction
Kai Han, Kwan-Yee K. Wong, Miaomiao Liu
International Journal of Computer Vision (IJCV), 2018.
BibTeX PDF arXiv DOI 
@article{han18dense,
 title   = {Dense Reconstruction of Transparent Objects by Altering Incident Light Paths through Refraction},
 author  = {Kai Han and Kwan-Yee K. Wong and Miaomiao Liu},
 journal = {International Journal of Computer Vision (IJCV)},
 year    = {2018}
}

2017

SCNet: Learning Semantic Correspondence
Kai Han, Rafael S. Rezende, Bumsub Ham, Kwan-Yee K. Wong, Minsu Cho, Cordelia Schmid, Jean Ponce
International Conference on Computer Vision (ICCV), 2017.
BibTeX PDF arXiv HAL Poster Project Code 
@inproceedings{han17scnet,
 title     = {SCNet: Learning Semantic Correspondence},
 author    = {Kai Han and Rafael S. Rezende and Bumsub Ham and Kwan-Yee K. Wong and Minsu Cho and Cordelia Schmid and Jean Ponce},
 booktitle = {International Conference on Computer Vision (ICCV)},
 year      = {2017}
}

2016

Single View 3D Reconstruction under an Uncalibrated Camera and an Unknown Mirror Sphere
Kai Han, Kwan-Yee K. Wong, Xiao Tan
International Conference on 3D Vision (3DV), 2016.
BibTeX PDF Poster 
@inproceedings{han16single,
 author    = {Kai Han and Kwan-Yee K. Wong and Xiao Tan},
 title     = {Single View 3D Reconstruction under an Uncalibrated Camera and an Unknown Mirror Sphere},
 booktitle = {International Conference on 3D Vision (3DV)},
 year      = {2016}
}
Mirror Surface Reconstruction Under an Uncalibrated Camera
Kai Han, Kwan-Yee K. Wong, Dirk Schnieders, Miaomiao Liu
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016.
BibTeX PDF Project Code 
@inproceedings{han16mirror,
 author    = {Kai Han and Kwan-Yee K. Wong and Dirk Schnieders and Miaomiao Liu},
 title     = {Mirror Surface Reconstruction under an Uncalibrated Camera},
 booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 year      = {2016}
}

2015

A Fixed Viewpoint Approach for Dense Reconstruction of Transparent Objects
Kai Han, Kwan-Yee K. Wong, Miaomiao Liu
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015.
BibTeX PDF 
@inproceedings{han15afixed,
 author    = {Kai Han and Kwan-Yee K. Wong and Miaomiao Liu},
 title     = {A Fixed Viewpoint Approach for Dense Reconstruction of Transparent Objects},
 booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
 year      = {2015}
}

PhD Dissertation

Single view reconstruction of transparent, mirror and diffuse surfaces
Kai Han
The University of Hong Kong, Pokfulam, Hong Kong, 2018.
BibTeX HKU Theses Online 
@phdthesis{han18phd,
 author = {Kai Han},
 title  = {Single view reconstruction of transparent, mirror and diffuse surfaces},
 school = {Department of Computer Science, The University of Hong Kong},
 year   = {2018}
}

Teaching

Awards and Honors

Professional Activities

Area Chair
Journal Editorship
Workshop Organization
Senior Program Committee
Conference Reviewer

CVPR (2018–2023) · ICCV (2019, 2021, 2023) · ECCV (2020, 2022) · NeurIPS (2022–2024) · ICLR (2022–2024) · ICML (2023–2026) · ACL (2026) · SIGGRAPH Asia (2024–2025) · ICRA (2023) · AAAI (2020–2022) · ACCV (2020, 2022)

Journal Reviewer

TPAMI · IJCV · TOG · JMLR · TIP · etc.