李秀

研究方向为人工智能、计算机视觉、强化学习。在国内外重要学术期刊或会议上发表学术论文200余篇，包括 CVPR, NeurIPS, ICML,ICLR,等人工智能领域顶级国际会议。论文在google scholar上被引用1万余次，2022年度~2024年度连续三年入选美国斯坦福大学与爱思唯尔数据库（Elsevier）联合发布的人工智能领域全球前2%顶尖科学家榜单。

2016年12月至今，清华大学深圳国际研究生院信息学部，教授

2016年06月-2017年2月，美国加州大学欧文分校计算机科学系，访问学者

2010年02月-2016年11月，清华大学深圳研究生院信息学部，副研究员

2007年10月-2008年10月，美国佐治亚理工学院，访问学者

2006年09月-2007年02月，香港理工大学，高级访问学者

2005年07月-2005年09月，香港大学访问学者

2003年12月-2010年01月，清华大学自动化系，副研究员

2002年05月-2003年11月，清华大学自动化系，讲师

2000年04月-2002年04月，清华大学自动化系，博士后 

中国人工智能学会智能融合专委会，秘书长

深圳市人工智能学会，副理事长

深圳市新一代互动媒体技术创新重点实验室，主任

《人工智能技术前沿与产业应用》

《人工智能基础》

《互联网思维与技术》

《人工智能实践课》负责人

具身智能：多智能体强化学习、人形机器人动态交互与控制

多模态理解与生成大模型：精准可控视频内容生成、2D/3D内容生成与编辑、共语动作驱动的数字人生成

AI4S：人工智能与生命科学的交叉研究

Group Website: THUSIGSICLAB

For the complete publication list, please refer to:

ORCID: 0000-0003-0403-1923

Google scholar: Xrh1OIUAAAAJ&hl=en&oi=ao

近三年代表作：

· SCI检索期刊论文：

【1】 Chunming He*, Yuqi Shen*, Chengyu Fang*, Fengyang Xiao, Longxiang Tang, Yulun Zhang, Wangmeng Zuo, Senior Member, IEEE, Zhenhua Guo, Xiu Li†. Diffusion Models in Low-Level Vision: A Survey[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2025, vol.47, no.6, pp.4630-4651.WOS:001484716600013. [Page]

【2】Hantao Zhou, Rui Yang, Yachao Zhang, Haoran Duan, Yawen Huang, Runze Hu†, Xiu Li†, Yefeng Zheng, Fellow IEEE. UniHead: Unifying Multi-Perception for Detection Heads[J]. IEEE Transactions on Neural Networks and Learning Systems, 2024,vol.36, no.5, pp.9565-9576. WOS:001252659600001. EI Accession number: 20242616353982 [Page] [Code]

【3】Jiafei Lyu, Le Wan, Zongqing Lu† and Xiu Li†. Off-Policy RL Algorithms Can be Sample-Efficient for Continuous Control via Sample Multiple Reuse[J]. Information Sciences, 2024, Volume:666. WOS:001210302700001. EI Accession number: 20241115739541[Page] [Code]

【4】 Jiafei Lyu, Le Wan, Xiu Li†s and Zongqing Lu. Understanding what affects generalization gap in visual reinforcement learning: Theory and empirical evidence[J]. Journal of Artificial Intelligence Research, 2024, Volume:80. WOS:001318537600001. EI Accession number: 20244117178921[Page]

【5】 Mengbei Yan*, Jiafei Lyu*, Xiu Li†. Enhancing visual reinforcement learning with State–Action Representation[J]. Knowledge-Based Systems, 304 (2024): 112487. WOS:001320006700001. EI Accession number: 20243917083150 [Page]

【6】 Aicheng Gong*, Kai Yang*, Jiafei Lyu, Xiu Li†. A two-stage reinforcement learning-based approach for multi-entity task allocation[J]. Engineering Applications of Artificial Intelligence, 136 (2024): 108906. WOS:001348574200001. EI Accession number: 20242816685744 [Page]

【7】Jingyi Tang, Zeyu Chen, Bowen Fu, Wenjie Lu, Shengquan Li, Xiu Li†, and Xiangyang Ji†. ROV6D: 6D Pose Estimation Benchmark Dataset for Underwater Remotely Operated Vehicles[J]. IEEE Robotics and Automation Letters, 2024, vol.9, no.1, pp.65-72. WOS:001257126000001. EI Accession number: 20234715087683 [Data]

【8】He C, Li K†, Xu G, Yan J, Tang L, Zhang Y, Wang Y, Li X†. Hqg-net: Unpaired medical image enhancement with high-quality guidance[J]. IEEE Transactions on Neural Networks and Learning Systems, 2023. WOS:001165741200001. EI Accession number: 20234314966831 [Page] [Code] [Data]

【9】 Meng C, Zhao Z, Guo W, Zhang Y, Wu H, Gao C, Li D, Li X†, et al. Coarse-to-fine knowledge-enhanced multi-interest learning framework for multi-behavior recommendation[J]. ACM Transactions on Information Systems, 2023, 42(1): 1-27. WOS:001040640300001. EI Accession number: 20234715075803 [Page] [Code] [Data]

【10】Lan S†, Li X, Guo Z†. An Adaptive Region-Based Transformer for Nonrigid Medical Image Registration With a Self-Constructing Latent Graph[J]. IEEE Transactions on Neural Networks and Learning Systems, 2023. WOS:001040640300001. EI Accession number: 20233114462548 [Page]

【11】Yu B, Li X†, Li W, Zhou J, Lu J. Discrepancy-Aware Meta-Learning for Zero-Shot Face Manipulation Detection[J]. IEEE Transactions on Image Processing, 2023, vol.32, pp.3759-3773. WOS:001028969300001. EI Accession number: 20232814386176 [Page]

【12】Zhang T, Lin Z, Wang Y, Ye D, Fu Q, Yang W, Wang X, Liang B, Yuan B†, Li X†. Dynamics-Adaptive Continual Reinforcement Learning via Progressive Contextualization[J]. IEEE Transactions on Neural Networks and Learning Systems, 2023. WOS:001005843700001. EI Accession number: 20232414220584 [Page]

【13】 Yuan Z, Pan W, Zhao X, Zhao F, Xu Z, Li X, Zhao Yi, Zhang Michael Q†, Yao Jianhua†. Publisher Correction: SODB facilitates comprehensive exploration of spatial omics data[J]. Nature methods, 2023, 20(4): 623. WOS:000953332800001. [Page]

【14】Tang M, Wang Z, Zeng Z, Li X†, et al. Stay in Grid: Improving Video Captioning via Fully Grid-level Representation[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2022, vol.33, no.7, pp.3319-3332. WOS:001022165700021. EI Accession number: 20230313395050 [Page]

【15】Huo Y, Li Xiang†, Zhang X, Li Xiu, et al. Adaptive Intention-Driven Variable Impedance Control for Wearable Robots With Compliant Actuators[J]. IEEE Transactions on Control Systems Technology, 2022, 31(3): 1308-1323. WOS:000890819200001. EI Accession number: 20225113272426 [Page]

【16】Yan J*, Chen H*, Li X†, Yao Jianhua†. Deep contrastive learning based tissue clustering for annotation-free histopathology image analysis[J]. Computerized Medical Imaging and Graphics, 2022, 97: 102053. JCR Q1. WOS:000787887200003. EI Accession number: 20221211816207  [Page]

· CCF-A类会议论文：

【1】 Yukang Lin*, Hokit Fung*, Jianjin Xu, Zeping Ren, Adela S.M. Lau, Guosheng Yin†, Xiu Li†. MVPortrait: Text-Guided Motion and Emotion Control for Multi-view Vivid Portrait Animation[C]. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR-25), 2025.[Page]

【2】Zunnan Xu, Zhentao Yu, Zixiang Zhou, Jun Zhou, Xiaoyu Jin, Fa-Ting Hong, Xiaozhong Ji, Junwei Zhu, Chengfei Cai, Shiyu Tang, Qin Lin, Xiu Li†, Qinglin Lu†. HunyuanPortrait: Implicit Condition Control for Enhanced Portrait Animation[C]. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR-25), 2025.[Page] [Code] [Demo]

【3】Haonan Han*, Xiangzuo Wu*, Huan Liao*, Zunnan Xu, Zhongyuan Hu, Ronghui Li, Yachao Zhang†, Xiu Li†. AToM: Aligning Text-to-Motion Model at Event-Level with GPT-4Vision Reward[C]. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR-25), 2025. EI Accession number: 20240517811 [Page]

【4】Zinqin Huang, Gu Wang, Chenyangguang Zhang, Ruida Zhang, Xiu Li†, Xiangyang Ji. GIVEPose: Gradual Intra-class Variation Elimination for RGB-based Category-Level Object Pose Estimation[C]. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR-25), 2025. [Page]

【5】Jiafei Lyu, Kang Xu, Jiacheng Xu, Mengbei Yan, Jingwen Yang, Zongzhang Zhang, Chenjia Bai†, Zongqing Lu, Xiu Li†. ODRL: A Benchmark for Off-Dynamics Reinforcement Learning[C]. In The Thirty-eight Conference on Neural Information Processing Systems Datasets and Benchmarks Track（NeurIPS-24）, 2024. EI Accession number:20240444723 [Page]

【6】 Runze Liu, Yali Du†, Fengshuo Bai, Jiafei Lyu, Xiu Li†. PEARL: Zero-shot Cross-task Preference Alignment and Robust Reward Learning for Robotic Manipulation[C].  In International Conference on Machine Learning（ICML-24）, 2024. EI Accession number:20243817050165 [Page]

【7】 Kai Yang*, Jian Tao*, Jiafei Lyu, Xiu Li†. Exploration and Anti-Exploration with Distributional Random Network Distillation[C]. In Forty-first International Conference on Machine Learning（ICML-24）, 2024. INSPEC:24464620. EI Accession number: 20243817052883[Page]

【8】  Jiafei Lyu, Chenjia Bai, Jingwen Yang, Zongqing Lu, Xiu Li†. Cross-Domain Policy Adaptation by Capturing Representation Mismatch[C]. In Forty-first International Conference on Machine Learning（ICML-24）, 2024, 2024. INSPEC:25316967. EI Accession number: 20240229502[Page] [Code]

【9】 Zunnan Xu, Yukang Lin, Haonan Han, Sicheng Yang, Ronghui Li, Yachao Zhang†, Xiu Li†. MambaTalk: Efficient Holistic Gesture Synthesis with Selective State Space Models[C]. Advances in Neural Information Processing Systems (NeurIPS-24), 2024. INSPEC:24837391. EI Accession number:20240124955[Page]

【10】Jiangshan Wang*, Yue Ma*, Jiayi Guo*, Yicheng Xiao, Gao Huang†, Xiu Li†. COVE: Unleashing the Diffusion Feature Correspondence for Consistent Video Editing[C]. Advances in Neural Information Processing Systems (NeurIPS-24), 2024. INSPEC:25205832. EI Accession number:20240265709[Page] [Code]

【11】 Chengyu Fang*, Fengyang Xiao, Chunming He*,†, Yulun Zhang†, Longxiang Tang, Yuelin Zhang, Kai Li, Xiu Li†. Real-world Image Dehazing with Coherence-based Label Generator and Cooperative Unfolding Network[C]. Advances in Neural Information Processing Systems (NeurIPS-24), 2024. INSPEC:25204966. EI Accession number:20240261457 [Page] [Code]

【12】 Yicheng Xiao, Lin Song, Shaoli Huang, Jiangshan Wang, Siyu Song, Yixiao Ge, Xiu Li, Ying Shan. MambaTree: Tree Topology is All You Need in State Space Model[C]. The Thirty-eighth Annual Conference on Neural Information Processing Systems(NeurIPS-24), 2024.[Page] [Code]

【13】Ronghui Li, YuXiang Zhang, Yachao Zhang, Hongwen Zhang, Jie Guo, Yan Zhang, Yebin Liu, Xiu Li†. Lodge: A Coarse to Fine Diffusion Network for Long Dance Generation Guided by the Characteristic Dance Primitives[C]. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR-24), 2024. INSPEC:24842557. EI Accession number: 20240129491[Page] [Code] [Data]

【14】Kai Yang*, Jian Tao*, Jiafei Lyu†, Chunjiang Ge, Qimai Li, Jiaxin Chen, Weihan Shen, Xiaolong Zhu, Xiu Li†. Using Human Feedback to Fine-tune Diffusion Models without Any Reward Model[C]. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR-24), 2024. INSPEC:24150778. EI Accession number: 20230432990[Page] [Code] [Data]

【15】 Yicheng Xiao*, Zhuoyan Luo*, Yong Liu, Yue Ma, Hengwei Bian, Yatai Ji, Yujiu Yang†, Xiu Li†. Bridging the Gap: A Unified Video Comprehension Framework for Moment Retrieval and Highlight Detection[C]. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR-24), 2024. INSPEC:24221468. EI Accession number: 20230428618[Page] [Code] [Data]

【16】 Zeping Ren，Shaoli Huang†, Xiu Li†. Realistic Human Motion Generation with Cross-Diffusion Models[C]. European Conference on Computer Vision (ECCV-24), 2024. EI Accession number: 20230458804[Page] [Code] [Data]

【17】 Yifan Pu*, Zhuofan Xia*, Jiayi Guo, Dongchen Han, Qixiu Li, Duo Li, Yuhui Yuan, Ji Li, Yizeng Han, Shiji Song, Gao Huang†, Xiu Li†. Efficient Diffusion Transformer with Step-wise Dynamic Attention Mediators[C]. European Conference on Computer Vision (ECCV-24), 2024. INSPEC:25484352. EI Accession number: 20240346913[Page] [Code]

【18】Jiangshan Wang*, Yifan Pu*, Yizeng Han, Jiayi Guo, Yiru Wang, Xiu Li†, Gao Huang†. GRA: Detecting Oriented Objects through Group-wise Rotating and Attention[C]. European Conference on Computer Vision (ECCV-24), 2024. EI Accession number: 20240127662[Page]

【19】Longxiang Tang, Zhuotao Tian, Kai Li, Chunming He, Hantao Zhou, Hengshuang Zhao, Xiu Li†, Jiaya Jia. Mind the Interference: Retaining Pre-trained Knowledge in Parameter Efficient Continual Learning of Vision-Language Models[C]. European Conference on Computer Vision (ECCV-24), 2024. EI Accession number:20240305908[Page] [Code] [Data]

【20】Yachao Zhang, Runze Hu, Ronghui Li, Yanyun Qu, Yuan Xie, Xiu Li†. Cross-Modal Match for Language Conditioned 3D Object Grounding[C]. In Proceedings of the Association for the Advance of Artificial Intelligence (AAAI-24), 2024. WOS:001239937300096. EI Accession number: 20241515870430[Page]

【21】Yue Ma*, Yingqing He*, Xiaodong Cun, Xintao Wang, Siran Chen, Ying Shan, Xiu Li†, Chen Qifeng†. Follow Your Pose: Pose-Guided Text-to-Video Generation using Pose-Free Videos[C]. In Proceedings of the Association for the Advance of Artificial Intelligence (AAAI-24), 2024. INSPEC:23392848. EI Accession number: 0241515870526[Page] [Code] [Demo]

【22】Zunnan Xu, Yachao Zhang, Sicheng Yang, Ronghui Li, Xiu Li†. Chain of Generation: Multi-Modal Gesture Synthesis via Cascaded Conditional Control[C]. In Proceedings of the Association for the Advance of Artificial Intelligence (AAAI-24), 2024. WOS:001239936300144. EI Accession number: 20241515867359 [Page] [Data]

【23】C He, K Li†, Y Zhang, L Tang, Y Zhang, X Li†. Camouflaged object detection with feature decomposition and edge reconstruction[C]. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR-23), 2023: 22046-22055. WOS:001062531306037. EI Accession number: 20234114867985 [Page] [Code] [Data]

【24】C He*, K Li*, Y Zhang, L Tang, Y Zhang, Z Guo, X Li†. Weakly-Supervised Concealed Object Segmentation with SAM-based Pseudo Labeling and Multi-scale Feature Grouping[C]. Advances in Neural Information Processing Systems (NeurIPS-23), 2023. WOS:001230083404048. EI Accession number: 20230197582 [Page] [Code] [Data]

【25】C He, K Li†, G Xu, Y Zhang, R Hu, Z Guo, X Li†. Degradation-Resistant Unfolding Network for Heterogeneous Image Fusion[C]. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV-23), 2023: 12611-12621. WOS:001169499005005. EI Accession number: 20240915635855 [Page] [Code] [Data] [Demo]

【26】Rui Yang*, Lin Song*,†, Yixiao Ge, Xiu Li†. BoxSnake: Polygonal Instance Segmentation with Box Supervision[C]. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV-23), 2023: 2303.11630. WOS:001159644301002. EI Accession number: 20230100468[Page] [Code] [Data]

1. 李秀；贾若楠；基于一致性约束建模的强化学习机器人控制方法及系统，中国，ZL202110768179.5

2. 李秀；贾若楠；减少过估计的模型化强化学习机器人控制方法及系统，中国，ZL202110757340.9

3. 李秀；杨锐；欧奕旻；严江鹏；一种海面船舶检测方法及系统，中国，ZL202111135426.4

4. 李秀；许菁；王梦凯；一种基于核主成分分析和LDA的主题分析方法及系统，中国，ZL202110709322.3

5. 李秀；许菁；一种基于多尺度注意网络的多标签胸部疾病图像分类方法，中国，202110514525.7

6. 李秀；马露凡；陶佳琪；一种动态分辨率实例分割方法及计算机可读存储介质，中国，202110400888.8

7. 李秀；宋恺祥；一种基于图像修复技术的弱监督语义分割方法和装置，中国，ZL202010129164.X

8. 李秀；吕加飞；杨瑞；一种基于强化学习的压水堆堆芯自动控制方法，中国，ZL202110031428.2

9. 李秀；杨瑞；吕加飞；杨宇；基于动态模型与事后经验回放的多目标机器人控制方法，中国，ZL202011281615.8

10. 李秀；徐哲；罗凤；马露凡；严江鹏；一种基于循环正则训练的跨模态医学图像配准方法及装置，中国，ZL202010667204.6

11. 李秀；徐哲；马露凡；罗凤；严江鹏；一种跨模态医学图像配准方法及装置，中国，ZL202010652606.9

12. 李秀；王亚伟；张明；一种对抗式模仿学习中奖励函数的选择方法，中国，ZL202010323155.4

13. 李秀；潘昭鸣；一种生成可执行代码的方法及计算机可读存储介质，中国，ZL202010172236.9

14. 李秀；董九阳；一种共聚焦显微镜的成像方法，中国，ZL202010152737.0

15. 李秀；段桂春；行人重识别网络搜索方法及行人重识别方法，中国，ZL202010144613.8

16. 李秀；陈洪鑫；一种基于深度强化学习的视频编码帧内码率控制方法，中国，ZL202010080042.6

17. 李秀；张凌霄；一种基于树莓派的智能迎宾机器人装置，中国，ZL201922274774.4

18. 李秀；张凌霄；一种延时照明装置，中国，ZL201922269021.4

19. 李秀；张凌霄；一种用户行为表征的方法及系统，中国，ZL201911304558.8

20. 李秀；宋恺祥；适用于2D卷积神经网络的可学习引导滤波模块和方法；中国，ZL201910867312.5

21. 李秀；严江鹏；一种基于深度学习的欠采样核磁共振图像重建方法；中国，ZL201910784735.0

22. 李秀；金坤；一种基于深度学习和语义分割的图像检索方法；中国，ZL201810615664.7

23. 李秀；龙如蛟；一种基于深度网络的使网络注意到数据的重要部分的方法，中国，ZL201810891937.0

24. 李秀；刘志鑫；门畅；学习行为动态预测方法、装置、设备及存储介质，中国，ZL201811144725.2

25. 李秀；闫欣伟；一种中文虚假顾客评论识别方法，中国，ZL201510164626.0

26. 李秀；陈连胜；汤友华；一种克服静止前景运动目标检测的方法，中国，ZL201510548886.8

27. 李秀；欧阳小刚；陈连胜；宋靖东；一种水下图像并行分割方法及装置，中国，ZL201510221256.X

28. 李秀；陈连胜；汤友华；一种运动目标检测的方法，中国，ZL201510549568.3

29. 李秀；宋靖东；黄容生；李静；基于Kepler科学工作流传感网服务组合方法及装置，中国，ZL201510072274.6

30. 李秀；宋靖东；科学工作流调度处理方法及装置，中国，ZL201410302064.7

31. 李秀；闫天翔；高福信；余瑾；一种从非关系型数据库到关系型数据库的数据迁移方法，中国，ZL201310443352.X

32. 李秀；黄容生；郭振华；马辉；用于海底观测网仪器智能配置的云配置方法，中国，ZL201310467742.0

1、第50届日内瓦国际发明展金奖；

2、2025年度发明创业奖创新二等奖；

3、2022年度广东省科技进步一等奖；

4、2022年度深圳市自然科学二等奖；

5、2015年广东省科技进步一等奖；

6、2015年深圳市科技进步二等奖；

7、2013年中国电子学会科学技术二等奖