人员简介

吴保元

职务/职称

香港中文大学（深圳）副教授

研究方向

人工智能安全与隐私、机器学习、计算机视觉、最优化

电子邮箱

wubaoyuan@cuhk.edu.cn

教育背景

中国科学院自动化研究所模式识别国家重点实验室，博士

美国伦斯勒理工学院，国家公派留学联合培养博士

北京科技大学信息工程学院，学士

个人介绍

吴保元博士现任香港中文大学（深圳）数据科学学院副教授，深圳市模式分析与感知计算重点实验室（筹）副主任，并担任腾讯AI Lab可信AI中心顾问。其研究方向包括机器学习、计算机视觉、最优化，在人工智能的顶级期刊和会议上发表论文50多篇，并曾入选人工智能顶级会议CVPR 2019最佳论文候选名单。其担任人工智能领域国际期刊Neurocomputing编委、国际会议ICLR 2022、AAAI 2022、ICIG 2021 领域主席，国际会议AAAI 2021、IJCAI 2020/2021高级程序委员、中国自动化学会模式识别与机器智能专委会副秘书长。作为项目负责人承担国家自然科学基金面上项目1项，深圳市优秀科技创新人才优秀青年基础研究项目1项，CCF-腾讯犀牛鸟基金1项，CCF-海康威视斑头雁基金1项，腾讯研究专项基金2项。

代表性论文

Technical Report:

AI安全的威胁风险矩阵

Tencent AI Lab (Baoyuan Wu, Yanbo Fan, Yong Zhang, Yiming Li, Zhifeng Li, Wei Liu), Tencent Zhuque Lab (viking, jifengzhu, allenszch, ucasjh, dylan, xunsu). 2020/09/18.

 

Selected Journal Publications:

  1. Yiming Li, Baoyuan Wu (corresponding author), Yan Feng, Yanbo Fan, Yong Jiang, Zhifeng Li, Shutao Xia (corresponding author), “Semi-supervised Robust Training with Generalized Perturbed Neighborhood”, Pattern Recognition, 2021. (CCF B, IF 7.74)
  2. Baoyuan Wu, Li Shen, Tong Zhang, Bernard Ghanem, ” MAP Inference via L2-Sphere Linear Program Reformulation”, International Journal of Computer Vision (IJCV), 128, pages1913–1936 (2020). (CCF A, IF 7.41)
  3. Yuyang Zhang, Shibiao Xu, Baoyuan Wu, Jian Shi, Weiliang Meng, Xiaopeng Zhang, “Unsupervised Multi-view Constrained Convolutional Network for Accurate Depth Estimation”, IEEE Transactions on Image Processing, Volume 29, pages 7019-7031, 2020. (CCF A, IF 12.87)
  4. Zechun Liu, Wenhan Luo, Baoyuan Wu, Xin Yang, Wei Liu, Kwang-Ting Cheng, “Bi-Real Net: Binarizing Deep Network Towards Real-Network Performance”, International Journal of Computer Vision (IJCV), 128, pages 202–219 (2020). (CCF A, IF 7.41)
  5. Baoyuan Wu, Bernard Ghanem, “Lp-Box ADMM: A Versatile Framework for Integer Programming”, IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), Volume 41(7), Page 1695-1708, 2019. (CCF A, IF 17.73)
  6. Zechun Liu, Wenhan Luo, Baoyuan Wu, Xin Yang, Wei Liu and Kang-Ting Cheng, “Bi-Real Net: Binarizing Deep Network Towards Real-Network Performance”, accepted to International Journal of Computer Vision (IJCV), 2019. (CCF A, IF 11.541)
  7. Baoyuan Wu, Li Shen, Bernard Ghanem and Tong Zhang, “MAP Inference via L2-Sphere Linear Program Reformulation”, accepted to International Journal of Computer Vision (IJCV) (CCF A, IF 11.541, under review, major revision)
  8. Baoyuan Wu, Fan Jia, Wei Liu, Bernard Ghanem and Siwei Lyu, “Multi-label Learning with Missing Labels using Mixed Graphs”, International Journal of Computer Vision (IJCV), Volume 126 (8), Page 875-896, 2018. (CCF A, IF 11.541)
  9. Shibiao Xu, Xingjia Pan, Er Li, Baoyuan Wu, Shuhui Bu, Weiming Dong, Shiming Xiang, Xiaopeng Zhang, “Automatic Building Rooftop Extraction From Aerial Images via Hierarchical RGB-D Priors”, IEEE Transactions on Geoscience and Remote Sensing, Volume 56 (12), Page 7369-7387, 2018. (CCF B, IF 5.63)
  10. Baoyuan Wu, Bao-Gang Hu, Qiang Ji, “A coupled hidden markov random field model for simultaneous face clustering and tracking in videos", Pattern Recognition, Volume 64, Page 361-373, 2017. (CCF B, IF 5.898)
  11. Yifan Zhang, Zhiqiang Tang, Baoyuan Wu (corresponding author), Qiang Ji, Hanqing Lu, “A coupled hidden conditional random field model for simultaneous face clustering and naming in videos", IEEE Transactions on Image Processing, Volume 25 (12), Page 5780-5792, 2016. (CCF A, IF 6.79)
  12. Yongqiang Li, Baoyuan Wu (corresponding author), Bernard Ghanem, Yongping Zhao, Hongxun Yao, Qiang Ji, “Facial action unit recognition under incomplete data based on multi-label learning with missing labels", Pattern Recognition, Volume 60, Page 890-900, 2016. (CCF B, IF 5.898)
  13. Baoyuan Wu, Siwei Lyu, Bao-Gang Hu, Qiang Ji, “Multi-label learning with missing labels for image annotation and facial action unit recognition”, Pattern Recognition, Volume 48, Page 2279-2289, 2015. (CCF B, IF 5.898)

Selected Conference Papers:

  1. Zeyu Qin, Yanbo Fan, Hongyuan Zha, Baoyuan Wu (corresponding author), “Random Noise Defense Against Query-Based Black-Box Attacks”, Accepted to NeurIPS 2021, to appear. (CCF A)
  2. Yuezun Li, Yiming Li, Baoyuan Wu (corresponding author), Longkang Li, Ran He, Siwei Lyu (corresponding author), “Invisible Backdoor Attack with Sample-Specific Triggers”, Accepted to ICCV 2021, to appear. (CCF A)
  3. Siyuan Liang, Baoyuan Wu (corresponding author), Yanbo Fan, Xingxing Wei, Xiaochun Cao (corresponding author), “Parallel Rectangle Flip Attack: A Query-based Black-box Attack against Object Detection”, Accepted to ICCV 2021, to appear. (CCF A)
  4. Weiwei Feng, Baoyuan Wu (corresponding author), Tianzhu Zhang (corresponding author), Yong Zhang, Yongdong Zhang, “Meta-Attack: Class-agnostic and Model-agnostic Physical Adversarial Attack”, Accepted to ICCV 2021, to appear. (CCF A)
  5. Gengcong Yang, Jingyi Zhang, Yong Zhang, Baoyuan Wu (corresponding author), Yujiu Yang (corresponding author), “Probabilistic Modeling of Semantic Ambiguity for Scene Graph Generation”, Accepted to CVPR 2021. (CCF A)
  6. Xunguang Wang, Zheng Zhang, Baoyuan Wu, Fumin Shen, Guangming Lu, “Prototype-supervised Adversarial Network for Targeted Attack of Deep Hashing”, Accepted to CVPR 2021. (CCF A)
  7. Weihao Xia, Yujiu Yang, Jing-Hao Xue, Baoyuan Wu, “ediGAN: Text-Guided Diverse Face Image Generation and Manipulation”, Accepted to CVPR 2021. (CCF A)
  8. Jindong Gu, Baoyuan Wu, Volker Tresp, “Effective and Efficient Vote Attack on Capsule Networks”, Accepted to ICLR 2021. (CCF A)
  9. Jiawang Bai, Baoyuan Wu (corresponding author), Yong Zhang, Yiming Li, Zhifeng Li, Shu-Tao Xia (corresponding author), “Targeted Attack against Deep Neural Networks via Flipping Limited Weight Bits”, Accepted to ICLR 2021. (CCF A)
  10. Tongqing Zhai, Yiming Li, Ziqi Zhang, Baoyuan Wu, Yong Jiang, Shu-Tao Xia, “Backdoor Attack Against Speaker Verification”, Accepted to ICASSP 2021, to appear. (CCF B)
  11. Chengcheng Ma, Weiliang Meng, Baoyuan Wu, Shibiao Xu, Xiaopeng Zhang, “Towards Effective Adversarial Attack Against 3D Point Cloud Classification”, Accepted to ICME 2021, to appear. (CCF B)
  12. Ruisong Zhang, Weize Quan, Baoyuan Wu, Zhifeng Li, Dong-Ming Yan, “Pixel-wise Dense Detector for Image Inpainting”, Accepted to Pacific Graphics 2020, to appear. (CCF B)
  13. Chengcheng Ma, Weiliang Meng, Baoyuan Wu, Shibiao Xu, Xiaopeng Zhang, “Efficient Joint Gradient Based Attack Against SOR Defense for 3D Point Cloud Classification”, Accepted to ACM MM 2020, to appear. (CCF A)
  14. Yanbo Fan*, Baoyuan Wu* (co-first authors, corresponding author), Tuanhui Li, Yong Zhang, Mingyang Li, Zhifeng Li, Yujiu Yang, “Sparse Adversarial Attack via Perturbation Factorization”, European Conference on Computer Vision (ECCV), 2020. (CCF A)
  15. Weilun Chen, Zhaoxiang Zhang, Xiaolin Hu, Baoyuan Wu., “Boosting Decision-based Black-box Adversarial Attacks with Random Sign Flip”, European Conference on Computer Vision (ECCV), 2020. (CCF A)
  16. Junbing Li, Changqing Zhang, Pengfei Zhu, Baoyuan Wu, Lei Chen, Qinghua Hu, “SPL-MLL: Selecting Predictable Landmarks for Multi-Label Learning”, European Conference on Computer Vision (ECCV), 2020. (CCF A)
  17. Yong Zhang, Haiyong Jiang, Baoyuan Wu (corresponding author), Yanbo Fan and Qiang Ji. "Context-aware Feature and Label Fusion for Facial Action Unit Intensity Estimation with Partially Labeled Data", ICCV 2019. (CCF A)
  18. Yan Xu*, Baoyuan Wu* (co-first authors, corresponding author), Fumin Shen, Yanbo Fan, Yong Zhang, Heng Tao Shen and Wei Liu. "Exact Adversarial Attack to Image Captioning via Structured Output Learning with Latent Variables", CVPR 2019. (CCF A)
  19. Tuanhui Li, Baoyuan Wu (corresponding author), Yujiu Yang, Yanbo Fan, Yong Zhang, and Wei Liu. "Compressing Convolutional Neural Networks via Factorized Convolutional Filters", CVPR 2019. (CCF A)
  20. Yong Zhang, Baoyuan Wu (corresponding author), Weiming Dong, Zhifeng Li, Wei Liu, Bao-Gang Hu and Qiang Ji. "Joint Representation and Estimator Learning for Facial Action Unit Intensity Estimation", CVPR 2019. (CCF A)
  21. Yinpeng Dong, Hang Su, Baoyuan Wu, Zhifeng Li, Wei Liu, Tong Zhang and Jun Zhu. "Efficient Decision-based Black-box Adversarial Attacks on Face Recognition", CVPR 2019. (CCF A)
  22. Kaihua Tang, Hanwang Zhang, Baoyuan Wu, Wenhan Luo and Wei Liu. "Learning to Compose Dynamic Tree Structures for Visual Contexts", CVPR 2019 (Best Paper Finalist). (CCF A)
  23. Xin Li, Chao Ma, Baoyuan Wu, Zhenyu He and Ming-Hsuan Yang. "Target-Aware Deep Tracking", CVPR 2019. (CCF A)
  24. Jia Wan, Wenhan Luo, Baoyuan Wu, Antoni Chan and Wei Liu. "Residual Regression with Semantic Prior for Crowd Counting", CVPR 2019. (CCF A)
  25. Baoyuan Wu, Weidong Chen, Wei Liu, Peng Sun, Bernard Ghanem and Siwei Lyu, “Tagging Like Humans: Diverse and Distinct Image Annotation”, CVPR, 2018. (CCF A)
  26. Linchao Bao, Baoyuan Wu and Wei Liu, “Video Object Segmentation via Inference inA Higher-Order Spatio-Temporal MRF”, CVPR, 2018. (CCF A)
  27. Zechun Liu, Baoyuan Wu, Wenhan Luo, Xin Yang, Wei Liu and Kang-Ting Cheng, “Bi-Real Net: Enhancing the Performance of 1-bit CNNs with Improved Representational Capability and Advanced Training Algorithm”, ECCV, 2018. (CCF B)
  28. Baoyuan Wu, Fan Jia, Wei Liu and Bernard Ghanem, “Diverse Image Annotation”, CVPR, 2017. (CCF A)
  29. Baoyuan Wu, Siwei Lyu, Bernard Ghanem, “Constrained Submodular Minimization for Missing Labels and Class Imbalance in Multi-label Learning”, AAAI, 2016. (CCF A)
  30. Baoyuan Wu, Siwei Lyu, Bernard Ghanem, “ML-MG: Multi-label Learning with Missing Labels Using a Mixed Graph”, ICCV, 2015. (CCF A)
  31. Baoyuan Wu, Siwei Lyu, Baogang Hu, Qiang Ji, “Simultaneous Clustering and Tracklet Linking for Multi-Face Tracking in Videos”, ICCV, 2013. (CCF A)

Baoyuan Wu, Yifan Zhang, Baogang Hu and Qiang Ji, “Constrained Clustering and Its Application to Face Clustering in Videos”, CVPR, 2013. (CCF A)