Hao Tang

Carnegie Mellon University
Office: 5000 Forbes Av, Pittsburgh, PA 15213, USA🇺🇸
Email: bjdxtanghao@gmail.com

Hey, thanks for stopping by! 👋

I am currently a postdoctoral fellow at CMU, USA🇺🇸. Before this, I held a postdoctoral position at ETH Zürich, Switzerland🇨🇭. My academic journey includes earning a master's degree from Peking University, China🇨🇳, and completing my Ph.D. at UoT, Italy🇮🇹. Additionally, I had the privilege of being a visiting scholar at University of Oxford, UK🇬🇧. Furthermore, I undertook a visiting internship at IIAI, UAE🇦🇪.

My research interests are AIGC, AI4Science, machine learning, and computer vision. Specifically, I focus on:

  • Generative AI (GANs, diffusion models) and its applications (e.g., image generation, image translation, text-to-image synthesis/editing, person image synthesis, semantic image synthesis, style transfer, video generation, graph-based generation, layout generation)
  • Efficient AI (e.g., pruning, distillation, quantization, NAS)
  • Interpretability AI
  • Robustness AI
  • Large language model (LLM)
  • AI for Science
  • Multi-modalities (e.g., audio-to-video synthesis, language-vision, point cloud)
  • Low-level vision (image/video restoration, super-resolution, denoising, deblurring, HDR deghosting)
  • High-level vision (depth estimation, segmentation, detection, recognition)
  • 3D vision (e.g., nerf, 3D-aware image/video generation, object reconstruction/generation, 3D pose transfer)
  • Medical image enhancement and analysis
  • Human pose estimation and motion prediction

Position Openings

If this resonates with you, we are actively hiring.

For prospective collaborators, we have multiple positions for Postdoc/Ph.D./Master/Intern researchers. If you are interested in joining/visitng our lab or remotely working with us, please email me with your self-introduction, the project of interest (what is the problem you are trying to solve? and how are you trying to solve this problem (be as specific as possible)?), transcript, and CV to bjdxtanghao@gmail.com.

News

2024-04 🎉🎉🎉I received offers from MIT and Harvard University.
2024-02 We have 7 papers (Explanation for ViT + Faithfulness of ViT + Diffusion Policy for Versatile Navigation + Subject-Driven Generation [Final rating: 455] + Diffusion Model for 3D Hand Pose Estimation + Adversarial Learning for 3D Pose Transfer + Efficient Diffusion Distillation [224->235]) accepted to CVPR 2024.
2024-01 We have 1 paper (Architectural Layout Generation) accepted to TPAMI 2024.
2023-12 We have 1 paper (Sign Pose Sequence Generation) accepted to AAAI 2024.
2023-10 We have 4 papers (BEV Perception + Efficient ViT + 3D Motion Transfer + Graph Distillation) accepted to NeurIPS 2023.
2023-09 We have 1 paper (Practical Blind Image Denoising) accepted to MIR 2023.
2023-07 We have 1 paper (Semantic Image Synthesis) accepted to TPAMI 2023.
2023-06 We have 1 paper (Visible-Infrared Person Re-ID) accepted to ICCV 2023.
2023-05 We have 2 papers (Image Restoration Dataset + 3D-Aware Video Generation) accepted to CVPRW 2023 and 1 paper (3D Face Generation) accepted to JSTSP 2023.
2023-04 We have 1 paper (Speed-Aware Object Detection) accepted to ICML 2023, 2 papers (Lottery Ticket Hypothesis for ViT + Zero-shot Character Recognition) accepted to IJCAI 2023, 1 paper (3D Human Pose Estimation) accepted to PR 2023, and 1 paper (SAR Target Recognition) accepted to TGRS 2023.
2023-03 We have 6 papers (HDR Deghosting + Point Cloud Registration + Graph-Constrained House Generation + Mathematical Architecture Design + Text-to-Image Synthesis + Efficient Semantic Segmentation) accepted to CVPR 2023.
2023-02 We have 3 papers (Camouflaged Object Detection + Brain Vessel Image Segmentation + Cross-View Image Translation) accepted to ICASSP 2023 and 1 paper (Camouflaged Object Detection) accepted to TCSVT 2023.
2023-01 We have 1 paper (Semantic Image Synthesis) accepted to ICLR 2023 and 1 paper (Human Reaction Generation) accepted to TMM 2023.
2022-11 We have 4 papers (Real-Time Segmentation + Wearable Design + Efficient ViT Training + Text-Guided Image Editing) accepted to AAAI 2023, 1 paper accepted (Person Pose and Facial Image Synthesis) to IJCV 2022, 1 paper (Salient Object Detection) accepted to TIP 2022, and 1 paper (Object Detection Transformer) accepted to TCSVT 2022.
2022-10 We have 1 paper (Sinusoidal Neural Radiance Fields) accepted to BMVC 2022 and 1 paper (Guided Image-to-Image Translation) accepted to TPAMI 2022.
2022-09 We have 1 paper (Facial Expression Translation) accepted to TAFFC 2022 and 1 paper (Ship Detection) accepted to TGRS 2022.
2022-07 We have 5 papers (Real-Time SR + Video SR + Soft Token Pruning for ViT + 3D-Aware Human Synthesis + Video Semantic Segmentation) accepted to ECCV 2022, 1 paper (Gaze Correction and Animation) accepted to TIP 2022, and 1 paper (Cross-view Panorama Image Synthesis) accepted to PR 2022.
2022-06 We have 2 papers (Character Image Restoration + Character Image Denoising) accepted to ACM MM 2022.
2022-04 We have 1 paper (Real-Time Portrait Stylization) accepted to IJCAI 2022, 1 paper (Wide-Context Transformer for Semantic Segmentation) accepted to TGRS 2022, and 1 paper (Incremental Learning for Semantic Segmentation) accepted to TMM 2022.
2022-03 We have 5 papers (Text-to-Image Synthesis + 3D Human Pose Estimation + Text-Driven Image Manipulation + 3D Face Modeling + 3D Face Restoration) accepted to CVPR 2022, 1 paper (Image Generation) accepted to TPAMI 2022, and 1 paper (Cross-View Panorama Image Synthesis) accepted to TMM 2022.
2021-12 We have 2 papers (Generalized 3D Pose Transfer + Audio-Visual Speaker Tracking) accepted to AAAI 2022.
2021-11 We have 1 paper (Building Extraction in VHR Remote Sensing Images) accepted to TIP 2021.
2021-10 We have 3 papers (Cross-View Image Translation + Data-driven 3D Animation + Natural Image Matting) accepted to BMVC 2021.
2021-08 We have 1 paper (Layout-to-Image Translation) accepted to TIP 2021 and 1 paper (Unpaired Image-to-Image Translation) accepted to TNNLS 2021.
2021-07 We have 2 papers (Continuous Pixel-Wise Prediction + Unsupervised 3D Pose Transfer) accepted to ICCV 2021.
2021-06 We have 1 paper (Cross-View Exocentric to Egocentric Video Synthesis) accepted to ACM MM 2021 and 1 paper (Total Generate) accepted to TMM 2021.
2020-08 We have 1 paper (Person Image Generation) accepted to BMVC 2020, 2 papers (Semantic Image Synthesis + Unsupervised Gaze Correction and Animation) accepted to ACM MM 2020, and 1 paper (Controllable Image-to-Image Translation) accepted to TIP 2020.
2020-07 We have 1 paper (Person Image Generation) accepted to ECCV 2020.
2020-05 We have 1 paper (Deep Dictionary Learning and Coding) accepted to TNNLS 2020 and 1 paper (Semantic Segmentation of Remote Sensing Images) accepted to TGRS 2020.
2020-02 We have 1 paper (Semantic-Guided Scene Generation) accepted to CVPR 2020.
2019-07 We have 1 paper (Keypoint-Guided Image Generation) accepted to ACM MM 2019.
2019-02 We have 1 paper (Cross-View Image Translation) accepted to CVPR 2019.
2018-06 We have 1 paper (Hand Gesture-to-Gesture Translation) accepted to ACM MM 2018.
2018-02 We have 1 paper (Monocular Depth Estimation) accepted to CVPR 2018.
2016-07 We have 1 paper (Large Scale Image Retrieval) accepted to IJCAI 2016.
2015-08 We have 1 paper (Gender Classification) accepted to ACM MM 2015.

Featured Publications

(Including NeurIPS, ICML, ICLR, CVPR, ICCV, ECCV, AAAI, IJCAI, ACM MM)

†Equal Contribution, *Corresponding Author(s)

  1. Arxiv
    Motion Mamba: Efficient and Long Sequence Motion Generation with Hierarchical and Bidirectional Selective SSM
     Zeyu Zhang,  Akide Liu,  Ian Reid,  Richard Hartley,  Bohan Zhuang,  Hao Tang*
    In Arxiv, 2024
    PDF Code
  2. Arxiv
    A Survey on Multimodal Wearable Sensor-based Human Action Recognition
     Jianyuan Ni,  Hao Tang,  Syed Tousiful Haque,  Yan Yan,  Anne HH Ngu
    In Arxiv, 2024
    PDF Code
  3. Arxiv
    Physical Adversarial Attack Neets Computer Vision: A Decade Survey
     Hui Wei,  Hao Tang,  Xuemei Jia,  Zhixiang Wang,  Hanxun Yu,  Zhubo Li,  Shin'ichi Satoh,  Luc Van Gool,  Zheng Wang
    In Arxiv, 2024
    PDF Code
  4. Arxiv
    StoryImager: A Unified and Efficient Framework for Coherent Story Visualization and Completion
     Ming Tao,  Bingkun Bao,  Hao Tang,  Yaowei Wang,  Changsheng Xu
    In Arxiv, 2024
    PDF Code
  5. Arxiv
    InstructGIE: Towards Generalizable Image Editing
     Zichong Meng,  Changdi Yang,  Jun Liu,  Hao Tang*,  Pu Zhao*,  Yanzhi Wang*
    In Arxiv, 2024
    PDF Code
  6. Arxiv
    Enlighten-Your-Voice: When Multimodal Meets Zero-shot Low-light Image Enhancement
     Xiaofeng Zhang,  Zishan Xu,  Hao Tang,  Chaochen Gu,  Wei Chen,  Shanying Zhu,  Xinping Guan
    In Arxiv, 2024
    PDF Code
  7. Arxiv
    MaskSAM: Towards Auto-prompt SAM with Mask Classification for Medical Image Segmentation
     Bin Xie,  Hao Tang,  Bin Duan,  Dawen Cai,  Yan Yan
    In Arxiv, 2024
    PDF Code
  8. Arxiv
    Efficient Pruning of Large Language Model with Adaptive Estimation Fusion
     Jun Liu,  Chao Wu,  Changdi Yang,  Hao Tang*,  Haoye Dong,  Zhenglun Kong,  Geng Yuan,  Wei Niu,  Dong Huang*,  Yanzhi Wang*
    In Arxiv, 2024
    PDF Code
  9. Arxiv
    StableGarment: Garment-Centric Generation via Stable Diffusion
     Rui Wang,  Hailong Guo,  Jiaming Liu,  Huaxia Li,  Haibo Zhao,  Xu Tang,  Yao Hu,  Hao Tang,  Peipei Li
    In Arxiv, 2024
    PDF Code
  10. Arxiv
    SCP-Diff: Photo-Realistic Semantic Image Synthesis with Spatial-Categorical Joint Prior
     Huan-ang Gao,  Mingju Gao,  Jiaju Li,  Wenyi Li,  Rong Zhi,  Hao Tang,  Hao Zhao
    In Arxiv, 2024
    PDF Code
  11. CVPR
    Highlight
    HandDiff: 3D Hand Pose Estimation with Diffusion on Image-Point Cloud
     Wencan Cheng,  Hao Tang,  Luc Van Gool,  Jong Hwan Ko
    In CVPR 2024, Seattle, USA
    PDF Code
  12. CVPR
    Versatile Navigation under Partial Observability via Value-guided Diffusion Policy
     Gengyu Zhang,  Hao Tang,  Yan Yan
    In CVPR 2024, Seattle, USA
    PDF Code
  13. CVPR
    Towards Robust 3D Pose Transfer with Adversarial Learning
     Haoyu Chen,  Hao Tang,  Ehsan Adeli,  Guoying Zhao
    In CVPR 2024, Seattle, USA
    PDF Code
  14. CVPR
    On the Faithfulness of Vision Transformer Explanations
     Junyi Wu,  Weitai Kang,  Hao Tang,  Yuan Hong,  Yan Yan
    In CVPR 2024, Seattle, USA
    PDF Code
  15. CVPR
    Token Transformation Matters: Towards Faithful Post-hoc Explanation for Vision Transformer
     Junyi Wu,  Bin Duan,  Weitai Kang,  Hao Tang,  Yan Yan
    In CVPR 2024, Seattle, USA
    PDF Code
  16. CVPR
    SSR-Encoder: Encoding Selective Subject Representation for Subject-Driven Generation
     Yuxuan Zhang,  Jiaming Liu,  Yiren Song,  Rui Wang,  Hao Tang,  Jinpeng Yu,  Huaxia Li,  Xu Tang,  Yao Hu,  Han Pan,  Zhongliang Jing
    In CVPR 2024, Seattle, USA
    PDF Code
  17. CVPR
    Workshop
    Towards Online Real-Time Memory-based Video Inpainting Transformers
     Guillaume Thiry,  Hao Tang*,  Radu Timofte,  Luc Van Gool
    In CVPR 2024, Seattle, USA
    PDF Code
  18. AAAI
    G2P-DDM: Generating Sign Pose Sequence from Gloss Sequence with Discrete Diffusion Model
     Pan Xie,  Qipeng Zhang,  Peng Taiying,  Hao Tang*,  Yao Du,  Zexian Li
    In AAAI 2024, Vancouver, Canada
    PDF Code
  19. NeurIPS
    HotBEV: Hardware-oriented Transformer-based Multi-View 3D Detector for BEV Perception
     Peiyan Dong,  Zhenglun Kong,  Xin Meng,  Pinrui Yu,  Yifan Gong,  Geng Yuan,  Hao Tang*, Yanzhi Wang
    In NeurIPS 2023, New Orleans, USA
    PDF Code
  20. NeurIPS
    PackQViT: Faster Sub-8-bit Vision Transformers via Full and Packed Quantization on the Mobile
     Peiyan Dong,  Lei Lu,  Chao Wu,  Cheng Lyu,  Geng Yuan,  Hao Tang*, Yanzhi Wang
    In NeurIPS 2023, New Orleans, USA
    PDF Code
  21. NeurIPS
    LART: Neural Correspondence Learning with Latent Regularization Transformer for 3D Motion Transfer
     Haoyu Chen,  Hao Tang,  Radu Timofte,  Luc Van Gool,  Guoying Zhao
    In NeurIPS 2023, New Orleans, USA
    PDF Code
  22. NeurIPS
    Does Graph Distillation See Like Vision Dataset Counterpart?
     Beining Yang,  Kai Wang,  Qingyun Sun,  Cheng Ji,  Xingcheng Fu,  Hao Tang,  Yang You,  Jianxin Li
    In NeurIPS 2023, New Orleans, USA
    PDF Code
  23. MIR
    Practical Blind Image Denoising via Swin-Conv-UNet and Data Synthesis
     Kai Zhang,  Yawei Li,  Jingyun Liang,  Jiezhang Cao,  Yulun Zhang,  Hao Tang,  Dengping Fan,  Radu Timofte,  Luc Van Gool
    Springer Machine Intelligence Research (MIR), 2023
    PDF Code
  24. ICCV
    Learning Concordant Attention via Target-aware Alignment for Visible-Infrared Person Re-identification
     Jianbing Wu,  Hong Liu,  Yuxin Su,  Wei Shi,  Hao Tang
    In ICCV 2023, Paris, France
    PDF Code
  25. ICML
    SpeedDETR: Speed-aware Transformers for End-to-end Object Detection
     Peiyan Dong,  Zhenglun Kong,  Xin Meng,  Peng Zhang,  Hao Tang*,  Yanzhi Wang,  Chih-Hsien Chou
    In ICML 2023, Hawaii, USA
    PDF Code
  26. IJCAI
    Data Level Lottery Ticket Hypothesis for Vision Transformers
     Xuan Shen,  Zhenglun Kong,  Minghai Qin,  Peiyan Dong,  Geng Yuan,  Xin Meng,  Hao Tang,  Xiaolong Ma,  Yanzhi Wang
    In IJCAI 2023, Macao, China
    PDF Code
  27. CVPR
    Graph Transformer GANs for Graph-Constrained House Generation
     Hao Tang,  Zhenyu Zhang,  Humphrey Shi,  Bo Li,  Ling Shao,  Nicu Sebe,  Radu Timofte,  Luc Van Gool
    In CVPR 2023, Vancouver, Canada
    PDF Code
  28. CVPR
    Unsupervised Deep Probabilistic Approach for Partial Point Cloud Registration
     Guofeng Mei,  Hao Tang,  Xiaoshui Huang,  Weijie Wang,  Juan Liu,  Jian Zhang,  Luc Van Gool,  Qiang Wu
    In CVPR 2023, Vancouver, Canada
    PDF Code
  29. CVPR
    DeepMAD: Mathematical Architecture Design for Deep Convolutional Neural Network
     Xuan Shen,  Yaohua Wang,  Ming Lin,  Yilun Huang,  Hao Tang,  Xiuyu Sun,  Yanzhi Wang
    In CVPR 2023, Vancouver, Canada
    PDF Code
  30. CVPR
    GALIP: Generative Adversarial CLIPs for Text-to-Image Synthesis
     Ming Tao,  Bingkun Bao,  Hao Tang,  Changsheng Xu
    In CVPR 2023, Vancouver, Canada
    PDF Code
  31. ICLR
    Edge Guided GANs with Contrastive Learning for Semantic Image Synthesis
     Hao Tang,  Xiaojuan Qi,  Guolei Sun,  Dan Xu,  Nicu Sebe,  Radu Timofte,  Luc Van Gool
    In ICLR 2023, Kigali, Rwanda
    PDF Code
  32. ECCV
    3D-Aware Semantic-Guided Generative Model for Human Synthesis
     Jichao Zhang,  Enver Sangineto,  Hao Tang,  Aliaksandr Siarohin,  Zhun Zhong,  Nicu Sebe,  Wei Wang
    In ECCV 2022, Tel Aviv, Israel
    PDF Code
  33. ECCV
    Towards Interpretable Video Super-Resolution via Alternative Optimization
     Jiezhang Cao,  Jingyun Liang,  Kai Zhang,  Wenguan Wang,  Qin Wang,  Yulun Zhang,  Hao Tang,  Luc Van Gool
    In ECCV 2022, Tel Aviv, Israel
    PDF Code
  34. CVPR
    MHFormer: Multi-Hypothesis Transformer for 3D Human Pose Estimation
     Wenhao Li,  Hong Liu,  Hao Tang,  Pichao Wang,  Luc Van Gool
    In CVPR 2022, New Orleans, USA
    PDF Code
  35. CVPR Oral
    DF-GAN: A Simple and Effective Baseline for Text-to-Image Synthesis
     Ming Tao,  Hao Tang,  Fei Wu,  Xiaoyuan Jing,  Bingkun Bao,  Changsheng Xu
    In CVPR 2022, New Orleans, USA
    PDF Code
  36. ACM MM Best
    Paper Candidate
    GestureGAN for Hand Gesture-to-Gesture Translation in the Wild
     Hao Tang,  Wei Wang,  Dan Xu,  Yan Yan,  Nicu Sebe
    In ACM MM 2018, Seoul, South Korea
    PDF Code