jay.png

I am a Ph.D. student at Show Lab, National University of Singapore, advised by Prof. Mike Zheng Shou and Prof. Wynne Hsu. My research interests lie in generative models for images, videos, 3D and 4D.

email: jay.zhangjie.wu [at] gmail.com

google scholar · github · linkedin · twitter

news

Jun 2025 Difix3D+ is recognized as the Best Paper Award Candidate at CVPR 2025.
Feb 2025 We’re organizing WorldModelBench: The First Workshop on Benchmarking World Foundation Models at CVPR 2025.
Feb 2024 Tutorial Diffusion-based Video Generative Models to appear at CVPR 2024.
Oct 2023 Code and model weights of Show-1 are released!
May 2023 Organized LOVEU-TGVE (Text-Guided Video Editing) competition at CVPR 2023.
Apr 2023 Searching for papers on video diffusion models? Check out our GitHub repo Awesome-Video-Diffusion .

publications

(*) denotes equal contribution
  1. chronoedit.gif
    ChronoEdit: Towards Temporal Reasoning for Image Editing and World Simulation
    Jay Zhangjie Wu*, Xuanchi Ren*, Tianchang Shen, Tianshi Cao, Kai He, Yifan Lu, Ruiyuan Gao, Enze Xie, Shiyi Lan, Jose M. Alvarez, Jun Gao, Sanja Fidler, Zian Wang, and Huan Ling*
    Technical Report · GitHub Repo stars
    Paper Project Page Code Demo
  2. sfvae.png
    Sparse Image Synthesis via Joint Latent and RoI Flow
    Ziteng Gao, Jay Zhangjie Wu, and Mike Zheng Shou
    NeurIPS 2025
    Paper
  3. difix3d.gif
    Difix3D+: Improving 3D Reconstructions with Single-Step Diffusion Models
    Jay Zhangjie Wu*, Yuxuan Zhang*, Haithem Turki, Xuanchi Ren, Jun Gao, Mike Zheng Shou, Sanja Fidler, Zan Gojcic, and Huan Ling
    CVPR 2025 (Oral) · Best Paper Award Candidate · GitHub Repo stars
    Paper Project Page Code Demo
  4. cosmosdrivedreams.gif
    Cosmos-Drive-Dreams: Scalable Synthetic Driving Data Generation with World Foundation Models
    Xuanchi Ren*, Yifan Lu*, Tianshi Cao*, Ruiyuan Gao*, Shengyu Huang, Amirmojtaba Sabour, Tianchang Shen, Tobias Pfaff, Jay Zhangjie Wu, Runjian Chen, Seung Wook Kim, Jun Gao, Laura Leal-Taixe, Mike Chen, Sanja Fidler, and Huan Ling
    White Paper · GitHub Repo stars
    Paper Project Page Code
  5. cosmostransfer1.gif
    Cosmos-Transfer1: Conditional World Generation with Adaptive Multimodal Control
    NVIDIA
    White Paper · GitHub Repo stars
    Paper Project Page Code
  6. cosmos.gif
    Cosmos World Foundation Model Platform for Physical AI
    NVIDIA (Jay Zhangjie Wu: Core contributor)
    White Paper · Best AI + Best Overall of CES 2025 · GitHub Repo stars
    Paper Website Project Page Code Video Blog Demo
  7. infinicube.gif
    InfiniCube: Unbounded and Controllable Dynamic 3D Driving Scene Generation with World-Guided Video
    Yifan Lu*, Xuanchi Ren*, Jiawei Yang, Tianchang Shen, Zhangjie Wu, Jun Gao, Yue Wang, Siheng Chen, Mike Chen, Sanja Fidler, and Jiahui Huang
    ICCV 2025 · GitHub Repo stars
    Paper Project Page Code
  8. scube.gif
    SCube: Instant Large-Scale Scene Reconstruction using VoxSplats
    Xuanchi Ren*, Yifan Lu*, Hanxue Liang, Zhangjie Wu, Huan Ling, Mike Chen, Sanja Fidler, Francis Williams, and Jiahui Huang
    NeurIPS 2024 · GitHub Repo stars
    Paper Project Page Code
  9. evolvedirector.png
    EvolveDirector: Approaching Advanced Text-to-Image Generation with Large Vision-Language Models
    Rui Zhao, Hangjie Yuan, Yujie Wei, Shiwei Zhang, Yuchao Gu, Lingmin Ran, Xiang Wang, Zhangjie Wu, Junhao Zhang, Yingya Zhang, and Mike Zheng Shou
    NeurIPS 2024
    Paper Code
  10. motiondirector.gif
    MotionDirector: Motion Customization of Text-to-Video Diffusion Models
    Rui Zhao, Yuchao Gu, Jay Zhangjie Wu, David Junhao Zhang, Jiawei Liu, Weijia Wu, Jussi Keppo, and Mike Zheng Shou
    ECCV 2024 (Oral) · GitHub Repo stars
    Paper Project Page Code
  11. freeatm.png
    Free-ATM: Exploring Unsupervised Learning on Diffusion-Generated Images with Free Attention Masks
    David Junhao Zhang, Mutian Xu, Jay Zhangjie Wu, Wenqing Zhang, Xiaoguang Han, Song Bai, and Mike Zheng Shou
    ECCV 2024
    Paper
  12. videoswap.gif
    VideoSwap: Customized Video Subject Swapping with Interactive Semantic Point Correspondence
    Yuchao Gu, Yipin Zhou, Bichen Wu, Licheng Yu, Jia-Wei Liu, Rui Zhao, Jay Zhangjie Wu, David Junhao Zhang, Mike Zheng Shou, and Kevin Tang
    CVPR 2024 · GitHub Repo stars
    Paper Project Page Code
  13. dynvideoe.gif
    DynVideo-E: Harnessing Dynamic NeRF for Large-Scale Motion- and View-Change Human-Centric Video Editing
    Jia-Wei Liu, Yan-Pei Cao, Jay Zhangjie Wu, Weijia Mao, Yuchao Gu, Rui Zhao, Jussi Keppo, Ying Shan, and Mike Zheng Shou
    CVPR 2024
    Paper Project Page
  14. show1.gif
    Show-1: Marrying Pixel and Latent Diffusion Models for Text-to-Video Generation
    David Junhao Zhang*, Jay Zhangjie Wu*, Jia-Wei Liu*, Rui Zhao, Lingmin Ran, Yuchao Gu, Difei Gao, and Mike Zheng Shou
    IJCV 2024 · GitHub Repo stars
    Paper Project Page Code
  15. mixofshow.png
    Mix-of-Show: Decentralized Low-Rank Adaptation for Multi-Concept Customization of Diffusion Models
    Yuchao Gu, Xintao Wang, Jay Zhangjie Wu, Yujun Shi, Yunpeng Chen, Zihan Fan, Wuyou Xiao, Rui Zhao, Shuning Chang, Weijia Wu, Yixiao Ge, Ying Shan, and Mike Zheng Shou
    NeurIPS 2024 · GitHub Repo stars
    Paper Project Page Code
  16. tuneavideo.gif
    Tune-A-Video: One-Shot Tuning of Image Diffusion Models for Text-to-Video Generation
    Jay Zhangjie Wu, Yixiao Ge, Xintao Wang, Stan Weixian Lei, Yuchao Gu, Yufei Shi, Wynne Hsu, Ying Shan, Xiaohu Qie, and Mike Zheng Shou
    ICCV 2023 · Most Influential ICCV’23 Papers #5 · GitHub Repo stars
    Paper Project Page Code Demo Two Minute Papers
  17. leocod.png
    Label-Efficient Online Continual Object Detection in Streaming Video
    Jay Zhangjie Wu, David Junhao Zhang, Wynne Hsu, Mengmi Zhang, and Mike Zheng Shou
    ICCV 2023
    Paper Code
  18. clvqa.png
    Symbolic Replay: Scene Graph as Prompt for Continual Learning on VQA Task
    Stan Weixian Lei*, Difei Gao*, Jay Zhangjie Wu, Yuxuan Wang, Wei Liu, Mengmi Zhang, and Mike Zheng Shou
    AAAI 2023 (Oral)
    Paper Code
  19. egfr.png
    Mining whole-lung information by artificial intelligence for predicting EGFR genotype and targeted therapy response in lung cancer: a multicohort study
    Shuo Wang*, He Yu*, Yuncui Gan*, Zhangjie Wu, and et al.
    The Lancet Digital Health 2022 (Impact Factor: 23.8)
    Paper PDF
  20. opixray.png
    Occluded Prohibited Items Detection: An X-ray Security Inspection Benchmark and De-occlusion Attention Module
    Yanlu Wei*, Renshuai Tao*, Zhangjie Wu, Yuqing Ma, Libo Zhang, and Xianglong Liu
    ACM MM 2020 (Oral)
    Paper Code