Wayne Wu

I am a Research Associate in the Department of Computer Science at the University of California, Los Angeles, working with Bolei Zhou. Previously, I served as a Research Scientist at Shanghai AI Lab, where I led the Virtual Human Group. I was also a Visiting Scholar at Nanyang Technological University, working with Chen Change Loy. In June 2022, I obtained my PhD in the Department of Computer Science and Technology at Tsinghua University.
Research
My research lies at the intersection of computer vision, computer graphics, and robotics. I aim to develop human-centric embodied AI systems capable of perceiving, understanding, and interacting with real-world environments populated by humans. To advance this vision, I address key challenges in the scalability of robot learning environments, the situational awareness of agents, and the realism of populated virtual humans. My work explores three primary directions:
- Scalable Embodied AI Simulators: Developing large-scale robot learning platforms with diverse assets and infinite urban scenes, and enabling high-efficiency robot training, as in MetaUrban, URBAN-SIM, Vid2Sim, and OmniObject3D.
- Situational Behavior Modeling: Building autonomous decision-making models of humans and other agents, to behave robustly based on multimodal understanding of surroundings -- including vision, audio, and language, as in EmbodiedHuman, PedGen, and Seeing-to-Experiencing.
- Realistic Virtual Humans: Constructing high-fidelity 4D volumetric capture systems and datasets, as in DNA-Rendering and RenderMe-360; and developing human foundation models to obtain generalizable representations, as in CosmicMan and MotionBert.
News
May, 2025 | We are organizing the workshop on Real-to-Sim: Bridging the Gap between Neural Rendering and Robot Learning at CVPR 2025. 🔥 |
May, 2025 | We are organizing the workshop on Embodied “Humans”: Symbiotic Intelligence Between Virtual Humans and Humanoid Robots at CVPR 2025. 🔥 |
Jan, 2025 | We released MetaUrban – a simulation platform for Embodied AI in urban spaces. Try it now! |
Mar, 2024 | We are organizing the workshop on Virtual Humans for Robotics and Autonomous Driving at CVPR 2024. |
Jun, 2023 | OmniObject3D is selected as Best Paper Candidate at CVPR 2023. |
Oct, 2022 | I am leading OpenXDLab – a new large-scale open-source data platform! |
Aug, 2020 | We are organizing Workshop on Sensing, Understanding and Synthesizing Humans, ECCV 2020. |
Jul, 2020 | We released MMAction2 – OpenMMLab’s Next Generation Action Understanding Toolbox. |
Jul, 2020 | We released MMEditing – OpenMMLab’s Image and Video Editing Toolbox. |
Oct, 2019 | We are organizing Workshop on Statistical Deep Learning for Computer Vision, ICCV 2019. |
Selected Publications
- Towards Autonomous Micromobility through Scalable Urban SimulationComputer Vision and Pattern Recognition (CVPR), 2025Highlight
- Vid2Sim: Realistic and Interactive Simulation from Video for Urban NavigationComputer Vision and Pattern Recognition (CVPR), 2025
- MetaUrban: An Embodied AI Simulation Platform for Urban MicromobilityInternational Conference on Learning Representations (ICLR), 2025Spotlight
- CosmicMan: A Text-to-Image Foundation Model for HumansComputer Vision and Pattern Recognition (CVPR), 2024Highlight
- RenderMe-360: A Large Digital Asset Library and Benchmarks Towards High-fidelity Head AvatarsNeural Information Processing Systems (NeurIPS), Datasets and Benchmarks, 2023
- DNA-Rendering: A Diverse Neural Actor Repository for High-Fidelity Human-centric RenderingInternational Conference on Computer Vision (ICCV), 2023
- SynBody: Synthetic Dataset with Layered Human Models for 3D Human Perception and ModelingInternational Conference on Computer Vision (ICCV), 2023
- OmniObject3D: Large-Vocabulary 3D Object Dataset for Realistic Perception, Reconstruction and GenerationConference on Computer Vision and Pattern Recognition (CVPR), 2023Best Paper Candidate
- StyleGAN-Human: A Data-Centric Odyssey of Human GenerationEuropean Conference on Computer Vision (ECCV), 2022
- Look at Boundary: A Boundary-Aware Face Alignment AlgorithmConference on Computer Vision and Pattern Recognition (CVPR), 2018