About Me
I am a Research Scientist at Beijing Institute for General Artificial Intelligence (BIGAI). I received my Ph.D. in the Department of Statistics at the University of California, Los Angeles (UCLA). I believe the ability to comprehensively perceive the world is central to next‐step general AI. Thus, my research focuses on enabling human‐like vision capabilities for machines, particularly in the areas of 3D scene understanding (e.g., reconstruction, grounding, human motion, and human‑object interaction).
Previously, I interned at Perceiving Systems, MPI for Intelligent Systems (advised by Prof. Michael Black and Prof. Dimitrios Tzionas), and Amazon (supervised by Li Erran Li).
News
New! 2025-04: We are organizing the 5th 3D Scene Understanding for Vision, Graphics and Robotics at CVPR 2025. Check out the workshop page for the awesome speaker lineup and exciting challenges!
New! 2025-03: Five papers accepted in CVPR 2025, including DPRecon, MOVIS, MetaScenes on 3D scene understanding, and MotionReFit, InteractAnything on human-scene interaction.
New! 2025-02: SynergAI, aiming to address human-robot perception alignment, accepted in ICRA 2025.
2024-10: PhyRecon accepted in NeurIPS 2024. Code for reconstruction and physical simulator all released! Congrats Junfeng et al.
2024-10: LINGO accepted in SIGGRAPH Asia 2024, which continuously pushes the boundaries of 3D human‑scene interaction. Congrats Nan et al.