Yang You (尤洋)

Yang You (尤洋)

Postdoc at the Geometric Computation group led by Prof. Leonidas Guibas in Stanford. Research interests include 3D graphics, 3D computer vision and robotics.

Visitors

About Me

I am a postdoctoral researcher in the Geometric Computation group led by Prof. Leonidas Guibas at Stanford. My research focuses on 3D graphics, 3D computer vision, and robotics, with an emphasis on robust perception and representation for real-world spatial understanding.

Research focus: 3D understanding, geometric learning, object-centric perception, and embodied AI.

I am interested in new and cool things, and in figuring out the principles beneath a particular technique. Feel free to contact me at yangyou at stanford.edu if you have any questions about my research or want to have a broad discussion in the field of 3D understanding. Including your biography when seeking a discussion would be highly appreciated.

A mug from KeypointNet with its annotated keypoints and its 9D pose: the box gives position and size, the axes give rotation. Drag to turn it.

News

  1. Our paper UniPose9D: Universal Category-Agnostic Object Pose Estimation is now available on arXiv. [Paper] [Code] [Project Page] [Demo]
  2. Our paper: OCH3R: Object-Centric Holistic 3D Reconstruction is accepted to the CVPR 2026 OpenSUN3D Workshop! [Paper]
  3. Our paper: DOT-Sim: Differentiable Optical Tactile Simulation with Precise Real-to-Sim Physical Calibration is accepted to ICRA 2026! [Paper] [Project Page]
  4. Our paper: Rodrigues Network for Learning Robot Actions is accepted to ICLR 2026 as an Oral! [Paper]
  5. Our paper: Img2CAD: Reverse Engineering 3D CAD Models from Images through VLM-Assisted Conditional Factorization is accepted to SIGGRAPH Asia 2025! [Paper] [Code] [Project Page]
  6. We’re hosting the cross-disciplinary workshop ``WCLOP: 1st ICCV Workshop and Challenge on Category-Level Object Pose Estimation in the Wild’’ at ICCV 2025, spotlighting advances in category-level object pose estimation for robotic manipulation scenarios. We are now accepting paper submissions and challenge participants! Awards totaling over $6000 are available! Please see our website for more details.
  7. Two papers (Robot Learning from Any Images, ARCH: Hierarchical Hybrid Learning for Long-Horizon Contact-Rich Robotic Assembly) got accepted to CoRL 2025!
  8. Our paper: AllTracker: Efficient Dense Point Tracking at High Resolution is accepted to ICCV 2025! [Paper] [Code] [Project Page]

All news