Skip to content
View wzzheng's full-sized avatar

Highlights

  • Pro

Block or report wzzheng

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please donโ€™t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this userโ€™s behavior. Learn more about reporting abuse.

Report abuse
wzzheng/README.md

Hi there ๐Ÿ‘‹

I'm Wenzhao Zheng, a postdoctoral fellow at BAIR, UC Berkeley, working with Prof. Kurt Keutzer. I received my Ph.D. in Automation from Tsinghua University, advised by Jie Zhou and Jiwen Lu, and my B.S. in Physics also from Tsinghua.

Previous Efforts

We worked on 3D perception, world models, and large language models with a focus on their applications in autonomous driving.

Some projects you may know:

  • 3D Occupancy Prediction: TPVFormer GitHub stars-> SurroundOcc GitHub stars -> PointOcc GitHub stars -> SelfOcc GitHub stars -> EmbodiedOcc GitHub stars
  • Gaussian Scene Representation-> GaussianFormer GitHub stars -> SยณGaussian GitHub stars -> GaussianWorld GitHub stars -> GaussianAD GitHub stars
  • 3D/4D Geometry Foundation Models: Point3R GitHub stars -> StreamVGGT GitHub stars -> DVGT GitHub stars -> IVGT GitHub stars -> SM4RT GitHub stars
  • World Models and Visual Generation: OccWorld GitHub stars -> OccSora GitHub stars -> SVG GitHub stars -> Astra GitHub stars -> MoRe4D GitHub stars
  • Autonomous Driving: SurroundDepth GitHub stars -> BEVerse GitHub stars-> GenAD GitHub stars -> Doe-1 GitHub stars -> LDM GitHub stars
  • Large Language Models: SparseVLM GitHub stars -> Skyra GitHub stars

Current Interests

Our target now is to build Physical Superintelligence (PSI) systems on more general robots.

๐Ÿฆ™ LLMs + ๐Ÿ’ป Digital Harness -> ๐Ÿง  AGI

๐Ÿฆ™ LLMs + ๐ŸŒ Physical Harness (๐Ÿ“ Spatial Intelligence, ๐ŸŒ World Modeling, and ๐Ÿฆพ Embodied Robots) -> ๐Ÿค– PSI

  • ๐Ÿ“ Spatial Intelligence: generative geometry foundation models with complete geometry, motion, and semantics
  • ๐ŸŒ World Modeling: world model agents with multimodal harness and embodied interaction control
  • ๐Ÿฆพ Embodied Robots: humanoids with whole-body intelligence and dexterous manipulation ability
  • ๐Ÿฆ™ LLMs: recurssive self-improvement for multi-agent system, GUI agents and spatial agents

Collaborations

I'm actively looking for research interns and collaborators interested in these directions. If you would like to work with me in person or remotely at UC Berkeley or Tsinghua University, feel free to email me at wenzhao.zheng@outlook.com.

Pinned Loading

  1. TPVFormer TPVFormer Public

    [CVPR 2023] An academic alternative to Tesla's occupancy network for autonomous driving.

    Python 1.4k 128

  2. StreamVGGT StreamVGGT Public

    [ICLR 2026] Streaming 4D Visual Geometry Transformer

    Python 994 55

  3. OccWorld OccWorld Public

    [ECCV 2024] 3D World Model for Autonomous Driving

    Python 597 45

  4. GenAD GenAD Public

    [ECCV 2024] GenAD: Generative End-to-End Autonomous Driving

    Python 499 55

  5. DVGT DVGT Public

    [CVPR 2026] Visual Geometry Transformer for Autonomous Driving

    Python 377 21

  6. LDM LDM Public

    Large Driving Models

    310 12