Zihan Wang

Department of Electrical Engineering and Computer Sciences, UC Berkeley

I am a first-year CS Ph.D. student in Berkeley Artificial Intelligence Research (BAIR) at UC Berkeley EECS, where I am fortunate to be advised by Prof. Angjoo Kanazawa, Prof. Jitendra Malik, and Prof. Pieter Abbeel. My research focuses on 4D vision for robot learning, particularly enabling robots to learn robust and generalizable skills from human videos. I was an Applied Scientist Intern at Frontier AI & Robotics (FAR), Amazon, working on humanoid perceptive locomotion & loco-manipulation.

I earned my M.S. in Computer Vision from Robotics Institute, School of Computer Science at Carnegie Mellon University, where I was fortunate to be advised by Prof. Deva Ramanan and to collaborate with Prof. Shubham Tulsiani. Before CMU, I received my B.Eng. (Hons) in Electrical & Electronics Engineering from The University of Edinburgh. During my undergraduate years, I also had the opportunity to work with Prof. Sebastian Scherer and Prof. Chen Wang at AirLab, CMU Robotics Institute.

Email  /  Google Scholar  /  GitHub  /  Twitter  /  LinkedIn

We live in a dynamic 4D world, constantly seeing, understanding, and interacting with objects and environments. My goal is to bridge the gap between reconstructed virtual worlds and practical robotic applications. I develop methods that help robots perceive, understand, and interact with objects and environments from sparse observations.

@inproceedings{wang2026contact,
  title={Contact-guided real2sim from monocular video with planar scene primitives},
  author={Wang, Zihan and Wang, Jiashun and Tan, Jeff and Zhao, Yiwen and Hodgins, Jessica and Tulsiani, Shubham and Ramanan, Deva},
  booktitle={International Conference on Learning Representations},
  volume={2026},
  pages={115452--115468},
  year={2026}
}

Reconstructs simulation-ready human motion and static scene geometry from monocular RGB video using planar primitives and human-scene contacts.

@inproceedings{wang2025monofusion,
  title={Monofusion: Sparse-view 4d reconstruction via monocular fusion},
  author={Wang, Zihan and Tan, Jeff and Khurana, Tarasha and Peri, Neehar and Ramanan, Deva},
  booktitle={2025 IEEE/CVF International Conference on Computer Vision (ICCV)},
  pages={8252--8263},
  year={2025},
  organization={IEEE}
}

Reconstructs dynamic 4D scenes from four sparse-view cameras by aligning monocular geometry and motion estimates for novel-view synthesis.

@inproceedings{wang2024airshot,
  title={AirShot: Efficient few-shot detection for autonomous exploration},
  author={Wang, Zihan and Li, Bowen and Wang, Chen and Scherer, Sebastian},
  booktitle={2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
  pages={11654--11661},
  year={2024},
  organization={IEEE}
}

Uses a Top Prediction Filter (TPF) on correlation maps for efficient few-shot object detection without novel-class fine-tuning at deployment.

A general framework that learns whole-body humanoid-object interaction skills directly from monocular RGB videos.

• EECS Department Fellowship, 2026
• Edinburgh Award, 2022
• Edinburgh Scholarship, 2022 & 2021
• Qianjiang Electric Scholarship, 2020

Academic Reviewer

• International Conference on Learning Representations (ICLR)
• European Conference on Computer Vision (ECCV)
• Conference on Robot Learning (CoRL)
• IEEE International Conference on Robotics and Automation (ICRA)
• IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

• Email: z1hanw@cs.berkeley.edu / zihanwa3@cs.cmu.edu
• Phone: +1-412-589-8046

I am confident that my "boundaryless", "website-stalking" friends have contributed to at least 74.47% of the traffic here.

Template adapted from Qitao Zhao (credit to Jon Barron).
Last updated: Aug 17, 2026