Yihan Wang
Ph.D. Candidate
New Jersey, United States
Summary
🤩
Rockstar🎓
Top SchoolYihan Wang is a Princeton Ph.D. candidate in computer science specializing in low-level vision and multi-view perception, advised by Prof. Jia Deng. With six years of research experience and a B.Eng. from Tsinghua’s elite Yao Class, Yihan has published state-of-the-art methods in optical flow, multi-agent RL, and efficient on-device pose estimation. At Princeton they delivered an optical flow method that cut error by up to 23% while speeding inference 24× and topped a Spring leaderboard (ECCV 2024), and they have built procedural 3D asset pipelines used in photorealistic world generation. Past roles at MIT and Tsinghua produced practical, hardware-efficient systems and algorithmic improvements that reduced complexity from exponential to linear. Yihan blends rigorous theory with hands-on implementation—authoring Blender Python tools for procedural assets and contributing extensively to the popular infinigen repo. Based in New Jersey, they focus on turning challenging perception problems into fast, production-ready solutions.
6 years of coding experience
2 years of employment as a software developer
Bachelor of Engineering - BE, Computer Science, Bachelor of Engineering - BE, Computer Science at 清华大学Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at Princeton University
Github Skills (9)
procedural-generation10
material-design10
geometry10
nodes10
python10
geom10
computer-graphics9
api8
apidoc8
Programming languages (2)
Github contributions (5)
princeton-vl/infinigen
Jun 2023 - Jul 2023
Infinite Photorealistic Worlds using Procedural Generation
Role in this project:
Back-end Developer
Back-end DeveloperContributions:1 branch in 18 days
Contributions summary:Yihan primarily contributed to the development of various asset generation scripts within the "infinigen" repository. They added and modified code related to procedural generation of diverse assets like horns, moss, ivy, bones, lichen, shells, particles, jellyfish, urchins, snails, trees, and more. Their work involved defining and implementing materials, geometry modifications, and animation techniques using Blender's Python API for the generation of photorealistic worlds.
princeton-vl/SEA-RAFT
May 2024 - Mar 2025
[ECCV2024 - Oral, Best Paper Award Candidate] SEA-RAFT: Simple, Efficient, Accurate RAFT for Optical Flow
Contributions:12 pushes, 1 branch, 23 comments in 10 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.