Chen Gao
Research Scientist at Meta Reality Labs
Bellevue, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolChen Gao is a Senior Research Scientist focused on 3D/4D simulation and reconstruction for autonomous driving, currently at Waymo after shipping Meta’s Hyperscape at Reality Labs. With a PhD from Virginia Tech and nine years in research roles across Meta, Google, and Facebook, he combines academic rigor with production-facing generative AI systems. His work spans high-fidelity 3D/4D reconstruction, novel-view synthesis, and video completion—contributions that include adapting RAFT-based optical flow and edge-guided techniques for ECCV-grade video completion. Based in Bellevue, WA, he blends deep computer vision research with practical system integration, routinely moving ideas from prototype to shipped product.
9 years of coding experience
Master’s Degree, Electrical and Computer Engineering, Master’s Degree, Electrical and Computer Engineering at University of MichiganDoctor of Philosophy, Computer Engineering, Doctor of Philosophy, Computer Engineering at Virginia TechBachelor’s Degree, Electrical and Computer Engineering, Computer Science, Bachelor’s Degree, Electrical and Computer Engineering, Computer Science at Oregon State University
Chinese, English
Github Skills (9)
video-processing10
pytorch10
computer-vision10
python10
optical-flow10
deep-learning8
machine-learning8
c-language4
cprogramming-language4
Programming languages (3)
Github contributions (5)
vt-vl-lab/FGVC
Sep 2020 - Dec 2021
[ECCV 2020] Flow-edge Guided Video Completion
Role in this project:
ML Engineer
ML EngineerContributions:12 commits, 10 pushes, 58 comments in 1 year 3 months
Contributions summary:Chen primarily contributed to the video completion project by implementing and refining the optical flow calculation and completion modules. Their work included adapting the RAFT model, integrating non-local flow estimation, and refining the overall video processing pipeline. Key changes involved adjusting the code for frame loading, ensuring only the first three channels were loaded, and integrating edge-guided completion techniques. These modifications indicate a focus on enhancing the accuracy and efficiency of the video completion process.
eccvcomputer-visionedgeeccv-2020video
vt-vl-lab/iCAN
Aug 2018 - Aug 2020
Contributions:9 commits, 6 pushes, 48 comments in 2 years
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.