Zhe Cao
Research Scientist at Google
United States
Summary
🤩
Rockstar🎓
Top SchoolZhe Cao is a research scientist with 11 years focused on computer vision and deep learning, currently advancing AR/VR and foundation models for human understanding at Google. He holds a PhD from UC Berkeley and has driven 3D virtual human and AR/VR research at Meta Reality Labs and industry research internships at Facebook/FAIR and FRL. His open-source work includes influential real-time multi-person pose estimation code (CVPR'17 oral), reflecting hands-on expertise in model inference, data pipelines, and evaluation at deployment scale. Comfortable bridging academia and product-focused research, he teaches and mentors as well as ships reproducible code—combining rigorous theory with practical systems thinking.
11 years of coding experience
7 years of employment as a software developer
Bachelor's degree, Computer Science, Bachelor's degree, Computer Science at Wuhan UniversityDoctor of Philosophy - PhD, Computer Vision, Deep Learning, Doctor of Philosophy - PhD, Computer Vision, Deep Learning at University of California, BerkeleyMaster's degree, Computer Vision, Master's degree, Computer Vision at Carnegie Mellon University
Github Skills (11)
computer-vision10
machine-learning10
eval10
caffe10
deep-learning10
evaluation10
pose-estimation10
coco9
matlab9
dataprep8
data-preprocessing8
Programming languages (3)
Github contributions (5)
ZheC/Realtime_Multi-Person_Pose_Estimation
Dec 2016 - Feb 2020
Code repo for realtime multi-person pose estimation in CVPR'17 (Oral)
Role in this project:
ML Engineer
ML EngineerContributions:144 commits, 8 PRs, 136 pushes in 3 years 2 months
Contributions summary:Zhe primarily contributed to the project by modifying and adding code related to model application and evaluation within the context of a real-time multi-person pose estimation system. Their changes include updates to the `applyModel` function, indicating involvement in the core inference process. Furthermore, they added code for generating LMDB files, COCO results, and MPII results, demonstrating an understanding of data preparation, evaluation metrics, and potentially training pipelines. The user's commits suggest a focus on the performance and evaluation of the pose estimation model.
pytorchperson-pose-estimationrealtimepersoncomputer-vision
ZheC/GTA-IM-Dataset
Jun 2020 - Jul 2020
Contributions:13 commits, 6 pushes, 6 comments in 24 days
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.