Zehao Yu
Zehao Yu is a PhD student at the University of Tübingen affiliated with the ELLIS program, co-supervised by Andreas Geiger and Torsten Sattler, and also affiliated with the Institute of Artificial Intelligence (TeleAI) at China Telecom. He is known for research in computer vision and machine learning, particularly 3D vision, neural surface reconstruction, and Gaussian splatting, with award-winning work including the CVPR 2024 Best Student Paper on Mip-Splatting. He is a first author of OASIS, a simulation-data-driven framework for real-world humanoid loco-manipulation.
“a policy trained entirely on sim-generated data achieves comparable or superior zero-shot performance on a real Unitree G1 humanoid compared to a policy trained on real-robot teleoperation data”
Source→“Listed as first author with Fudan University affiliation.”
Source→AI-extracted from podcast / newsletter / paper summaries. May contain errors.