Max Robotics — Global Service Robot Directory + US Certification

About

Yueh-Hua Wu is a robotics and machine learning researcher whose work spans dexterous robotic manipulation, imitation learning, and vision-language models. His research tackles some of the most challenging problems at the intersection of computer vision and robot learning, with a particular focus on enabling robots to perform complex, human-like manipulation tasks in unstructured environments. Wu's most influential contribution, "DexMV: Imitation Learning for Dexterous Manipulation from Human Videos" (2022, 117 citations), demonstrated how human video demonstrations could be leveraged to teach multi-finger robotic hands intricate manipulation skills—significantly improving sample efficiency over traditional reinforcement learning approaches. Building on this, his work on generalizable dexterous manipulation explored how robots can transfer grasp affordances to novel objects, addressing a critical bottleneck in real-world deployment. His subsequent research broadened in scope, with GNFactor introducing generalizable neural feature fields for multi-task robot learning, and DNAct combining neural rendering with diffusion-based policy learning for robust 3D manipulation. More recently, Wu has extended his expertise into vision-language model compression through VLsI. Collectively, his growing citation record reflects an emerging research voice bridging perception, manipulation, and scalable AI systems for embodied intelligence.