Yang Fu

About

Yang Fu is a leading researcher at the intersection of computer vision and robotics, with a primary focus on dexterous manipulation and 6D object pose estimation. His most influential work, "DexMV: Imitation Learning for Dexterous Manipulation from Human Videos" (2022), has garnered 117 citations, pioneering a novel framework that bridges human demonstration and robotic learning to enable complex, multi-fingered hand control. This breakthrough addresses a critical bottleneck in robotics: transferring fine-grained human skills to machines. In the domain of 6D pose estimation, Fu has made foundational contributions through his semi-supervised and self-supervised approaches. His paper "Category-Level 6D Object Pose Estimation in the Wild" (29 citations) introduced a new dataset and learning paradigm that allows models to generalize to unseen object instances without exhaustive annotation. Extending this, his self-supervised work (8 citations) further reduces reliance on labeled data, tackling the persistent challenge of real-world deployment. By advancing both imitation learning and geometric correspondence, Fu’s research is shaping the future of autonomous systems that can perceive and interact with the physical world with unprecedented dexterity and adaptability.

Research Focus

Key Achievements

Top Papers

1. 1
   DexMV: Imitation Learning for Dexterous Manipulation from Human Videos

   117 citations · 2022
2. 2
   Category-Level 6D Object Pose Estimation in the Wild: A Semi-Supervised Learning Approach and A New Dataset

   29 citations · 2022
3. 3
   Self-Supervised Geometric Correspondence for Category-Level 6D Object Pose Estimation in the Wild

   8 citations · 2022

Key Collaborators

Contact & Links