Zerui Chen
Papers
1
Total Citations
2
H-Index
1
About
Zerui Chen is a rising researcher in robot learning, with a focus on vision-based dexterous manipulation and imitation learning from human demonstrations. His work addresses the critical challenge of enabling multi-fingered robot hands to manipulate diverse objects in varied poses, leveraging human video data to bypass the need for expensive teleoperation. His most-cited paper, "ViViDex: Learning Vision-based Dexterous Manipulation from Human Videos" (2024), introduces a unified vision-based policy that learns directly from noisy human video trajectories, tackling the performance degradation caused by estimation errors. This approach has already garnered early citations, signaling its potential to advance scalable robot learning. Chen’s contributions are particularly notable for bridging the gap between human demonstration data and robust robotic control, a key bottleneck in dexterous manipulation. His work is positioned to impact both the robotics and computer vision communities, offering a practical pathway to more adaptable and capable robotic hands. With a focus on data-driven policy learning, Chen is establishing himself as a promising voice in the intersection of imitation learning and manipulation.
Research Focus
Key Achievements
Top Papers
- 1ViViDex: Learning Vision-based Dexterous Manipulation from Human Videos2 citations · 2024