Peihao Chen
Papers
3
Total Citations
38
H-Index
3
About
Peihao Chen is a leading researcher at the intersection of computer vision, robotics, and embodied AI, with a core focus on vision-and-language navigation (VLN) and 3D world models. His work tackles the fundamental challenge of enabling robots to understand and act within complex 3D environments using natural language instructions. Chen’s major contributions include pioneering weakly-supervised learning for VLN, as demonstrated in his 2022 paper "Weakly-Supervised Multi-Granularity Map Learning for Vision-and-Language Navigation" (16 citations), which reduces the need for expensive human annotations by learning to map language descriptions to environmental objects. He also introduced "3D-VLA: A 3D Vision-Language-Action Generative World Model" (2024, 14 citations), a groundbreaking framework that extends 2D VLA models into the 3D physical world, enabling agents to reason about world dynamics and object relations before acting. Additionally, his work on "Learning Active Camera for Multi-Object Navigation" (2022, 8 citations) addresses efficient exploration with camera sensors, advancing autonomous navigation. With over 38 citations across his top papers, Chen’s research is shaping the future of intelligent, language-guided robotic systems.
Research Focus
Key Achievements
Top Papers
- 1
- 23D-VLA: A 3D Vision-Language-Action Generative World Model14 citations · 2024
- 3Learning Active Camera for Multi-Object Navigation8 citations · 2022