Weihao Xuan
Papers
1
Total Citations
9
H-Index
1
About
Weihao Xuan is a robotics researcher whose work focuses on advancing self-supervised visual odometry and depth estimation for mobile robotic systems. His most cited paper, "MaskVO: Self-Supervised Visual Odometry with a Learnable Dynamic Mask" (2022, 9 citations), introduces a novel approach that addresses a critical limitation in deep learning-based ego-motion estimation: the confounding effects of dynamic objects in the scene. By incorporating a learnable dynamic mask, Xuan enables robots to jointly learn camera pose and depth maps more robustly, filtering out moving elements that traditionally degrade performance. This contribution is significant for autonomous navigation in real-world environments, where static scene assumptions often fail. Xuan's work sits at the intersection of computer vision and robotics, leveraging self-supervised learning to reduce reliance on costly labeled data. While his citation count is still growing, the innovative use of dynamic masking represents a meaningful step toward more reliable and adaptable visual odometry systems—a core capability for mobile robots operating in unpredictable settings.
Research Focus
Key Achievements
Top Papers
- 1MaskVO: Self-Supervised Visual Odometry with a Learnable Dynamic Mask9 citations · 2022