Shouyi Lu

Papers

Total Citations

H-Index

About

Shouyi Lu is a researcher advancing the field of visual simultaneous localization and mapping (VSLAM), with a particular focus on stereo visual odometry (VO) for autonomous navigation. His work addresses critical limitations in classical VO systems, which often rely on rigid assumptions that degrade performance in real-world environments. In his highly cited 2021 paper, "Stereo Visual Odometry Pose Correction through Unsupervised Deep Learning," Lu introduced an innovative approach that leverages deep learning to correct pose estimation errors without requiring labeled training data. This contribution has garnered 13 citations, reflecting its significance in improving the robustness and accuracy of ego-motion estimation for robots and autonomous vehicles. By integrating unsupervised learning techniques, Lu's research bridges the gap between traditional geometric methods and modern data-driven solutions, offering a more adaptable framework for VSLAM systems. His work is particularly impactful for applications in positioning and navigation, where reliable visual odometry is essential. Lu's achievements highlight his role in pushing the boundaries of autonomous perception, making him a notable figure in the robotics and computer vision communities.