Sumin Zhang
Papers
1
Total Citations
13
H-Index
1
About
Sumin Zhang is a researcher advancing the field of visual simultaneous localization and mapping (VSLAM), with a focus on enhancing autonomous navigation through deep learning. Their key research areas include stereo visual odometry, unsupervised learning for pose estimation, and robust ego-motion tracking for robots and autonomous systems. Zhang’s most notable contribution, “Stereo Visual Odometry Pose Correction through Unsupervised Deep Learning” (2021), addresses critical limitations in classical visual odometry—such as reliance on rigid assumptions—by introducing a data-driven approach that corrects pose estimates without requiring labeled ground truth. This work has garnered 13 citations, reflecting its growing influence in the VSLAM community. By bridging traditional geometric methods with modern unsupervised learning, Zhang’s research offers a scalable path toward more reliable positioning and navigation in complex environments. Their work is particularly relevant for students and researchers exploring how deep learning can overcome the fragility of classical VO systems, making autonomous robots more adaptable in real-world settings.
Research Focus
Key Achievements
Top Papers
- 1Stereo Visual Odometry Pose Correction through Unsupervised Deep Learning13 citations · 2021