Xiangsai Feng
Papers
1
Total Citations
12
H-Index
1
About
Xiangsai Feng is a researcher in robotics and computer vision, with a primary focus on visual simultaneous localization and mapping (SLAM). His key contributions center on improving the robustness and accuracy of loop closure detection—a critical component for correcting drift in long-term SLAM systems. Feng’s most-cited work, "Loop Closure Detection based on Image Covariance Matrix Matching for Visual SLAM" (2021, 12 citations), introduces an innovative method that leverages image covariance matrices to enhance place recognition. By matching statistical features rather than raw pixel data, his approach reduces computational overhead while maintaining high recall, making it particularly suitable for resource-constrained platforms. This work addresses a fundamental challenge in autonomous navigation, enabling more reliable map consistency in dynamic or repetitive environments. Though early in his career, Feng’s contributions are gaining traction among researchers seeking efficient, real-time solutions for visual SLAM. His focus on covariance-based techniques represents a promising direction for scalable, low-latency perception systems in robotics and augmented reality applications.
Research Focus
Key Achievements
Top Papers
- 1