Sang Feng
Papers
2
Total Citations
10
H-Index
1
About
Sang Feng is an emerging researcher specializing in visual simultaneous localization and mapping (V-SLAM), with a focused emphasis on enabling robust autonomous navigation in dynamic and complex real-world environments. His work addresses one of the most persistent challenges in intelligent robotics: maintaining accurate localization when scenes contain moving objects that confound traditional mapping systems. His most cited contribution, "ADM-SLAM: Accurate and Fast Dynamic Visual SLAM with Adaptive Feature Point Extraction, Deeplabv3pro, and Multi-View Geometry" (2024), has already garnered 9 citations since its publication, reflecting strong early interest from the robotics and computer vision communities. This work integrates deep semantic segmentation with geometric reasoning to effectively identify and discard dynamic objects, significantly improving localization accuracy and processing speed. His follow-up work, "SEGL-SLAM" (2025), extends this research by incorporating transformer-based segmentation and line feature enhancement, pushing the boundaries of SLAM performance in challenging environments. Feng's research sits at the intersection of deep learning, geometric computer vision, and autonomous systems — a rapidly growing field with broad applications in self-driving vehicles, drone navigation, and service robotics. His consistent focus on practical, high-performance solutions positions him as a promising contributor to next-generation autonomous navigation research.
Research Focus
Key Achievements
Top Papers
- 1
- 2