Papers
1
Total Citations
7
H-Index
1
About
Junpeng Lin’s research centers on advancing computer vision, with a particular focus on pedestrian detection—a critical task for autonomous driving, intelligent surveillance, and robotics. His most-cited work, “Feature Fusing of Feature Pyramid Network for Multi-Scale Pedestrian Detection” (2018, 7 citations), tackles the persistent challenge of detecting pedestrians at varying scales in real-world images. Lin’s major contribution lies in enhancing Feature Pyramid Networks through innovative feature fusion techniques, which improve detection accuracy for both small and large pedestrians. This work addresses a fundamental bottleneck in vision systems, where scale variation often leads to missed detections or false positives. By refining how multi-scale features are integrated, Lin’s approach boosts robustness in cluttered environments, directly impacting safety-critical applications like autonomous driving. His research has garnered attention from peers working on object detection and scene understanding, with citations reflecting its relevance to ongoing advancements in deep learning architectures. Lin’s contributions underscore his role in bridging theoretical improvements with practical deployment, making him a notable figure in the computer vision community. For students and researchers, his work exemplifies how targeted architectural innovations can solve real-world vision problems.
Research Focus
Key Achievements
Top Papers
- 1