Papers
2
Total Citations
16
H-Index
2
About
Yuheng Pan is a researcher focused on the intersection of computer vision, autonomous systems, and smart city applications. Their primary contributions lie in developing efficient, real-time frameworks for automated parking violation detection (PVD) using drone technology. Pan introduced a novel "suspect-and-investigate" framework that can be embedded directly into drones, enabling autonomous surveillance and ticketing. A key innovation within this work is SwiftFlow, an efficient and accurate convolutional neural network (CNN) designed for unsupervised optical flow, which allows drones to detect and track moving vehicles with minimal computational overhead. Their most-cited paper, with 11 citations, lays the groundwork for this approach, while a subsequent 2020 publication refines the system for practical deployment. Pan’s work has significant implications for urban traffic management, law enforcement automation, and the broader field of edge AI, demonstrating how lightweight neural networks can enable complex real-world tasks on resource-constrained platforms. Their research is a notable step toward fully autonomous drone-based infrastructure monitoring.
Research Focus
Key Achievements
Top Papers
- 1
- 2ATG-PVD: Ticketing Parking Violations on A Drone5 citations · 2020