Papers
1
Total Citations
23
H-Index
1
About
Jingu Kim is a researcher in computer vision and real-time object tracking, with a focus on developing efficient, practical solutions for dynamic visual environments. His most cited work, "A real-time multi-class multi-object tracker using YOLOv2" (2017, 23 citations), addresses a critical bottleneck in multi-class multi-object tracking: achieving real-time performance without sacrificing accuracy. Kim’s key contribution lies in integrating the YOLOv2 detection framework with a tracking-by-detection paradigm, enabling simultaneous classification and tracking of multiple object classes—such as pedestrians, vehicles, and gestures—in live video streams. This work has direct implications for surveillance systems, gesture recognition, and robotics, where low-latency processing is essential. By demonstrating that high-speed detection can be effectively paired with robust tracking, Kim advanced the feasibility of deploying multi-object trackers in resource-constrained, real-world applications. His research underscores the importance of balancing computational efficiency with algorithmic complexity, a challenge central to modern computer vision. With 23 citations, this paper serves as a reference for subsequent studies on real-time tracking, reflecting Kim’s impact on bridging the gap between academic models and industrial deployment.
Research Focus
Key Achievements
Top Papers
- 1A real-time multi-class multi-object tracker using YOLOv223 citations · 2017