Sungil Kang

Papers

Total Citations

H-Index

About

Sungil Kang is a robotics researcher whose work centers on real-time perception systems for mobile robots, with a particular focus on vision-based people and pedestrian detection. His key contributions lie in leveraging Graphics Processing Units (GPUs) to accelerate computationally intensive computer vision tasks, enabling mobile robots to operate in dynamic, human-populated environments. In his 2011 paper on people detection using an omnidirectional camera and GPU-accelerated optical flow, Kang demonstrated how to efficiently identify regions of interest by analyzing ego-motion compliance—a method that has garnered 10 citations for its practical approach to real-time robotics. His companion work on pedestrian detection with stereo vision further advanced the field by using GPUs to compute dense disparity maps and extract edge-based regions of interest, achieving 9 citations for its contribution to safe robot navigation. Together, these papers showcase Kang’s impact on making mobile robots more aware of their surroundings, a critical step toward autonomous systems that can interact safely with humans. His research remains relevant for students and engineers developing low-latency perception pipelines for service and field robotics.