Byeongkeun Kang

University of California San Diego, Seoul National University of Science and Technology

Papers

Total Citations

H-Index

About

Byeongkeun Kang is a researcher whose work bridges computer vision, robotics, and human-computer interaction, with a particular focus on robust perception systems for real-world applications. His key research areas include hand segmentation, unsupervised image segmentation, and autonomous systems. Kang’s most impactful contribution is his work on hand segmentation for hand-object interaction using depth maps, which addresses a critical limitation of color-based methods—their vulnerability to skin-colored objects and varying skin tones. This foundational work, published in 2017, has garnered 38 citations and is essential for applications in augmented reality, medical imaging, and human-robot interaction. He further advanced the field with a pixel-level clustering network for unsupervised image segmentation (2023, 28 citations), offering a novel approach to segmentation without labeled data. Notably, Kang has also applied his expertise to autonomous EV charging systems, developing image-to-image translation-based data augmentation for robust charging inlet detection (2022, 13 citations), a practical solution for improving the reliability of autonomous charging robots. His work demonstrates a consistent commitment to solving real-world challenges through innovative computer vision techniques, making him a notable contributor to both theoretical and applied research.