Kyoung Sig Roh

Samsung (South Korea), Korea Advanced Institute of Science and Technology

Papers

Total Citations

H-Index

About

Kyoung Sig Roh is a researcher in computer vision and mobile robotics, with a focus on projective geometry and its application to autonomous navigation. His work centers on developing mathematically elegant, calibration-free methods for 3D object recognition and robot self-localization. In his most cited paper, "3-D Object Recognition Using a New Invariant Relationship by Single-View" (2000, 16 citations), Roh introduced a novel invariant relationship that enables robust object recognition from a single perspective, reducing dependence on complex sensor setups. His follow-up work, "Obstacle Detection and Self-Localization without Camera Calibration Using Projective Invariants" (2002, 14 citations), proposed two practical vision-based methods for indoor mobile robot navigation: a self-localization algorithm leveraging projective invariants, and a simple yet effective obstacle detection technique using image differencing and relative positioning within corridor environments. Though his citation counts are modest, Roh’s contributions are notable for their theoretical rigor and practical simplicity, offering efficient solutions for real-world robotics challenges. His work remains relevant for researchers seeking calibration-robust approaches to visual navigation and object recognition.