Hyunjun Lim

Korea Advanced Institute of Science and Technology

Papers

Total Citations

H-Index

About

Hyunjun Lim is a robotics researcher whose work sits at the intersection of visual SLAM, depth perception, and autonomous navigation. His primary research focuses on enabling robots to build accurate maps of their environments while understanding their own position within them—a fundamental challenge for service and field robotics. Lim’s most influential contribution is **Struct-MDC**, a novel unsupervised depth completion method that leverages structural regularities from visual SLAM to transform sparse feature-based depth maps into dense, mesh-refined reconstructions. This work, which has garnered 15 citations, directly addresses the critical sparsity problem in feature-based SLAM, offering a practical solution for robots that need rich environmental data without expensive sensors. Beyond depth completion, Lim has explored unconventional platforms, such as designing a mole-like excavate robot with specialized localization methods for deep-ground operations, demonstrating his versatility in tackling real-world robotic challenges. His recent **CLOi-Mapper** framework further showcases his commitment to practical deployment, providing a consistent, lightweight, and robust SLAM solution optimized for embedded systems in commercial service robots. With a growing citation record and a focus on bridging algorithmic innovation with hardware constraints, Hyunjun Lim is a rising figure in robotic perception and mapping.