Papers
10
Total Citations
784
H-Index
8
About
Giseop Kim is a leading researcher in robotics, specializing in LiDAR-based place recognition, lifelong mapping, and SLAM (Simultaneous Localization and Mapping). His major contributions include the development of Scan Context++, a structural place recognition method robust to rotation and lateral variations in urban environments, which has garnered 288 citations. He also pioneered Removert, a static point cloud map construction algorithm that effectively removes dynamic objects, cited 189 times. His work on long-term LiDAR localization, demonstrated in "1-Day Learning, 1-Year Localization," achieves robust year-round performance with just a single day of training, earning 136 citations. Kim's impact extends to open-source frameworks like LT-mapper for lifelong mapping and SC-LiDAR-SLAM, a versatile SLAM system. Notably, his HeLiPR dataset addresses inter-LiDAR place recognition under spatiotemporal variations, advancing heterogeneous sensor fusion. With over 700 total citations, Kim's research is foundational for robust robot navigation in dynamic, real-world environments, making him a key figure in autonomous systems and spatial AI.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4LT-mapper: A Modular Framework for LiDAR-based Lifelong Mapping47 citations · 2022
- 5
- 6SC-LiDAR-SLAM: A Front-end Agnostic Versatile LiDAR SLAM System33 citations · 2022
- 7
- 8
- 9Ephemerality Meets Lidar-Based Lifelong Mapping5 citations · 2025
- 10