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

8
H-Index
10
Papers
784
Total Citations
78
Avg Citations/Paper
🏆 Most Cited Paper
Scan Context++: Structural Place Recognition Robust to Rotation and Lateral Variations in Urban Environments
288 citations · 2021
📈 Most Prolific Year: 2021 (2 Papers)
🤝 Key Collaborators: 16
🏛 Institutions: Korea Advanced Institute of Science and Technology, Naver (South Korea)

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago