Papers
13
Total Citations
156
H-Index
6
About
Soohwan Kim is a robotics researcher whose work spans probabilistic mapping, vision-based localization, and human-robot interaction. He is perhaps best known for his contributions to Gaussian process-based robotic mapping, most notably the GPmap framework (2014, 47 citations), which established a unified approach to spatial representation using sparse Gaussian processes. Building on this foundation, his work on recursive Bayesian updates for occupancy mapping and surface reconstruction introduced an elegant single-framework solution capable of generating both occupancy maps and surface meshes, advancing the field beyond batch-processing limitations. Kim's earlier research made significant strides in vision-based global localization, developing hybrid map representations that combine topological and metric approaches to help mobile robots reliably determine their position in complex indoor environments (2009, 32 citations). His contextual object recognition work, particularly for door detection, demonstrated how incorporating robot-specific contextual knowledge can meaningfully improve recognition efficiency. More recently, Kim has addressed the growing demand for multimodal sensing with a targetless, structureless approach to spatiotemporal camera-LiDAR calibration. Across his career, his research reflects a consistent drive to make robotic perception more robust, scalable, and practically deployable — contributions that continue to influence both academic robotics research and real-world autonomous systems development.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Recursive Bayesian Updates for Occupancy Mapping and Surface Reconstruction18 citations · 2014
- 4Context-based object recognition for door detection14 citations · 2011
- 5Human augmented mapping for indoor environments using a stereo camera13 citations · 2009
- 6
- 7Vision-based global localization based on a hybrid map representation6 citations · 2008
- 8Hierarchical Gaussian Processes for Robust and Accurate Map Building5 citations · 2015
- 9Real-time arm motion imitation for human–robot tangible interface4 citations · 2009
- 10