Papers
6
Total Citations
70
H-Index
4
About
Ue-Hwan Kim is a leading researcher at the intersection of robotics, computer vision, and artificial intelligence, with a focus on enabling intelligent systems to perceive, navigate, and interact with dynamic, real-world environments. His work spans three core areas: **smart home robotics and human-robot interaction**, **visual odometry and depth estimation**, and **robust visual perception for autonomous systems**. Kim’s major contributions include developing the Stabilized Feedback Episodic Memory (SF-EM) framework for robot-IoT collaboration in smart homes, and the SimVODIS++ system, a novel neural visual odometry approach that maintains accuracy in dynamic environments by leveraging semantic understanding. He has also advanced change detection through dual-task learning that exploits both dense correspondences and mis-correspondences, and has revisited self-supervised monocular depth estimation to improve its reliability. With over 70 citations across his most influential works, Kim’s research has been recognized for its practical impact on autonomous navigation and service robotics. Notably, his work on SimVODIS++ and change detection addresses fundamental challenges in deploying robots and autonomous vehicles in unpredictable, real-world settings. Kim’s contributions continue to shape how machines perceive and act within complex, human-centered spaces.
Research Focus
Key Achievements
Top Papers
- 1
- 2SimVODIS++: Neural Semantic Visual Odometry in Dynamic Environments20 citations · 2022
- 3
- 4Revisiting Self-supervised Monocular Depth Estimation9 citations · 2022
- 5
- 6On the Study of Data Augmentation for Visual Place Recognition2 citations · 2023