Papers
2
Total Citations
7
H-Index
2
About
Juyeop Han is a rising researcher in robotics and autonomous navigation, with a focus on visual-inertial systems and 3D environmental perception. His work bridges cutting-edge computer vision techniques with robust state estimation for real-world robotic applications. Han’s most notable contribution is **NVINS**, a novel framework that fuses traditional Visual Inertial Navigation (VINS) with Neural Radiance Fields (NeRF) to enhance camera pose regression and uncertainty quantification. This work addresses critical challenges in real-time navigation, such as computational cost and artifact-induced degradation in NeRF-based reconstruction. Additionally, Han developed **DS-K3DOM**, a 3D dynamic occupancy mapping approach that integrates kernel inference with Dempster-Shafer evidential theory, enabling robots to represent and reason about changing environments more effectively than conventional 2D methods. Though early in his career, with papers accumulating citations in the single digits, Han’s work demonstrates strong potential to impact autonomous systems, particularly in scenarios requiring robust perception under uncertainty. His research is especially relevant for students and engineers working on SLAM, sensor fusion, and robotic navigation in dynamic, unstructured settings.
Research Focus
Key Achievements
Top Papers
- 1
- 2