I am a Researcher at the Field Robotics Research Section, ETRI. I received my Ph.D. in Electrical Engineering from KAIST, advised by Professor Hyun Myung.
My research interests lie in vision-language navigation (VLN), geo-localization, and domain robustness. I conduct full-stack research on VLN, spanning photorealistic digital twins, synthetic data generation, vision-language-action models, and real-robot deployment. I also developed cross-view geo-localization and object detection models to improve localization accuracy under domain shifts. This research contributed to the deployment of autonomous vehicles and robotic systems in real-world environments.
dnwn24.etri [at] gmail.com
Field Robotics Research Section, ETRI, Daejeon, Republic of Korea
B.S in Mechanical Engineering, 2019
Korea University, Korea
M.S in Robotics Program, 2021
KAIST, Korea
Ph.D in Electrical Engineering, 2025
KAIST, Korea
Advisor: Prof. Hyun Myung
NeuDonatello: Uncertainty-Aware Framework for Accurate Neural SDF Learning
BMVC 2026, Accepted
[ Paper ]
VIRD: View-Invariant Representation through Dual-Axis Transformation for Cross-View Pose Estimation
CVPR 2026, Highlight
[ Paper ]
CoCoA-Mix: Confusion-and-Confidence-Aware Mixture Model for Context Optimization
ICML 2025
[ Paper ]
Adaptive Coverage Path Planning Policy for a Cleaning Robot with Deep Reinforcement Learning
ICCE 2022
[ Paper ]
Sample-efficient and occlusion-robust reinforcement learning for robotic manipulation via multimodal fusion dualization and representation normalization
Neural Networks 2025
[ Paper ]
X-MAS: Extremely Large-Scale Multi-Modal Sensor Dataset for Outdoor Surveillance in Real Environments
RA-L 2023
[ Paper ]
ECCV 2026 VLNVerse EMR Challenge, 1st Place
European Computer Vision Association (ECVA), Sep. 2026
Surround Modulation-Inspired Neural Network for Robust Image Classification
AFCV (Asian Federation of Computer Vision) Best Paper Award, KROC 2021
[ Paper ]