Hyeongjun Yoo

Gwangju Institute of Science and Technology

Papers

Total Citations

H-Index

About

Dr. Hyeongjun Yoo is a rising researcher in computer vision and robotic perception, with a primary focus on developing robust visual systems that perform reliably under adverse environmental conditions. His most influential work, "Learning to See in the Rain via Disentangled Representation" (2021), addresses a critical challenge in autonomous navigation and robotics: image degradation caused by rain and haze. Yoo's key contribution lies in pioneering disentangled representation learning for de-raining, a novel approach that separates rain-specific features from underlying scene content, enabling more effective and generalizable restoration. This work has garnered 3 citations, establishing a foundation for subsequent research in weather-robust vision. His research sits at the intersection of deep learning, image restoration, and robot vision, aiming to bridge the gap between laboratory-trained models and real-world deployment where weather conditions are unpredictable. By tackling the fundamental problem of visual analytics failure in rain, Yoo's contributions hold significant promise for advancing the reliability of autonomous vehicles, drones, and outdoor robotic systems. His work represents an important step toward creating vision systems that can "see" clearly regardless of weather, a crucial capability for safe and practical real-world AI.