Byeong-Uk Lee

Korea Advanced Institute of Science and Technology

Papers

Total Citations

H-Index

About

Byeong-Uk Lee is a researcher at the forefront of computer vision and robotics, tackling some of the field’s most challenging real-world problems. His work centers on developing intelligent systems that can perceive and navigate complex environments, with a particular focus on image dehazing and disaster response. Lee’s major contribution is a pioneering CNN-based approach that simultaneously performs image dehazing and depth estimation, a breakthrough that addresses the critical difficulty cameras and depth sensors face in hazy scenes. This work, published in 2020, has already garnered 21 citations, underscoring its impact on advancing robust visual perception. Building on this, Lee created a large-scale virtual dataset for egocentric localization in disaster scenarios, a vital resource that overcomes the shortage of training data in this domain. This 2021 paper, with 8 citations, highlights his commitment to applying computer vision to life-saving applications, enabling more effective visual observation for rescue and safety operations. By bridging the gap between theoretical computer vision and practical, high-stakes environments, Byeong-Uk Lee is shaping a future where machines can see clearly and act decisively, even in the most adverse conditions.