Joong-Hwan Baek

Papers

Total Citations

H-Index

About

Joong-Hwan Baek’s research spans medical imaging, mobile robotics, and computer vision, with a focus on practical, real-time systems. His most impactful work, “Unsupervised Monocular Depth Estimation for Colonoscope System Using Feedback Network” (2021, 26 citations), addresses a critical challenge in colorectal cancer screening: improving polyp detection rates during colonoscopy. By developing a feedback network that estimates depth from monocular video without ground-truth data, Baek enhances the spatial awareness of endoscopic systems, potentially increasing diagnostic accuracy and reducing missed abnormalities. Earlier, he contributed to autonomous navigation with “Real Time Obstacle Avoidance for Mobile Robot Using Limit-Cycle and Vector Field Method” (2006, 16 citations), a method that enables robots to avoid obstacles smoothly in dynamic environments. His work on “The Character Recognition System of Mobile Camera Based Image” (2010, 3 citations) extends his expertise to assistive technologies, such as text recognition for visually impaired navigation aids. Across these projects, Baek demonstrates a consistent drive to translate algorithmic innovations into deployable solutions for healthcare, robotics, and accessibility, making his research both technically rigorous and socially valuable.