Durkhyun Cho

Papers

Total Citations

H-Index

About

Durkhyun Cho is a researcher at the forefront of privacy-preserving computer vision and human-robot interaction. His work primarily addresses two critical challenges: safeguarding personal privacy in an increasingly surveilled world and enabling natural, real-time communication between humans and machines. Cho’s most impactful contribution is the "CLEANIR: Controllable Attribute-Preserving Natural Identity Remover" (2020), which has garnered 27 citations. This work tackles the urgent need for privacy in the age of smartphones, service robots, and surveillance cameras, moving beyond simple masking or blurring to develop a sophisticated system that removes identifiable features while preserving natural attributes—a crucial advance for ethical AI deployment. Earlier in his career, Cho focused on enhancing human-robot interaction through robust facial feature tracking. His 2013 and 2014 papers, with 5 and 2 citations respectively, innovatively combined particle filters with active appearance models to overcome the slow processing speeds that hindered real-time applications. By improving the efficiency and accuracy of facial feature tracking, Cho laid essential groundwork for more responsive and intuitive robotic systems. His research trajectory—from enabling machines to see us clearly to ensuring they can do so without compromising our privacy—marks him as a thoughtful engineer addressing both the technical and ethical dimensions of modern AI.