Tae-Kyun Kim

Tae-Kyun Kim is a Faculty member at the School of Computing, KAIST, where he leads the Computer Vision and Learning Lab (KCVL). He has been an adjunct reader at Imperial College London since 2020 and previously directed the Computer Vision and Learning Lab there from 2010 to 2020. His research focuses on machine (deep) learning for 3D computer vision, generative AI, and AI with physics, including topics such as 3D hand/body reconstruction, face analysis, and 6D object pose estimation. He has co-authored over 120 academic papers and has organized several workshops in the field. He is also an Associate Editor of several journals and has served as an Area Chair for top-tier conferences.

Tae-Kyun (T-K) Kim is a researcher in computer vision, deep learning, machine learning, robotics, and action recognition. His work focuses on advancing techniques for object tracking, action classification, and 3D pose estimation. He has contributed to benchmarks such as BOP for 6D object pose estimation and developed methods for hand pose estimation using RGB-D data. His research includes discriminative learning, canonical correlation analysis, and latent regression forests for structured estimation. Kim's publications address challenges in action recognition, multimodal classification, and real-time hand pose estimation. His work emphasizes the integration of geometric and semantic information for robust and efficient vision systems.