Tae-Hyun Oh

Tae-Hyun Oh is a faculty member in the School of Computing at KAIST. His research focuses on vision-centric multi-modal perception, spatial and physics-aware AI for embodied agents, and synthetic data generation. He leads the Advanced Machine Intelligence Lab (AMI Lab.) at KAIST, where his team explores audio-visual understanding, vision-language understanding, and alternative sensor fusion. Prof. Oh is also involved in spatial and physics-aware AI for embodied agents, including 3D scene and human understanding through synthesis. His work has been recognized with awards, including the Best Paper and Outstanding Poster Presentation Award from IPIU’26. He is affiliated with the AMI Lab. and can be contacted via email at taehyun.oh@kaist.ac.kr or through the lab's website at https://ami.kaist.ac.kr.

Tae-Hyun Oh is an Associate Professor at KAIST with expertise in Machine Learning, Computer Vision, and Graphics. His research focuses on developing advanced techniques for audio-visual scene understanding, including sound source localization, action recognition through audio-visual fusion, and robust image processing methods. He has contributed to areas such as federated learning, video motion magnification, and high dynamic range imaging. His work also explores cross-modal learning, such as speech-to-face synthesis and weakly-supervised image retargeting. Oh's research emphasizes the integration of deep learning with physical modeling for realistic image and video synthesis.