πŸ‘‹ Home

πŸ‘€ About Me

Nice to meet you! I’m Hail Song, a Ph.D. student at the UVR Lab, KAIST Graduate School of Culture Technology, under the supervision of Prof. Woontack Woo (with a minor in the Graduate School of Metaverse).

My research lies at the intersection of neural rendering, 3D human avatar reconstruction, and immersive XR systems. I develop both reconstruction methods and real-time XR avatar systems for creating and deploying high-fidelity 3D avatars from accessible inputs such as single images, monocular videos, and sparse egocentric signals. My work spans single-image head avatar generation, full-body avatar reconstruction, and real-time 3D Gaussian avatar systems for immersive XR.

Previously, I worked as a computer vision engineer at InBody, and I have hands-on experience in C++, Python, and building end-to-end AR/VR systems. I’m deeply interested in bridging cutting-edge research and real-world application, especially in the context of extended reality (XR).

If you’re curious about my work or interested in collaboration, feel free to get in touch!

My avatar rendered with VRGaussianAvatar, presented at IEEE VR 2026 (TVCG Journal Track).

I find joy in bringing virtual humans to life through my research. Take a look at my virtual self in this video! The dancing sequence is an implementation of LHM, a fascinating recent paper, whereas the other parts are directly tied to my own research. For more details, please see my publications.


πŸ“ Contact

βœ‰οΈ hail96@kaist.ac.kr
πŸ”— LinkedIn ・ GitHub