Byungjun Kim

I am currently a MS/PhD student at Seoul National University (SNU), advised by Hanbyul Joo. My research primarily focuses on 3D digital human modeling through the lens of generative models, with a particular interest in compositional modeling.
Recently, I’ve been expanding my interest toward human-environment interaction, which naturally extends to robotics. My long-term goal is to bridge digital human simulation with real-world robotics to enable collaborative agents.

Email  /  CV  /  Github

We present HairCUP, a universal prior model for 3D head avatars with hair compositionality, which enables hairstyle swapping and efficient personalization.

We present GALA, a framework that takes as input a single-layer clothed 3D human mesh and decomposes it into complete multi-layered 3D assets.

We present, PEGASUS, a method for constructing personalized generative 3D face avatars from monocular video sources.

We present Guess The Unseen, a method to reconstruct the world and multiple dynamic humans in 3D from a monocular video input.

We propose Chupa, a 3D human generation pipeline that combines the generative power of diffusion models and neural rendering techniques to create diverse, realistic 3D humans.

We propose a novel self-supervised learning framework for snow points removal in LiDAR point clouds.