3D Computer Vision / Pose and Mesh Reconstruction
I am a combined MS/PhD student at the Artificial Intelligence Graduate School (AIGS), UNIST, and a member of the UNIST Vision and Learning Lab (UVLL), advised by Prof. Seungryul Baek. I received my B.S. degree in Statistics from Chonnam National University.
My research focuses on 3D human and hand-object reconstruction from monocular videos, with broader interests in multi-person tracking, 3D computer vision, and human-object interaction.
Our paper "Multi-THuMBS: Multi-person Tracking of 3D Human Meshes Beyond Video Shots" has been accepted to .
Our paper "BIGS: Bimanual Category-agnostic Interaction Reconstruction from Monocular Videos via 3D Gaussian Splatting" has been accepted to .
Won 1st place at the HANDS Workshop Challenge, ARCTIC Track.
Won 3rd place at the HANDS Workshop Challenge, Consistent Motion Reconstruction.
Multi-THuMBS: Multi-person Tracking of 3D Human Meshes Beyond Video Shots
1st Place Solution to the 8th HANDS Workshop Challenge - ARCTIC Track: 3DGS-based Bimanual Category-agnostic Interaction Reconstruction
Class-Wise Buffer Management for Incremental Object Detection: An Effective Buffer Training Strategy
InstantSplat: Sparse-view SfM-free Gaussian Splatting in Seconds (arXiv, 2024)
Can Generative Video Models Help Pose Estimation? (Inter-pose)
D-SCo: Dual-Stream Conditional Diffusion for Monocular Hand-Held Object Recon...
HORSE: 3D Reconstruction of Objects in Hands without Real World 3D Supervision
HUGS: Human Gaussian Splats
HOLD: Category-agnostic 3D Reconstruction of Interacting Hands and Objects fr...
Shape of Motion: 4D Reconstruction from a Single Video
Dance In the Wild: Monocular Human Animation with Neural Dynamic Appearance S...