Hello, I'm Donghwan Kim

I am a PhD student in Computer Vision and Learning Lab at KAIST advised by Prof. Tae-Kyun Kim.

My dream is making people (also including me) happy. I feel happy in tackling challenging problems and exploring innovative ways to solve them. To make people happy, I am interested in research topics that enhance entertainment experiences particularly in fields such as AR/VR, telepresence, and creative content generation. Recently, my research has focused on 3D human reconstruction and human pose and shape estimation.

CV Scholar Github Linkedin

News

  • Oct. 2024: MHCDIFF has been accepted to NeurIPS 2024.
  • Jul. 2023: HOGraspNet has been accepted to ECCV 2024.
  • Sep. 2023: FourierHandFlow has been accepted to NeurIPS 2023.

Publications

Multi-hypotheses Conditioned Point Cloud Diffusion for 3D Human Reconstruction from Occluded Images

Multi-hypotheses Conditioned Point Cloud Diffusion for 3D Human Reconstruction from Occluded Images

Donghwan Kim, Tae-Kyun (T-K) Kim
NeurIPS, 2024

Project Page PDF arXiv
Dense Hand-Object(HO) GraspNet with Full Grasping Taxonomy and Dynamics

Dense Hand-Object(HO) GraspNet with Full Grasping Taxonomy and Dynamics

Woojin Cho, Jihyun Lee, Minjae Yi, Minje Kim, Taeyun Woo, Donghwan Kim, Taewook Ha, Hyokeun Lee, Je-Hwan Ryu, Woontack Woo, Tae-Kyun (T-K) Kim
ECCV, 2024

Project Page PDF arXiv
FourierHandFlow: Neural 4D Hand Representation Using Fourier Query Flow

FourierHandFlow: Neural 4D Hand Representation Using Fourier Query Flow

Jihyun Lee, Junbong Jang, Donghwan Kim, Minhyuk Sung, Tae-Kyun (T-K) Kim
NeurIPS, 2023

Project Page PDF arXiv