Hyunjin Kim

I'm a 3D Vision Engineer / Researcher at KRAFTON AI Research Center. I received my Master's degree at KAIST Visual AI Group, advised by Prof. Minhyuk Sung. I received my Bachelor's degree double majoring in Mathematics and Computer Science from the KAIST in 2021.

I'm interested in various topics for both 2D and 3D computer vision, such as generative AI, manipulation, segmentation, NeRF, and multimodal tasks. I really enjoy learning new things, and I am thoroughly prepared to adapt quickly and immerse myself in any field!

My Erdős number is 4 (Hyunjin Kim → Minhyuk Sung → Leonidas Guibas → Frances Yao → Paul Erdős)

Email  /  Google Scholar  /  CV

profile photo

Publications
PartSTAD: 2D-to-3D Part Segmentation Task Adaptation
Hyunjin Kim, Minhyuk Sung
ECCV, 2024
arXiv / Project / Code
SyncDiffusion: Coherent Montage via Synchronized Joint Diffusions
Yuseung Lee, Kunho Kim, Hyunjin Kim, Minhyuk Sung
NeurIPS, 2023
arXiv / Project / Code
Pop-Out Motion: 3D-Aware Image Deformation via Learning Shape Laplacian
Jihyun Lee*, Minhyuk Sung*, Hyunjin Kim, Tae-Kyun (T-K) Kim (* denotes equal contribution)
CVPR, 2022
arXiv / Project / Code / Video

Projects
Reimplementing BARF
Reimplementing BARF (Bundle-Adjusting Neural Radiance Fields) from scratch.
KAIST CS492 Machine Learning for 3D Data, Spring 2022
Report / Supplementary
TCNet
Improving amodal instance segmentation method BCNet via trilayer structure.
KAIST CS580 Artificial Intelligence and Machine Learning, Spring 2022
Report

Teaching Experiences

Teaching Assistant, KAIST CS479 Machine Learning for 3D Data, Fall 2023

Teaching Assistant, KAIST CS380 Introduction to Computer Graphics, Spring 2023

Academic Services

Reviewer

  • CVPR 2024
  • CVPR 2023

    		•

    Template from Jon Barron