I am a researcher at ADD, the Korean counterpart to the U.S. DARPA. I received my bachelor’s degree (Summa Cum Laude) in Computer Science and Engineering at POSTECH, where I was fortunate to work with Prof. Tae-Hyun Oh and Prof. Suha Kwak. During my undergraduate, I also studied at Stanford, UIUC and UC Berkeley.
My research interests lie in computer vision and machine learning. The goal of my research is to build intelligent agents that can perceive and reason about our dynamic 3D world, with the ability to interact as humans do. To this end, I focus on (i) efficient 3D/4D reconstruction from real-life photo/video, (ii) multi-modal understanding of real-world situations, and (iii) realistic generation of natural human behaviors. I have published first-authored papers at CVPR, ECCV and ICCV, and been recognized as an outstanding reviewer at ICCV 2023.
News
10/2023 | I am selected as an outstanding reviewer (top 1.89%) at ICCV 2023. |
---|---|
07/2023 | Our paper on domain generalization, PromptStyler, is accepted to ICCV 2023. |
07/2022 | Our paper on 3D human reconstruction, FastMETRO, is accepted to ECCV 2022. |
03/2022 | Our paper on situation recognition, CoFormer, is accepted to CVPR 2022. |
01/2022 | I won the best research project award from POSTECH CSE. |
10/2021 | Our paper on situation recognition, GSRTR, is accepted to BMVC 2021. |
01/2021 | I won the minister’s award from Ministry of Science and ICT. |
Education
02/2018 - 02/2022 |
POSTECH, Pohang, South Korea B.S. in Computer Science and Engineering Summa Cum Laude |
---|---|
06/2021 - 08/2021 |
Stanford University, California, United States International Exchange Program (Summer 2021) GPA: 4.3/4.3 |
01/2021 - 05/2021 |
UIUC, Illinois, United States International Exchange Program (Spring 2021) GPA: 4.0/4.0 |
08/2018 - 12/2018 |
UC Berkeley, California, United States International Exchange Program (Fall 2018) |
Selected Publications
-
Object-Centric Domain Randomization for 3D Shape Reconstruction in the WildUnder Review, 2024
-
PromptStyler: Prompt-driven Style Generation for Source-free Domain GeneralizationIEEE/CVF International Conference on Computer Vision (ICCV), 2023
-
Cross-Attention of Disentangled Modalities for 3D Human Mesh Recovery with TransformersEuropean Conference on Computer Vision (ECCV), 2022
-
Collaborative Transformers for Grounded Situation RecognitionIEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022
Honors and Awards
ICCV Outstanding Reviewer (2023), IEEE/CVF International Conference on Computer Vision
|
Professional Services
Conference Reviewer
|