Sangmin Kim

I am currently a second year student in the Integrated M.S. and Ph.D. Program at Seoul National University, where I am advised by Jaesik Park. I received my B.S. degree in Computer Science and Engineering from Chung-Ang University with honors of magna cum laude in 2024.

My research interests are in 3D computer vision and deep learning. Especially, I am interested in computer vision applications for post-production and visual contents.

Email / CV / LinkedIn / Google Scholar / Github

🚩 News

Feb 27, 2025 A paper about TV show dynamic reconstruction has been accepted to CVPR 2025.

🔬 Work Experience

Internship
Assisted consulting on revision of Shinhan Bank's ICT regulation.
Deloitte Korea (Nov 2021 - Feb 2022)

🏆 Honors and Awards

M.S. Fellowship
Funded by Korean Ministry of Education
750 M.S. students selected nationwide (Sep 2024 - Aug 2025)
Excellence Award
Awarded by Chung-Ang University
Capstone Design Competition (June 2024)
Excellence Award
Awarded by National IT Industry Promotion Agency of Korea
SW Development Competition (June 2023)
Academic Scholarship
Awarded by Chung-Ang University (Jan 2023)

📚 Publications

Optimized Minimal 4D Gaussian Splatting
Minseo Lee*, Byeonghyeon Lee*, Lucas Yunkyu Lee, Eunsoo Lee, Sangmin Kim, Seunghyeon Song, Joo Chan Lee, Jong Hwan Ko, Jaesik Park, Eunbyung Park
arXiv, 2025
paper / code

We present OMG4 (Optimized Minimal 4D Gaussian Splatting), a framework that constructs a compact set of salient Gaussians capable of faithfully representing 4D Gaussian models.

ShowMak3r: Compositional TV Show Reconstruction
Sangmin Kim, Seunguk Do, Jaesik Park
CVPR, 2025
paper / code

A comprehensive pipeline that reconstructs the dynamic radiance field from TV shows, enabling viewpoint editing like how the video clip is made in a production control room.

Website's source code is referenced from Jon Barron's.