Junmyeong Lee

junmyeong_profile.jpeg

I am a Ph.D. student at the Graduate School of Artificial Intelligence at POSTECH and a member of the Computer Vision Lab, advised by Prof. Minsu Cho.

My research focuses on 3D vision and representation learning, with the goal of reconstructing expressive and geometrically consistent 3D/4D scene representations from visual observations. I am particularly interested in dynamic scene reconstruction, novel view synthesis, and integrating semantic cues to improve the geometric and temporal consistency of learned representations. If you are interested in my research, please feel free to contact me.

Education

Mar, 2024 - Present Pohang University of Science and Technology (POSTECH), Pohang, South Korea
Ph.D. Student, Graduate School of Artificial Intelligence
Advisor: Prof. Minsu Cho
Mar, 2018 - Feb, 2024 Gwangju Institute of Science and Technology (GIST), Gwangju, South Korea
B.S. in Electrical Engineering and Computer Science
Advisor: Prof. Hae-Gon Jeon
GPA: 4.13/4.5 (Magna Cum Laude)

Publications

  1. MoGaF: Space-Time Forecasting of Dynamic Scenes with Motion-aware Gaussian Grouping
    Junmyeong Lee*, Hoseung Choi*, and Minsu Cho(* : equal contribution)
    IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026
  2. RoDyGS: Robust Dynamic Gaussian Splatting for Casual Videos
    Yoonwoo Jeong*, Junmyeong Lee*, Hoseung Choi*, et al.(* : equal contribution)
    arXiv preprint, 2024
  3. CanonicalFusion: Generating Drivable 3D Human Avatars from Multiple Images
    Jisu Shin, Junmyeong Lee, Seongmin Lee, et al.
    European Conference on Computer Vision (ECCV), 2024

Experience

Aug, 2021 - Dec, 2022 VoyagerX, Seoul, Korea
Team vFlat - ML engineer
  • On-device inference optimization using TFLite and ONNX.
  • Build Millon-scale data pipeline for training and evaluation.
Jan, 2021 - May, 2021 Search CIC, Naver Corp., Seoul, Korea
Research Intern
  • Research Topic : Large-scale Vector Retrieval Optimization