I am a Ph.D. candidate in Scalable Graphics, Vision, and Robotics (SGVR) Lab at KAIST advised by Prof. Sung-Eui Yoon. My research focuses on generative AI for Physical AI, particularly in 3D scene understanding and action planning under incomplete real-world observations. I am broadly interested in developing models for autonomous driving and robotics that can understand and reconstruct dynamic 3D environments, enabling more reliable reasoning and interaction in the physical world.
jmlee [at] kaist.ac.kr
291 Daehak-ro, Yuseong-gu, Daejeon, Korea
Research Intern, Robot Vision & Learning Team, NAVER Labs
Nov. 2024 – May. 2025
B.S. in Artificial Intelligence and Mechanical Engineering, SungKyunKwan University
Sep. 2017 - Feb. 2022
What You Ask is What You Ground: Bridging Question Intent to Temporal Evidence for Grounded VideoQA
European Conference on Computer Vision (ECCV) 2026
MorphGS: Morphable Gaussian Splatting for 3D Scene Representation
European Conference on Computer Vision (ECCV) 2026
Visual-RRT: Finding Paths toward Visual-Goals via Differentiable Rendering
Computer Vision and Pattern Recognition (CVPR) 2026
Pose-free 3D Gaussian splatting via shape-ray estimation
International Conference in Image Processing (ICIP) 2025
Diffusion Probabilistic Models for Scene-Scale 3D Categorical Data
Image Processing and Image Understanding (IPIU) 2023