Under Review, 2026
M.S. Researcher · KAIST CVLAB
Seonghu Jeon.
I'm interested in how machines come to understand and represent the physical world, and how they can build on that to generate and act in it. I'm a first-year M.S. researcher at KAIST CVLAB. My research so far has centered on multi-view diffusion and 4D generation, and I'm now exploring 3D foundation models and geometric action models for robotics, including GAM, currently under review. Reach me at seonghu.jeon@kaist.ac.kr.
01
most recent first
News & notes
2026.06
GLD accepted to ECCV 2026 for repurposing geometric foundation models for multi-view diffusion.
2026.06
GAM is under review for geometric action modeling in visuomotor control.
2026.03
Joined KAIST CVLAB as an M.S. student under Prof. Seungryong Kim.
2026.02
CAMEO accepted to CVPR 2026 for correspondence-attention alignment in multi-view diffusion.
2025.07
ReMoTE accepted to ITC-CSCC.
2025.02
Graduated from Korea University, B.S. in Computer Science & Engineering, with Great Honors.
02
05 entries
Selected publications
03
three open threads
Focus
01
3D / 4D vision
Reconstructing static and dynamic scenes from sparse views, and studying how geometry-trained foundation models transfer to view synthesis, including GLD at ECCV 2026.
02
Generative models
Diffusion and flow matching with structured conditioning across correspondence, geometry, and motion. Architecture work and better source distributions.
03
Robotics
The destination. If a robot can imagine a scene's geometry forward in time, it can plan in it, and that loop closes through generative models.
04
Experience
KAIST CVLAB
2026 - now
M.S. Student
Advised by Prof. Seungryong Kim. Multi-view diffusion, geometric foundation models, 4D scene generation.
CVLAB, Korea University → KAIST
Dec 2023 - Feb 2026
Research Intern
Pre-graduate research on motion transfer and correspondence-attention. Multiple co-authored papers from this period.
Korea University
2022 - 2026
B.S., Computer Science & Engineering
Graduated with Great Honors (4.46 / 4.50). Coursework in vision, graphics, and deep learning.