I’m Jiwan Seo, an AI researcher at KETI. My research focuses on object-centric video analysis, particularly Video Instance Segmentation (VIS) and Referring Video Object Segmentation (RVOS). Currently, I am researching 3D Object Reconstruction and 4D Gaussian Splatting (4DGS).

My CV (PDF)

🔥 News

• 2026.03: Joined Hologram Research Center, KETI.
• 2026.02:  🎉🎉 1 paper accepted to CVPR 2026

📝 Publications

ICCV 2025

CAVIS: Context-Aware Video Instance Segmentation

Seunghun Lee*, Jiwan Seo*, Kiljoon Han, Minwoo Choi, Sunghoon Im

Paper | Project | Code |

ICCV 2025

[LOMM: Latest Object Memory Management for Temporally Consistent Video Instance Segmentation]

Seunghun Lee*, Jiwan Seo, Minwoo Choi, Kiljoon Han, Jaehoon Jeong, Zane Durante, Ehsan Adeli, Sang Hyun Park, Sunghoon Im (DGIST X Stanford)

Paper | Project | Code

CVPR 2026

[CVA: Context-aware Video-text Alignment for Video Temporal Grounding]

Sungho Moon, Seunghun Lee, Jiwan Seo, Sunghoon Im

Under review

[Temporal Grounding as a Learning Signal for Referring Video Object Segmentation]

Seunghun Lee, Jiwan Seo*, Jeonghoon Kim, Siwon Kim, Hyokyung Jeon, Haeun Yoon, Zane Durante, Sang Hyun Park, Sunghoon Im (DGIST X Stanford)

Paper | Project | Code

🖥️ Industrial Projects

Industry project

[Lightweight vision model for retail checkout automation]

DGIST & Ki*** Electronics collaborated work

(Keywords: Object Detection, Edge Device)

📖 Educations

• 2024.03 - 2026.02, M.S. course, Artificial Intelligence, DGIST CVLAB
• 2018.03 - 2024.02, Undergraduate, Computer Engineering, Yeungnam University

📖 Academic Activities

• Reviewer
  • Reviewer, IEEE Robotics and Automation Letters (RA-L), 2025
• Workshop
  • Presenter, Human-inspired Computer Vision Workshop, 2025
  • Presenter, Memory and Vision Workshop, 2025

📖 Language Skills

• English
  • 2026.03, TOEIC 910