Scholar
Sohyun Lee
Google Scholar ID: S_IGo4UAAAAJ
POSTECH
Citations & Impact
Citations
313
H-index
5
i10-index
5
Publications
10
Co-authors
16
list available
Publications
Resume (English only)
Academic Achievements
- 1. Sep 19, 2025, Paper 'GaRA-SAM: Robustifying Segment Anything Model with Gated-Rank Adaptation' accepted to NeurIPS 2025.
- 2. Jul 2, 2024, Paper on robust segmentation under multiple adverse conditions accepted to ECCV 2024.
- 3. Sep 22, 2023, Paper on active learning accepted to NeurIPS 2023.
- 4. Jun 10, 2023, Won the POSTECHIAN Fellowship.
- 5. Feb 28, 2023, Paper on low-light image recognition accepted to CVPR 2023.
- 6. Feb 27, 2023, FIFO won the grand prize at BK21 Best Paper Award from POSTECH GSAI.
- 7. Feb 8, 2023, Won the excellence award at 3rd POSTECH Research Performance Contest.
- 8. Nov 7, 2022, Three papers honored as winners at the Qualcomm Innovation Fellowship 2022.
- 9. Jul 4, 2022, A paper on active domain adaptation accepted to ECCV 2022.
- 10. Jun 21, 2022, Paper on foggy scene segmentation nominated as a best paper finalist in CVPR 2022.
- 11. Mar 3, 2022, Two papers (including one best paper finalist) accepted to CVPR 2022.
Research Experience
- 1. May, 2025 - Aug., 2025, ETH Zürich, Visiting Research Student, Host: Dr. Christos Sakaridis, Prof. Konrad Schindler.
- 2. Jan, 2025 - Present, Google Zürich, Research Collaboration, Working with Lukas Hoyer.
- 3. Mar, 2024 - May, 2024, Tübingen AI Center, University of Tübingen, Visiting Research Student, Host: Prof. Seong Joon Oh.
- 4. Sep, 2020 - Present, Computer Vision Lab, POSTECH, Research and Teaching Assistant.
Education
- 1. Sep, 2020 - Present, Pohang University of Science and Technology (POSTECH), Integrated M.S. & Ph.D. Student in Artificial Intelligence, Advisor: Prof. Suha Kwak.
- 2. Mar, 2015 - Aug, 2020, Pohang University of Science and Technology (POSTECH), B.S in Mechanical Engineering, Advisor: Prof. Junsuk Rho.
Background
- Research Interests: Computer Vision and Deep Learning; Work on robust recognition in adverse visual conditions, domain adaptation, and generalization.
