Sohyun Lee

SohyunLee.png

I am a graduate student in the Graduate School of Artificial Intelligence integrated M.S. & Ph.D. program at POSTECH. I am a member of the Computer Vision Lab at POSTECH, advised by Prof. Suha Kwak. Previously, I completed my B.S. in Mechanical Engineering at POSTECH.

My research interests lie in computer vision and deep learning. I’ve worked on the robust recognition in adverse visual conditions, domain adaptation, and generalization.

If you are interested in my research projects, please feel free to contact me.

News

Jul 2, 2024 📝 Our paper on robust segmentation under multiple adverse conditions is accepted to ECCV 2024.
Sep 22, 2023 📝 Our paper on active learning is accepted to NeurIPS 2023.
Jun 10, 2023 :trophy: I won the POSTECHIAN Fellowship.
Feb 28, 2023 📝 Our paper on low-light image recognition is accepted to CVPR 2023.
Feb 27, 2023 :trophy: FIFO won the grand prize at BK21 Best Paper Award from POSTECH GSAI.
Feb 8, 2023 :trophy: I won the excellence award at 3rd POSTECH Research Performance Contest.
Nov 7, 2022 :trophy: Three papers got honored to be the winners at the Qualcomm Innovation Fellowship 2022.
Jul 4, 2022 📝 A paper on active domain adaptation is accepted to ECCV 2022.
Jun 21, 2022 :trophy: Our paper on foggy scene segmentation is nominated as a best paper finalist in CVPR 2022.
Mar 3, 2022 📝 Two papers (including one best paper finalist) are accepted to CVPR 2022.

Experience

Mar, 2024 - May, 2024 Tübingen AI Center, University of Tübingen, Tübingen, Germany
Visiting Research Student
  • Host: Prof. Seong Joon Oh
Sep, 2020 - Present Computer Vision Lab, POSTECH, Pohang, South Korea
Research and Teaching Assistant

Education

Sep, 2020 - Present Pohang University of Science and Technology (POSTECH), Pohang, South Korea
Integrated M.S. & Ph.D. Student in Artificial Intelligence
Advisor: Prof. Suha Kwak.
Mar, 2015 - Aug, 2020 Pohang University of Science and Technology (POSTECH), Pohang, South Korea
B.S in Mechanical Engineering
Advisor: Prof. Junsuk Rho.

Publications

  1. FREST: Feature RESToration for Semantic Segmentation under Multiple Adverse Conditions
    Sohyun Lee, Namyup Kim, Sungyeon Kim, and Suha Kwak
    European Conference on Computer Vision (ECCV), 2024
  2. Active Learning for Semantic Segmentation with Multi-class Label Query
    Sehyun Hwang,  Sohyun Lee, Hoyoung Kim, Minhyeon Oh, Jungseul Ok, and Suha Kwak
    Conference on Neural Information Processing Systems (NeurIPS), 2023
  3. Human Pose Estimation in Extremely Low-Light Conditions
    Sohyun Lee*, Jaesung Rim*, Boseung Jeong, Geonu Kim, ByungJu Woo, Haechan Lee, Sunghyun Cho, and Suha Kwak (*equal contribution)
    IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023
  4. Combating Label Distribution Shift for Active Domain Adaptation
    Sehyun Hwang,  Sohyun Lee, Sungyeon Kim, Jungseul Ok, and Suha Kwak
    European Conference on Computer Vision (ECCV), 2022
    Qualcomm Innovation Fellowship Winner
  5. FIFO: Learning Fog-invariant Features for Foggy Scene Segmentation
    Sohyun Lee, Taeyoung Son, and Suha Kwak
    IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022
    (Best Paper Finalist, Oral Presentation)
    Invited paper talk at V4AS Workshop @ CVPR 2022
    Qualcomm Innovation Fellowship Winner
  6. Style Neophile: Constantly Seeking Novel Styles for Domain Generalization
    Juwon Kang,  Sohyun Lee, Namyup Kim, and Suha Kwak
    IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022
    Qualcomm Innovation Fellowship Winner

Professional Services

Organizer
  • Women in Computer Vision Workshop (WiCV) at ACCV 2024
Journal Reviewer
  • IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
Conference Reviewer
  • ICLR, NeurIPS, ICCV, CVPR, ECCV, AAAI, WACV, ACCV

Honors and Awards

  • POSTECHIAN fellowship awards, POSTECH, 2023
  • POSTECH Research Performance Contest (Excellence Award), POSTECH, 2023
  • BK21 Best Paper Award (Grand Prize), POSTECH GSAI, 2023
  • Qualcomm Innovation Fellowship Winner (3 times), Qualcomm Korea Corp., 2022
    • FIFO: Learning Fog-invariant Features for Foggy Scene Segmentation (CVPR 2022, Best Paper Finalist)
    • Style Neophile: Constantly Seeking Novel Styles for Domain Generalization (CVPR 2022)
    • Combating Label Distribution Shift for Active Domain Adaptation (ECCV 2022)
  • CVPR Best Paper Finalist, CVPR, 2022
    • Awarded to Top 0.4% (33 of 8161 papers)
    • FIFO: Learning Fog-invariant Features for Foggy Scene Segmentation
  • IPIU Best Paper Award (Gold Prize), IPIU, 2022
  • POSTECH Creative Self-Research Scholarship, POSTECH GSAI, 2020