Korea University
Aug 2024 – PresentPh.D. Candidate, Department of Electrical Engineering
Advisor: Prof. Seung-Won Jung · Deep Image Processing Lab
Ph.D. Candidate, Department of Electrical Engineering
Deep Image Processing Lab, Korea University
I am an AI research engineer specializing in computer vision, with a primary focus on low-level vision and image restoration. My current research addresses image degradation challenges, including deblurring, low-light enhancement, and display-induced distortions. My broader research experience also spans video understanding, including temporal action localization and pedestrian attribute recognition.
Ph.D. Candidate, Department of Electrical Engineering
Advisor: Prof. Seung-Won Jung · Deep Image Processing Lab
B.S. & M.S., Department of Software
Advisor: Prof. Jungchan Cho · Visual AI Lab
Research topic: video shot boundary detection
Research topic: pedestrian state understanding in real-world CCTV domains
Android TV application programming
Intrinsic Decomposition-Based Curriculum Learning for Low-Light Image Enhancement
IEEE SPL 2025IEEE Signal Processing Letters, vol. 32, pp. 4019–4023, 2025
Background-Aware Robust Context Learning for Weakly-Supervised Temporal Action Localization
IEEE Access 2022IEEE Access, vol. 10, pp. 65315–65325, 2022
DnSAD: Defocus and Spatially Adaptive Deblurring for Mura Compensation in Display Panels
JSID 2026Journal of the Society for Information Display, vol. 34, no. 3, pp. 99–108, 2026
Underexposed Image Enhancement for Mura Compensation in Display Panels
JSID 2025Journal of the Society for Information Display, vol. 33, no. 9, pp. 955–963, 2025
Improving Visual Pedestrian Attributes Discernment With Textual Reconstruction
IEEE Access 2024IEEE Access, vol. 12, pp. 164178–164189, 2024
Datasets and Research Trends for Local Motion Deblurring
IPIU 2026Workshop on Image Processing and Image Understanding (IPIU), 2026
A Study on Limitations of Temporal Video Localization Based on Point-Level Annotation and an Improvement Method Through Point Distribution Modeling
KINGC 2023The Journal of Korean Institute of Next Generation Computing, 2023
Improvements on Real-World Domain and Person Understanding via Optimal Combination of Training Dataset and Model
KIISE 2022Korean Institute of Information Scientists and Engineers (KIISE), 2022
Funded by LG Display
Funded by SK Telecom
Funded by LG Display
Funded by ETRI (Electronics and Telecommunications Research Institute)