Donghun Ryou
I am currently at the SNU Computer Vision Lab, advised by Prof. Bohyung Han. I am broadly interested in computer vision & deep learning. I focus on applying deep learning techniques to real-world scenarios, emphasizing robustness and efficiency.
NEWS
Apr 2025
Our paper Learning to Translate Noise for Robust Image Denoising will be presented at DG-EBF Workshop @ CVPR 2025.
Apr 2025
Our team, SNUCV (Donghun Ryou, Inju Ha, Bohyung Han), won 1st place in the Image Super-Resolution (x4) perceptual track and 2nd place in the Image Denoising at the NTIRE Workshop and Challenges @ CVPR 2025.
Feb 2024
Our paper Robust Image Denoising through Adversarial Frequency Mixup is accepted at CVPR 2024
Education
- Seoul National University
Ph.D. Student, Computer Vision Lab
Sep, 2022 ~ present
- POSTECH
B.S, Electrical Engineering
Feb, 2018 ~ Aug, 2022
- RWTH Aachen University
Exchange Student
Sep, 2021 ~ Feb, 2022
Publication
Learning to Translate Noise for Robust Image Denoising [paper] [code] [project page]
Inju Ha*, Donghun Ryou*, Seonguk Seo, Bohyung Han
* indicates equal contribution
ArXiv
Multi-stage adaptive rank statistic pruning for lightweight human 3D mesh recovery model [paper]
Donghun Ryou, Kim Youwang and Tae-Hyun Oh
The Visual Computer Journal, Springer (TVCJ) 2023
Projects
Web application development
- Web shopping application based short-form video
- Used programming language : React, TypeScript
- website : digkey.shop
Camera software system design & development
- 100day Camera creation project
- Cheap and vintage aesthetic camera
- instagram : luke__p100
