Ouyang Mingyu
National University of Singapore
Hiii 😀:)
I am currently a Master's student at the Department of Mathematics, National University of Singapore. I received my B.Sc in Spatial Informatics from Wuhan University in 2023. There I joined IIP Lab under the supervision of Prof. Zhenzhong Chen.
My research mainly focuses on computer vision, including low-level image singnal processing (compression, super-resolution, denoising, etc) and vision-embedded LLM agents. I am also experienced in efficient network design and building AI ISP algorithms for camera/phone in-built modules.
📣Latest Updates:
▪ [05/2024]: First authored paper accepted by IEEE TIP (IF=10.8)! Check out the paper/news.
▪ [03/2024]: Collab on one paper to appear in CVPR 24'.
▪ [12/2023]: Start an internship at Sensetime, Singapore.
▪ [08/2023]: Great honored to be a part of ShowLab, NUS. Cheers!
🎓Education
M.Sc in Data Science and Machine Learning National University of Singapore Aug, 2023 - present
B.Sc in Spatial Informatics and Digitalized Technology Wuhan University 2019 - 2023
▪ GPA: 87.54/100, 2021-22 Annual Department Assessment Ranking: 3/52
▪ First Class Undergraduate Scholarship 2022' (top 5%)
Certificated Program Imperial College London 2022
▪ Best Final Project (1 of 16 groups) [Certificate]
🧑💻 Internship
I spent some wonderful and meaningful time as an intern with:
Sensetime, Singapore
CV Research Intern
Dec. 2023 - Present
Tencent Technology, Beijing
ML Engineer Intern, Tencent Map
Mar. 2022 – Oct. 2022
🍀 Publications
[IEEE TIP] JPEG Quantized Coefficient Recovery via DCT Domain Spatial-Frequential Transformer [Link]
Mingyu Ouyang, Zhenzhong Chen†
• A dual-branch Transformer for JPEG artifact removal in the DCT frequency domain (DCTransformer), with an efficient yet effective learning scheme utilizing quantization priors.
• Paper submitted to IEEE TIP 07/2023, revised 02/2024, accepted 05/2024.
[CVPR 24'] ASSISTGUI: Task-Oriented Desktop Graphical User Interface Automation [Link]
D. Gao, L. Ji, Z. Bai, M. Ouyang, P. Li, D. Mao, Q. Wu, W. Zhang, P. Wang, X. Guo, H. Wang, L. Zhou, Mike Z. Shou
• A novel benchmark for multi-modality assistant agents, namely AssistGUI, capable of manipulating the mouse and keyboard on the system OS in response to user-requested tasks.
🐾 Extracurricular
Wuhan University Football Team
• Captain (2021-2023). Champion of the 16th Hubei Province Games (2022), and 2nd place at the Province University Tournament (2020)
Website Stat: