Qing Jiang
Research Intern at International Digital Economy Academy 粤港澳大湾区数字经济研究院
Guangzhou City, Guangdong Province, China
Summary
🤩
Rockstar🎓
Top SchoolQing Jiang is a first-year Ph.D. candidate at South China University of Technology specializing in object perception and understanding, currently interning at the International Digital Economy Academy after prior research work at Shanghai AI Laboratory. With five years of experience in research and engineering, Qing contributes to open-source AI tooling—notably improving OpenMMLab's MMOCR by refactoring ResNets, adding JSONL support, and fixing training-loss bugs to enhance model stability. Their work bridges model architecture, data-format interoperability, and practical training robustness, reflecting both theoretical depth and production-minded implementation. Based in Guangzhou, Qing is committed to sustainable AI development through open source and long-term collaborative research.
5 years of coding experience
博士, 信息与通信工程, 博士, 信息与通信工程 at 华南理工大学
Github Skills (12)
ocr10
image-recognition10
machine-learning10
text-recognition10
ocra10
python10
text-detection9
deep-learning9
pytorch9
tensorflow8
computer-vision7
object-detection7
Programming languages (2)
Github contributions (5)
open-mmlab/mmocr
Mar 2022 - Dec 2022
OpenMMLab Text Detection, Recognition and Understanding Toolbox
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:142 reviews, 87 commits, 86 PRs in 9 months
Contributions summary:Qing implemented support for JSONL format in the recognition converter, enhancing the tool's flexibility for handling different annotation formats. They also refactored ResNets, a crucial component of many image recognition models, suggesting a focus on model architecture and optimization. Furthermore, the user addressed several bugs related to loss functions, including ignore_index in SARLoss and inplace operation errors, indicating involvement in refining the model training process and ensuring its stability.
panettextsnakeopenmmlabcrnnunderstanding
Mountchicken/2021-Resolution
May 2021 - Jan 2022
Contributions:236 pushes, 1 branch in 8 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.