Zhe Chen
Professor at Nanjing University of Aeronautics and Astronautics
Nanjing City, Jiangsu, China
Summary
🤩
Rockstar🎓
Top SchoolZhe Chen is a Professor at Nanjing University of Aeronautics and Astronautics with a PhD from INSA Toulouse and over a decade of academic experience focused on software engineering, programming languages, and formal methods for building highly dependable systems. He has authored 60+ peer-reviewed publications in top venues (TSE, TOSEM, OOPSLA, ISSTA) and won an ACM SIGSOFT Distinguished Paper Award in 2019, reflecting a strong research impact. Beyond theory, he contributes to cutting-edge computer vision and transformer projects on GitHub—working on InternImage, PVTv2, and ViT-Adapter implementations—demonstrating practical ML engineering skills such as backbone modification, segmentation adaptation, and FLOPs-aware optimizations. A member of ACM and senior member of CCF serving on its formal methods technical committee, he bridges rigorous formal techniques with applied ML and systems work, often bringing efficiency-minded changes to large-scale models.
9 years of coding experience
14 years of employment as a software developer
Master, Computer Science, Master, Computer Science at Central South UniversityPh.D., Computer Science, Ph.D., Computer Science at Institut national des Sciences appliquées de ToulousePh.D., Computer Science, Ph.D., Computer Science at Shanghai Jiao Tong University
Chinese, English, French
Github Skills (20)
pytorch10
semantic-segmentation10
machine-learning10
image-segmentation10
deep-learning10
segmentation10
trainings10
vision-transformer10
model-optimization10
adapter10
object-detection10
computer-vision10
transformer10
backbone10
modeling10
Programming languages (6)
Github contributions (5)
OpenGVLab/InternImage
Nov 2022 - Mar 2023
[CVPR 2023 Highlight] InternImage: Exploring Large-Scale Vision Foundation Models with Deformable Convolutions
Role in this project:
ML Engineer
ML EngineerContributions:3 releases, 2 reviews, 17 commits in 4 months
Contributions summary:Zhe primarily focused on releasing detection and segmentation models within the InternImage framework, which is a vision foundation model utilizing deformable convolutions. Their contributions involved modifications to the backbone architecture, specifically the InternImage model, within the context of object detection and segmentation tasks. They also added scripts for calculating FLOPs, demonstrating an emphasis on model efficiency and evaluation. The user also added support for cityscapes dataset.
backbonedeformable-convolutionfoundation-modelobject-detectionsemantic-segmentation
whai362/PVT
Jun 2021 - Feb 2022
Role in this project:Back-end Developer
Back-end DeveloperContributions:31 commits, 10 PRs, 25 pushes in 8 months
Contributions summary:Zhe's primary contribution involves updating and modifying the `pvt_v2.py` file, indicating work on the core implementation of the Pyramid Vision Transformer V2 model. These changes involve adjustments to the MLP and Attention modules and likely aim to improve the model's performance or add new features. The presence of the `pvt_v2.py` file within the `whai362/pvt` repository and commits related to the model's architecture highlight the user's focus on the back-end aspects of the project.
backbonedetectionpvttransformersegmentation
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.