Wanli Zhong
Student at OpenCV China
SUSTech
Summary
🤩
RockstarWanli Zhong is a master’s student at SUSTech with six years of software engineering experience and a growing focus on computer vision and deep learning. As an active contributor and member of OpenCV China, he has made substantive backend improvements to OpenCV’s DNN module—refactoring the Slice layer, broadening ONNX compatibility, adding robust unit tests, and extending CUDA support for better GPU performance. His work blends ML engineering, QA, and test automation, frequently generating and validating ONNX models to cover edge cases like broadcasting and negative axes. Wanli combines research-oriented rigor with practical engineering, demonstrating an uncommon depth in both low-level library internals and cross-version model compatibility.
6 years of coding experience
Github Skills (12)
computer-vision10
pytorch10
opencv10
c-language10
deep-learning10
onnx10
cprogramming-language10
python10
image-processing10
test-automation10
cuda9
numpy8
Programming languages (11)
Github contributions (5)
opencv/opencv_extra
Jun 2022 - Dec 2022
OpenCV extra data
Role in this project:
ML Engineer & QA Engineer / Test Automation Engineer
ML Engineer & QA Engineer / Test Automation EngineerContributions:6 reviews, 6 commits, 23 PRs in 5 months
Contributions summary:Wanli contributed significantly to the testing of ONNX models within the OpenCV ecosystem, primarily focusing on the `dnn` module. Their commits involve adding tests and generating ONNX models for various scenarios, including slice, unsqueeze, and MatMul operations, often related to ONNX compatibility and functionalities. The contributions reveal an emphasis on ensuring the proper behavior and coverage of different ONNX model configurations, including tests for edge cases and broadcast behavior. This work strongly suggests a QA/Test Automation and ML engineering focus on the project.
opencv/opencv
May 2022 - Dec 2022
Open Source Computer Vision Library
Role in this project:Back-end Developer
Back-end DeveloperContributions:79 reviews, 11 commits, 40 PRs in 7 months
Contributions summary:Wanli primarily focused on enhancing and refactoring the Slice layer within the OpenCV DNN module, a core component for deep learning inference. They implemented improvements to handle different input scenarios, including constant inputs, negative axes, and broadcasting, which expanded the layer's functionality. The contributions included adding unit tests to validate the changes and ensuring compatibility across different ONNX versions. Furthermore, the user extended support for CUDA to specific operators in the NaryEltwise layer, improving performance on NVIDIA GPUs.
computer-visionc-plus-plusdeep-learningimage-processing
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.