Yuantao Feng
Software Engineer at OpenCV
Shenzhen, Guangdong Province, China
Summary
🤩
Rockstar🎓
Top SchoolYuantao Feng is a software engineer with 10 years’ experience specializing in machine learning, computer vision, and deep learning inference frameworks, currently based in Shenzhen. As a core OpenCV developer, he has improved the DNN module across CPU architectures (x64, ARM, RISC-V, LoongSon), managed the OpenCV Model Zoo, and integrated backends like CANN to make model deployment more robust. His open-source work includes training and exporting face-detection models to ONNX and contributing operator support and conformance test models to the widely used opencv/opencv and opencv/opencv_extra repositories. He blends research-grade model training with production-oriented engineering—building CI systems, fixing edge-case operator bugs, and enabling dynamic input support. Now building his own "AI farm," he combines deep technical breadth with hands-on system-level optimizations that accelerate real-world ML inference.
10 years of coding experience
Master's degree, Computer Science, Master's degree, Computer Science at Shenzhen University
English, Chinese, Japanese
Stackoverflow
191reputation
44kreached
8answers
5questions
Github Skills (20)
face-detection10
pytorch10
opencv10
c-language10
python10
machine-learning10
onnx10
deep-learning10
trainings10
model-testing10
computer-vision10
cprogramming-language10
modeling10
nvidia6
neural-network6
Programming languages (15)
Github contributions (5)
ShiqiYu/libfacedetection.train
Mar 2020 - Feb 2022
The training program for libfacedetection for face detection and 5-landmark detection.
Role in this project:
ML Engineer
ML EngineerContributions:16 reviews, 30 commits, 20 PRs in 1 year 11 months
Contributions summary:Yuantao primarily contributed to training and testing a face detection and landmark detection model. Their work involved adding datasets for the WIDER Face dataset, implementing a testing script, and exporting the model to ONNX format. They also refined the model, updated its architecture, fixed bugs, and added support for dynamic input sizes and evaluation.
face-detection
opencv/opencv_extra
Oct 2021 - Dec 2022
OpenCV extra data
Role in this project:ML Engineer
ML EngineerContributions:10 reviews, 11 commits, 41 PRs in 1 year 1 month
Contributions summary:Yuantao primarily contributed to the repository by adding and updating test data and models related to ONNX, specifically for the DNN (Deep Neural Network) module. Their work includes incorporating new models for various ONNX operators, generating model files using Python scripts and ONNXScript, and updating existing models, especially for conformance tests. They focused on features like quantized convolutions, GEMM (General Matrix Multiplication) and Expand operators, thereby contributing to the development and testing of the DNN capabilities within the project.
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.