Kuang-Yu Jeng
軟體工程師 at 聯發科
Taiwan
Summary
👤
Senior🎓
Top SchoolKuang-Yu Jeng is a software engineer based in Taiwan with 10 years of experience specializing in deep learning, image processing, and computer vision. At MediaTek he built production-ready systems that automate image quality assessment and chart/pose detection, including a Kubernetes-based AI service to improve robustness and accessibility. He engineered a deep-learning hand-eye calibration tool that cut robotic setup time by 90% and a novel pose estimator combining DL with epipolar geometry that raised automation rates to 98%. His work on objective image and video quality metrics — from brightness/contrast and autofocus to Bokeh and motion blur via PSF estimation and 3D techniques — consistently aligns with human judgment and achieves high AUROC/mAP scores. Comfortable bridging research and production, he integrates photogrammetry and 3D Gaussian Splatting into practical pipelines to reduce subjective bias. Fluent in turning complex computer-vision theory into measurable, deployable solutions, he holds advanced studies in computer science from National Taiwan University.
10 years of coding experience
Computer Science, Computer Science at 國立中正大學碩士, 資工, 碩士, 資工 at 國立臺灣大學
繁體中文, English
Github Skills (29)
ipa9
apple-silicon8
keymapping7
decoder6
encoder6
swiftui5
png5
intel5
ios5
swift4
arm4
macos3
robotics2
ros2
graph1
Programming languages (5)
Github contributions (5)
peter0749/WGAN-GP-Anime-with-Auxiliary-Classifier
Apr 2018 - Nov 2018
Contributions:1 release, 63 commits, 155 pushes in 6 months
peter0749/Multimedia-Content-Analysis
Oct 2017 - Jul 2022
Contributions:22 pushes, 1 branch in 4 years 9 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.