Hoseong Lee
Principal Deep Learning Researcher at Cognex Corporation
Seoul, South Korea
Summary
🤩
Rockstar🎓
Top SchoolHoseong Lee is a Principal Deep Learning Researcher based in Seoul with 8 years of hands-on experience building industrial computer vision and anomaly detection systems. At Cognex (and previously SUALAB) he has led research spanning unsupervised anomaly detection, object detection, segmentation, barcode image restoration, edge-oriented network compression, and defect generation. His work emphasizes production-ready solutions—label-noise detection, hyperparameter optimization, and image matching—to bridge research prototypes and deployed inspection systems. He holds bachelor’s and master’s degrees in Electrical and Information Engineering from Seoul National University. An active practitioner, he shares code and insights on GitHub and his technical blog, signaling a commitment to reproducible research and community knowledge sharing. Colleagues describe him as a researcher who balances deep algorithmic rigor with practical constraints of edge deployment.
8 years of coding experience
5 years of employment as a software developer
석사, 전기정보공학, 석사, 전기정보공학 at Seoul National University
Github Skills (20)
anomaly-detection10
classification10
machine-learning10
deep-learning10
deep-neural-networks10
object-detection10
pytorch9
statistics9
paper9
jekyll9
jekyll-theme8
style-transfer8
visualization8
image-classification7
semantic-segmentation7
Programming languages (2)
Github contributions (5)
hoya012/semantic-segmentation-tutorial-pytorch
Jul 2020 - Nov 2021
A simple PyTorch codebase for semantic segmentation using Cityscapes.
Contributions:16 commits, 8 PRs, 13 pushes in 1 year 3 months
pytorchsemantic-segmentation
hoya012/carrier-of-tricks-for-classification-pytorch
Jun 2020 - Nov 2021
carrier of tricks for image classification tutorials using pytorch.
Contributions:60 commits, 2 PRs, 44 pushes in 1 year 5 months
image-classificationpytorchdeep-learningdeep-neural-networksclassification-pytorch
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.