Yunjey Choi
Tech Lead & Research Scientist at NAVER Corp
South Korea
Summary
🤩
Rockstar🎓
Top SchoolYunjey Choi is a tech lead and research scientist with 10 years of experience building and shipping generative vision models at NAVER AI Lab, where he now leads the Generation Team focused on visual imagination. His research contributions include core roles in StarGAN and StarGAN v2—CVPR papers with widespread academic and community impact—and practical work on evaluation metrics presented at ICML 2020. He bridges research and engineering, authoring reproducible PyTorch and TensorFlow implementations, dataset pipelines, evaluation tooling (LPIPS, FID), and training infrastructure that power both papers and open-source users. Known for turning cutting-edge ideas into well-documented code, he has influenced models that together have thousands of citations and GitHub stars. Trained at 고려대학교 with top academic standing, he combines strong theoretical grounding with hands-on ML engineering and a knack for making complex generative systems usable in practice.
10 years of coding experience
2 years of employment as a software developer
Bachelor's degree, Computer Science, Major GPA: 4.38/4.5, Bachelor's degree, Computer Science, Major GPA: 4.38/4.5 at 고려대학교
Korean, English
Github Skills (22)
pytorch10
python10
evaluation10
dcgan10
machine-learning10
cgan10
data-loading10
dqn10
deeplearning-ai10
unsupervised-learning10
deep-learning10
image-generation10
tensorflow10
neural-network10
generative-model10
Programming languages (4)
Github contributions (5)
clovaai/stargan-v2
Dec 2019 - Jun 2022
StarGAN v2 - Official PyTorch Implementation (CVPR 2020)
Role in this project:
ML Engineer
ML EngineerContributions:31 commits, 3 PRs, 23 pushes in 2 years 7 months
Contributions summary:Yunjey contributed to the development and evaluation of a machine learning model for image-to-image translation, specifically within the context of the StarGAN v2 project. Their work includes adding scripts for downloading and preparing datasets, implementing crucial evaluation metrics like LPIPS and FID, and creating the necessary infrastructure for model assessment. Additionally, the user updated the project's dependencies, including adding model checkpoints and download links, and creating core Python scripts for model execution and testing.
pytorchstarganimage-to-image-translationgenerative-modelsstargan-v2
yunjey/stargan
Nov 2017 - Apr 2020
StarGAN - Official PyTorch Implementation (CVPR 2018)
Role in this project:ML Engineer
ML EngineerContributions:45 commits, 29 pushes, 53 comments in 2 years 5 months
Contributions summary:Yunjey's commits primarily focused on adding and modifying core components related to data loading, logging, and model training within the StarGAN project. This included implementing a `CelebDataset` class for handling image data and labels, creating a `Logger` class for tracking training progress, and adding a `main.py` file to orchestrate the training process. These changes suggest a focus on setting up and executing the machine learning pipeline for image-to-image translation tasks.
pytorchstarganimage-to-image-translationgenerative-modelscvpr2018
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.