Kelvin Chan
Senior Research Scientist at Google DeepMind
Seattle, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolKelvin Chan is a Staff Research Scientist at Google DeepMind in Seattle with seven years of research and engineering experience in generative and image/video modeling. He has driven key image-generation capabilities across the Gemini and Imagen families and recently contributed to Gemini Omni, blending research rigor with product-scale deployment. Kelvin’s open-source contributions include adding video super-resolution (BasicVSR) and Real-ESRGAN integrations to the widely used OpenMMLab mmagic toolbox, reflecting practical expertise in diffusion and restoration pipelines. He holds a PhD in Computer Science from Nanyang Technological University and dual bachelor’s degrees in Information Engineering and Mathematics, anchored by an MPhil in Applied Mathematics. Known for bridging novel architectures (optical-flow/SPyNet components, test-time ensembles) into reproducible training and demo workflows, he thrives where cutting-edge research meets production engineering. Colleagues would note his mix of deep theoretical grounding and hands-on implementation across both research and open-source ecosystems.
7 years of coding experience
2 years of employment as a software developer
Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at Nanyang Technological UniversityThe Chinese University of Hong Kong (CUHK)
English, Chinese, Chinese
Github Skills (11)
diffusion-models10
computer-vision10
pytorch10
machine-learning10
deeplearning-ai10
deep-learning10
generative-ai10
model-testing9
evaluation9
eval9
image-generation9
Programming languages (2)
Github contributions (5)
open-mmlab/mmagic
Apr 2021 - Apr 2022
OpenMMLab Multimodal Advanced, Generative, and Intelligent Creation Toolbox. Unlock the magic 🪄: Generative-AI (AIGC), easy-to-use APIs, awsome model zoo, diffusion models, for text-to-image generation, image/video restoration/enhancement, etc.
Role in this project:
ML Engineer
ML EngineerContributions:322 reviews, 126 commits, 200 PRs in 1 year
Contributions summary:Kelvin contributed to the development of video super-resolution (VSR) models, specifically focusing on incorporating BasicVSR and Real-ESRGAN architectures. This included implementing components like the SPyNet for optical flow estimation and integrating Real-ESRGAN model and its associated training configurations. The user also worked on adding training configurations, test-time ensemble strategies, and the integration of the models with the testing and demo functionalities.
mattingimage-synthesistext-to-imagevideo-super-resolutionvideo-restoration
ckkelvinchan/RealBasicVSR
Nov 2021 - Apr 2022
Official repository of "Investigating Tradeoffs in Real-World Video Super-Resolution"
Contributions:34 commits, 17 PRs, 30 pushes in 5 months
resolutioncomputer-graphicsvideo-super-resolutionsuper-resolutionsiggraph
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.