Syed Zamir is an AI/ML research scientist with a PhD in image processing and over a decade of experience building and deploying state-of-the-art models across computer vision, NLP, speech and multimodal systems for sectors from healthcare to remote sensing. He has published 45+ papers with 15,000+ citations, holds four patents, and maintains popular open-source projects—his Restormer contributions tie directly to a CVPR-oral, SOTA image-restoration repo that has helped accelerate practical adoption. At research labs including Inception Institute of Artificial Intelligence, Inria and Microsoft, he has led work on vision-language models, efficient diffusion and restoration algorithms, RAG systems, and synthetic-data pipelines with an eye for resource efficiency and real-world deployment. Based in Abu Dhabi, he combines deep academic rigor and psychophysical evaluation experience with hands-on ML engineering, often optimizing code for large-image inference and CPU-friendly demos that broaden accessibility.
7 years of coding experience
12 years of employment as a software developer
Doctor of Philosophy (Ph.D.), Image Processing, Doctor of Philosophy (Ph.D.), Image Processing at Universitat Pompeu Fabra
Bachelor of Science (B.Sc.), Computer Engineering, Distinction, Bachelor of Science (B.Sc.), Computer Engineering, Distinction at COMSATS Institute of Information Technology
Master of Science (M.Sc.), Digital Signal Processing, Distinction, Master of Science (M.Sc.), Digital Signal Processing, Distinction at Queen Mary, U. of London
[CVPR 2022--Oral] Restormer: Efficient Transformer for High-Resolution Image Restoration. SOTA for motion deblurring, image deraining, denoising (Gaussian/real data), and defocus deblurring.
Role in this project:
ML Engineer
Contributions:1 release, 28 commits, 28 pushes in 7 months
Contributions summary:Syed contributed to the image restoration project by addressing code issues related to image saving, fixing paper references, and adding a demo file. They also updated the evaluation scripts for performance assessment, incorporated parallel computing for faster evaluation, and enabled the demo to run on CPU if a GPU is unavailable. Furthermore, the user added an option for testing on large images by implementing a tiling strategy.
Official repository for "Learning Enriched Features for Real Image Restoration and Enhancement" (ECCV 2020). SOTA results for image denoising, super-resolution, and image enhancement.
Contributions:17 commits, 13 pushes, 12 comments in 2 years 1 month
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.