Abdullah Al Shafi
Computer Vision · Generative Models · Medical Image Analysis
Computer Vision · Generative Models · Medical Image Analysis
I recently completed my B.Sc. in Computer Science and Engineering at Khulna University of Engineering & Technology (KUET), Bangladesh, and am keen to keep working on research in computer vision.
So far, the work has spanned generative modelling, image restoration and medical image analysis: how far diffusion models can be compressed before their samples suffer, how real-world degradations differ from the synthetic ones restoration methods learn from, and how medical imaging models can let related tasks inform one another and recognise when their own predictions are unreliable. Beneath all three is a concern with whether our evaluations measure what we believe they measure.
At KUET, I completed my undergraduate thesis under the supervision of Prof. Kazi Saeed Alam, worked with Prof. Sk. Md. Masudul Ahsan on my third-year system development project, and conducted research in the lab of Prof. Sk Imran Hossain. With him and Prof. Engelbert Mephu Nguifo of Université Clermont Auvergne, I have worked on compact diffusion models and on multi-task learning for medical image analysis.
| Sep 2026 | presented the guidance-aware quantization paper as an oral at IEEE HPEC |
|---|---|
| Sep 2026 | OneHealth+ published in SoftwareX |
| Jul 2026 | graduated with a B.Sc. in Computer Science and Engineering from KUET |
| Jul 2026 | guidance-aware quantization paper accepted as an oral at IEEE HPEC |
| Jun 2026 | GoRent published in SoftwareX; preprints out for DASH and BTI-Net |
| May 2026 | released iPhoneBlur: 7,400 stratified blur–sharp pairs, with fine-tuned weights |
Release artifacts for the papers above are linked from each entry. Beyond those:
| 2022–26 | B.Sc. in Computer Science and Engineering, Khulna University of Engineering & Technology |
|---|
| 3rd year | Dean's List Award, Faculty of Electrical and Electronic Engineering, KUET, for academic excellence |
|---|