Controllable blind deblurring with diffusion models
2026-08-24 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors focus on fixing blurry photos taken with professional cameras when the exact blur pattern is unknown. They created a method called SuperSharpen that uses a special type of AI called diffusion models to sharpen images while letting users control how strong the sharpening is. They tested two ways to set up the AI and found that fully retraining it works better to keep the photo realistic without making up fake details. Their method works well on both simulated and real blurry photos and improves how clear the images look.
blind deblurringdiffusion modelsimage restorationisotropic blurControlNetfinetuninginverse problemsphotorealismperceptual qualitydegradation kernel
Authors
Imane Si Salah, Emile Cribelier, Thomas Veit, Wolf Hauser, Arthur Leclaire
Abstract
Image acquisition with a camera involves several degradations due to the optical system, sensor, or low-level processing steps. We address blind deblurring in professional photography: we aim to invert unknown isotropic blur without knowledge of the degradation kernel.For such inverse problems,where some high-frequency information is lost, it is challenging to use generative models to produce details that are both photo-realistic and faithful to the input. We propose SuperSharpen, a diffusion-based blind deblurring method offering explicit control over restoration strength through a blur measure. We compare two conditioning strategies: a ControlNet-style adapter on a frozen backbone, and full finetuning of the diffusion prior. Our experiments show that finetuning achieves better fidelity with fewer hallucinated details. We validate our approach on synthetic and real-world blur, demonstrating improved perceptual quality and controllable restoration strength.