PG-ControlNet: A Physics-Guided ControlNet for Generative Spatially Varying Image Deblurring

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Motorcu, Hakki, Cetin, Mujdat
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915638045310976
author Motorcu, Hakki
Cetin, Mujdat
author_facet Motorcu, Hakki
Cetin, Mujdat
contents Spatially varying image deblurring remains a fundamentally ill-posed problem, especially when degradations arise from complex mixtures of motion and other forms of blur under significant noise. State-of-the-art learning-based approaches generally fall into two paradigms: model-based deep unrolling methods that enforce physical constraints by modeling the degradations, but often produce over-smoothed, artifact-laden textures, and generative models that achieve superior perceptual quality yet hallucinate details due to weak physical constraints. In this paper, we propose a novel framework that uniquely reconciles these paradigms by taming a powerful generative prior with explicit, dense physical constraints. Rather than oversimplifying the degradation field, we model it as a dense continuum of high-dimensional compressed kernels, ensuring that minute variations in motion and other degradation patterns are captured. We leverage this rich descriptor field to condition a ControlNet architecture, strongly guiding the diffusion sampling process. Extensive experiments demonstrate that our method effectively bridges the gap between physical accuracy and perceptual realism, outperforming state-of-the-art model-based methods as well as generative baselines in challenging, severely blurred scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2511_21043
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle PG-ControlNet: A Physics-Guided ControlNet for Generative Spatially Varying Image Deblurring
Motorcu, Hakki
Cetin, Mujdat
Computer Vision and Pattern Recognition
Spatially varying image deblurring remains a fundamentally ill-posed problem, especially when degradations arise from complex mixtures of motion and other forms of blur under significant noise. State-of-the-art learning-based approaches generally fall into two paradigms: model-based deep unrolling methods that enforce physical constraints by modeling the degradations, but often produce over-smoothed, artifact-laden textures, and generative models that achieve superior perceptual quality yet hallucinate details due to weak physical constraints. In this paper, we propose a novel framework that uniquely reconciles these paradigms by taming a powerful generative prior with explicit, dense physical constraints. Rather than oversimplifying the degradation field, we model it as a dense continuum of high-dimensional compressed kernels, ensuring that minute variations in motion and other degradation patterns are captured. We leverage this rich descriptor field to condition a ControlNet architecture, strongly guiding the diffusion sampling process. Extensive experiments demonstrate that our method effectively bridges the gap between physical accuracy and perceptual realism, outperforming state-of-the-art model-based methods as well as generative baselines in challenging, severely blurred scenarios.
title PG-ControlNet: A Physics-Guided ControlNet for Generative Spatially Varying Image Deblurring
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.21043