Abstract:Blind image deblurring (BID) has been extensively studied in computer vision and adjacent fields. Modern methods for BID can be grouped into two categories: single-instance methods that deal with individual instances using statistical inference and numerical optimization, and data-driven methods that train deep-learning models to deblur future instances directly. Data-driven methods can be free from the difficulty in deriving accurate blur models, but are fundamentally limited by the diversity and quality of the training data -- collecting sufficiently expressive and realistic training data is a standing challenge. In this paper, we focus on single-instance methods that remain competitive and indispensable. However, most such methods do not prescribe how to deal with unknown kernel size and substantial noise, precluding practical deployment. Indeed, we show that several state-of-the-art (SOTA) single-instance methods are unstable when the kernel size is overspecified, and/or the noise level is high. On the positive side, we propose a practical BID method that is stable against both, the first of its kind. Our method builds on the recent ideas of solving inverse problems by integrating the physical models and structured deep neural networks, without extra training data. We introduce several crucial modifications to achieve the desired stability. Extensive empirical tests on standard synthetic datasets, as well as real-world NTIRE2020 and RealBlur datasets, show the superior effectiveness and practicality of our BID method compared to SOTA single-instance as well as data-driven methods. The code of our method is available at: \url{<a class="link-external link-https" href="https://github.com/sun-umn/Blind-Image-Deblurring" rel="external noopener nofollow">this https URL</a>}.

Fast ℓ0-Norm-based Single Image Blind Deblurring: A Comparative Study.

Blind Deconvolution Using an Improved L0 Sparse Representation

Natural Image Deblurring Based on L0-regularization and Kernel Shape Optimization

Text Image Deconvolution Using Re-blurry Process And <tex>$L_{0}$</tex> Norm

Motion Image Deblurring Based on L0 Norms Regularization Term

Surface-Aware Blind Image Deblurring

Blind Deblurring Based on an L_(1/2)/L_2 Regularization

Blind Image De-blurring Based on Sparse Prior and Relative Total Variation

Deblurring Natural Image Using Super-Gaussian Fields

Blind Image Motion Deblurring with L-0-Regularized Priors

Single Image Blind Deblurring Based on Salient Edge-Structures and Elastic-Net Regularization

Blind natural image deblurring with edge preservation based on L0-regularized gradient prior

Blind image deblurring using kernel error for p -shrinkage operator optimization model

Fast -Regularized Kernel Estimation for Robust Motion Deblurring

Blind Deconvolution with Scale Ambiguity

Blind Image Deblurring with Unknown Kernel Size and Substantial Noise

Blind Image Deblurring Via Content Adaptive Method

A Simple Local Minimal Intensity Prior and an Improved Algorithm for Blind Image Deblurring

Non-blind Deconvolution with ℓ 1 -Norm of High-Frequency Fidelity

Blind and Non-Blind Deconvolution-Based Image Deblurring Techniques for Blurred and Noisy Image

Blind Image Deblurring with Local Maximum Gradient Prior.