Leveraging Classic Deconvolution and Feature Extraction in Zero-Shot Image Restoration

TL;DR

This paper introduces a novel zero-shot non-blind deconvolution approach that combines classic iterative algorithms with deep learning features, enabling high-quality image restoration without large training datasets.

Contribution

It integrates classic Richardson-Lucy deconvolution with deep feature extraction and zero-shot optimization, reducing data dependency and computational resources.

Findings

Significant improvement in real-world deconvolution tasks

Reduced need for large training datasets

Faster reconstruction with high-quality results

Abstract

Non-blind deconvolution aims to restore a sharp image from its blurred counterpart given an obtained kernel. Existing deep neural architectures are often built based on large datasets of sharp ground truth images and trained with supervision. Sharp, high quality ground truth images, however, are not always available, especially for biomedical applications. This severely hampers the applicability of current approaches in practice. In this paper, we propose a novel non-blind deconvolution method that leverages the power of deep learning and classic iterative deconvolution algorithms. Our approach combines a pre-trained network to extract deep features from the input image with iterative Richardson-Lucy deconvolution steps. Subsequently, a zero-shot optimisation process is employed to integrate the deconvolved features, resulting in a high-quality reconstructed image. By performing the preliminary reconstruction with the classic iterative deconvolution method, we can effectively utilise a smaller network to produce the final image, thus accelerating the reconstruction whilst reducing the demand for valuable computational resources. Our method demonstrates significant improvements in various real-world applications non-blind deconvolution tasks.