BluRef: Unsupervised Image Deblurring with Dense-Matching References

TL;DR

This paper presents an unsupervised image deblurring method that uses unpaired images and dense matching to generate pseudo-ground truth, achieving state-of-the-art results without requiring paired training data.

Contribution

It introduces a simple, unsupervised deblurring approach that does not rely on paired datasets or pre-trained models, enhancing adaptability and performance.

Findings

Achieves state-of-the-art deblurring performance

Does not require paired training data

Compatible with low-resource devices

Abstract

This paper introduces a novel unsupervised approach for image deblurring that utilizes a simple process for training data collection, thereby enhancing the applicability and effectiveness of deblurring methods. Our technique does not require meticulously paired data of blurred and corresponding sharp images; instead, it uses unpaired blurred and sharp images of similar scenes to generate pseudo-ground truth data by leveraging a dense matching model to identify correspondences between a blurry image and reference sharp images. Thanks to the simplicity of the training data collection process, our approach does not rely on existing paired training data or pre-trained networks, making it more adaptable to various scenarios and suitable for networks of different sizes, including those designed for low-resource devices. We demonstrate that this novel approach achieves state-of-the-art performance, marking a significant advancement in the field of image deblurring.