Rethinking Blur Synthesis for Deep Real-World Image Deblurring

TL;DR

This paper introduces a realistic blur synthesis pipeline to improve training data quality and a multi-path transformer-enhanced UNet for better real-world image deblurring, resulting in superior performance on real datasets.

Contribution

It presents a novel blur synthesis method to bridge the domain gap and a multi-path transformer module for enhanced feature learning in deblurring models.

Findings

Enhanced deblurring performance on real datasets

Better robustness to real-world blur

Outperforms state-of-the-art methods

Abstract

In this paper, we examine the problem of real-world image deblurring and take into account two key factors for improving the performance of the deep image deblurring model, namely, training data synthesis and network architecture design. Deblurring models trained on existing synthetic datasets perform poorly on real blurry images due to domain shift. To reduce the domain gap between synthetic and real domains, we propose a novel realistic blur synthesis pipeline to simulate the camera imaging process. As a result of our proposed synthesis method, existing deblurring models could be made more robust to handle real-world blur. Furthermore, we develop an effective deblurring model that captures non-local dependencies and local context in the feature domain simultaneously. Specifically, we introduce the multi-path transformer module to UNet architecture for enriched multi-scale features learning. A comprehensive experiment on three real-world datasets shows that the proposed deblurring model performs better than state-of-the-art methods.