Scale-recurrent Network for Deep Image Deblurring

TL;DR

This paper introduces a scale-recurrent neural network for single image deblurring that simplifies the architecture, reduces parameters, and outperforms existing methods on complex motion datasets.

Contribution

Proposes a novel scale-recurrent network architecture for image deblurring that is simpler, more efficient, and achieves superior results compared to prior approaches.

Findings

Outperforms state-of-the-art methods quantitatively

Produces better qualitative deblurring results

Has a smaller, easier-to-train network structure

Abstract

In single image deblurring, the "coarse-to-fine" scheme, i.e. gradually restoring the sharp image on different resolutions in a pyramid, is very successful in both traditional optimization-based methods and recent neural-network-based approaches. In this paper, we investigate this strategy and propose a Scale-recurrent Network (SRN-DeblurNet) for this deblurring task. Compared with the many recent learning-based approaches in [25], it has a simpler network structure, a smaller number of parameters and is easier to train. We evaluate our method on large-scale deblurring datasets with complex motion. Results show that our method can produce better quality results than state-of-the-arts, both quantitatively and qualitatively.