SL-CycleGAN: Blind Motion Deblurring in Cycles using Sparse Learning

TL;DR

This paper presents SL-CycleGAN, a novel GAN-based model for blind motion deblurring that integrates sparse learning with cycle consistency, achieving state-of-the-art results on benchmark datasets.

Contribution

Introduction of a sparse ResNet-block with HTM-based spatial pooler in a CycleGAN framework for blind motion deblurring, demonstrating superior performance.

Findings

Achieved a PSNR of 38.087 dB on GoPro dataset.

Outperformed recent methods by 5.377 dB in PSNR.

Validated effectiveness through extensive qualitative and quantitative experiments.

Abstract

In this paper, we introduce an end-to-end generative adversarial network (GAN) based on sparse learning for single image blind motion deblurring, which we called SL-CycleGAN. For the first time in blind motion deblurring, we propose a sparse ResNet-block as a combination of sparse convolution layers and a trainable spatial pooler k-winner based on HTM (Hierarchical Temporal Memory) to replace non-linearity such as ReLU in the ResNet-block of SL-CycleGAN generators. Furthermore, unlike many state-of-the-art GAN-based motion deblurring methods that treat motion deblurring as a linear end-to-end process, we take our inspiration from the domain-to-domain translation ability of CycleGAN, and we show that image deblurring can be cycle-consistent while achieving the best qualitative results. Finally, we perform extensive experiments on popular image benchmarks both qualitatively and quantitatively and achieve the record-breaking PSNR of 38.087 dB on GoPro dataset, which is 5.377 dB better than the most recent deblurring method.