From Attention to Frequency: Integration of Vision Transformer and FFT-ReLU for Enhanced Image Deblurring

TL;DR

This paper introduces a dual-domain image deblurring architecture combining Vision Transformers with an FFT-ReLU module to effectively suppress blur artifacts and enhance image details, outperforming existing methods.

Contribution

It presents a novel integration of spatial attention and frequency sparsity in a unified model for improved image deblurring performance.

Findings

Achieves superior PSNR and SSIM on benchmark datasets.

Outperforms state-of-the-art models in perceptual quality.

Validated through quantitative, qualitative, and human preference evaluations.

Abstract

Image deblurring is vital in computer vision, aiming to recover sharp images from blurry ones caused by motion or camera shake. While deep learning approaches such as CNNs and Vision Transformers (ViTs) have advanced this field, they often struggle with complex or high-resolution blur and computational demands. We propose a new dual-domain architecture that unifies Vision Transformers with a frequency-domain FFT-ReLU module, explicitly bridging spatial attention modeling and frequency sparsity. In this structure, the ViT backbone captures local and global dependencies, while the FFT-ReLU component enforces frequency-domain sparsity to suppress blur-related artifacts and preserve fine details. Extensive experiments on benchmark datasets demonstrate that this architecture achieves superior PSNR, SSIM, and perceptual quality compared to state-of-the-art models. Both quantitative metrics, qualitative comparisons, and human preference evaluations confirm its effectiveness, establishing a practical and generalizable paradigm for real-world image restoration.