DeblurDiNAT: A Compact Model with Exceptional Generalization and Visual Fidelity on Unseen Domains

TL;DR

DeblurDiNAT is a compact Transformer-based deblurring model that excels in generalizing to unseen domains and improves perceptual image quality, addressing limitations of prior methods focused on distortion metrics.

Contribution

It introduces a novel Dilated Neighborhood Attention mechanism, a local cross-channel learner, and a dual-stage feature fusion module for enhanced generalization and perceptual fidelity.

Findings

Outperforms state-of-the-art models in perceptual metrics

Demonstrates superior generalization to unseen domains

Maintains a compact model size

Abstract

Recent deblurring networks have effectively restored clear images from the blurred ones. However, they often struggle with generalization to unknown domains. Moreover, these models typically focus on distortion metrics such as PSNR and SSIM, neglecting the critical aspect of metrics aligned with human perception. To address these limitations, we propose DeblurDiNAT, a deblurring Transformer based on Dilated Neighborhood Attention. First, DeblurDiNAT employs an alternating dilation factor paradigm to capture both local and global blurred patterns, enhancing generalization and perceptual clarity. Second, a local cross-channel learner aids the Transformer block to understand the short-range relationships between adjacent channels. Additionally, we present a linear feed-forward network with a simple while effective design. Finally, a dual-stage feature fusion module is introduced as an alternative to the existing approach, which efficiently process multi-scale visual information across network levels. Compared to state-of-the-art models, our compact DeblurDiNAT demonstrates superior generalization capabilities and achieves remarkable performance in perceptual metrics, while maintaining a favorable model size.