Efficient Transformer for High Resolution Image Motion Deblurring

TL;DR

This paper improves the Restormer architecture for high-resolution image motion deblurring by reducing complexity and enhancing training, resulting in more efficient models with maintained or better performance across multiple datasets.

Contribution

The authors introduce architectural modifications and an enhanced training pipeline that together reduce model complexity and improve robustness for high-resolution image deblurring.

Findings

18.4% reduction in model complexity

Improved convergence and reduced training time

Competitive performance on multiple datasets

Abstract

This paper presents a comprehensive study and improvement of the Restormer architecture for high-resolution image motion deblurring. We introduce architectural modifications that reduce model complexity by 18.4% while maintaining or improving performance through optimized attention mechanisms. Our enhanced training pipeline incorporates additional transformations including color jitter, Gaussian blur, and perspective transforms to improve model robustness as well as a new frequency loss term. Extensive experiments on the RealBlur-R, RealBlur-J, and Ultra-High-Definition Motion blurred (UHDM) datasets demonstrate the effectiveness of our approach. The improved architecture shows better convergence behavior and reduced training time while maintaining competitive performance across challenging scenarios. We also provide detailed ablation studies analyzing the impact of our modifications on model behavior and performance. Our results suggest that thoughtful architectural simplification combined with enhanced training strategies can yield more efficient yet equally capable models for motion deblurring tasks. Code and Data Available at: https://github.com/hamzafer/image-deblurring