CFAT: Unleashing TriangularWindows for Image Super-resolution

TL;DR

This paper introduces CFAT, a novel transformer-based model for image super-resolution that combines triangular and rectangular window techniques to better capture features and reduce boundary distortion, leading to improved performance.

Contribution

The paper proposes a new non-overlapping triangular window technique integrated with rectangular windows in a transformer model for enhanced image super-resolution.

Findings

Achieves 0.7 dB higher PSNR than state-of-the-art models

Effectively reduces boundary distortion in super-resolution tasks

Captures long-range, multi-scale features for better image quality

Abstract

Transformer-based models have revolutionized the field of image super-resolution (SR) by harnessing their inherent ability to capture complex contextual features. The overlapping rectangular shifted window technique used in transformer architecture nowadays is a common practice in super-resolution models to improve the quality and robustness of image upscaling. However, it suffers from distortion at the boundaries and has limited unique shifting modes. To overcome these weaknesses, we propose a non-overlapping triangular window technique that synchronously works with the rectangular one to mitigate boundary-level distortion and allows the model to access more unique sifting modes. In this paper, we propose a Composite Fusion Attention Transformer (CFAT) that incorporates triangular-rectangular window-based local attention with a channel-based global attention technique in image super-resolution. As a result, CFAT enables attention mechanisms to be activated on more image pixels and captures long-range, multi-scale features to improve SR performance. The extensive experimental results and ablation study demonstrate the effectiveness of CFAT in the SR domain. Our proposed model shows a significant 0.7 dB performance improvement over other state-of-the-art SR architectures.