Efficient Image Super-Resolution with Feature Interaction Weighted Hybrid Network

TL;DR

This paper introduces FIWHN, a hybrid neural network architecture combining CNN and Transformer components, designed to enhance image super-resolution by effectively preserving and fusing features while maintaining low computational costs.

Contribution

The paper proposes a novel Feature Interaction Weighted Hybrid Network (FIWHN) that integrates Wide-residual Distillation Interaction Blocks, feature shuffling, and Transformer modules for improved super-resolution.

Findings

Achieves a good balance between performance and efficiency.

Effectively preserves intermediate features during reconstruction.

Demonstrates superior results on low- and high-level tasks.

Abstract

Lightweight image super-resolution aims to reconstruct high-resolution images from low-resolution images using low computational costs. However, existing methods result in the loss of middle-layer features due to activation functions. To minimize the impact of intermediate feature loss on reconstruction quality, we propose a Feature Interaction Weighted Hybrid Network (FIWHN), which comprises a series of Wide-residual Distillation Interaction Block (WDIB) as the backbone. Every third WDIB forms a Feature Shuffle Weighted Group (FSWG) by applying mutual information shuffle and fusion. Moreover, to mitigate the negative effects of intermediate feature loss, we introduce Wide Residual Weighting units within WDIB. These units effectively fuse features of varying levels of detail through a Wide-residual Distillation Connection (WRDC) and a Self-Calibrating Fusion (SCF). To compensate for global feature deficiencies, we incorporate a Transformer and explore a novel architecture to combine CNN and Transformer. We show that our FIWHN achieves a favorable balance between performance and efficiency through extensive experiments on low-level and high-level tasks. Codes will be available at \url{https://github.com/IVIPLab/FIWHN}.