DLGSANet: Lightweight Dynamic Local and Global Self-Attention Networks for Image Super-Resolution

TL;DR

DLGSANet introduces a lightweight, efficient Transformer-based network for image super-resolution that combines local and global self-attention modules to improve high-resolution image reconstruction with fewer parameters.

Contribution

The paper proposes a novel hybrid dynamic-Transformer architecture with local and global self-attention modules, reducing computational costs while maintaining competitive super-resolution performance.

Findings

Achieves competitive super-resolution results with fewer parameters.

Introduces a sparse global self-attention module for better feature utilization.

Demonstrates lower computational costs compared to state-of-the-art methods.

Abstract

We propose an effective lightweight dynamic local and global self-attention network (DLGSANet) to solve image super-resolution. Our method explores the properties of Transformers while having low computational costs. Motivated by the network designs of Transformers, we develop a simple yet effective multi-head dynamic local self-attention (MHDLSA) module to extract local features efficiently. In addition, we note that existing Transformers usually explore all similarities of the tokens between the queries and keys for the feature aggregation. However, not all the tokens from the queries are relevant to those in keys, using all the similarities does not effectively facilitate the high-resolution image reconstruction. To overcome this problem, we develop a sparse global self-attention (SparseGSA) module to select the most useful similarity values so that the most useful global features can be better utilized for the high-resolution image reconstruction. We develop a hybrid dynamic-Transformer block(HDTB) that integrates the MHDLSA and SparseGSA for both local and global feature exploration. To ease the network training, we formulate the HDTBs into a residual hybrid dynamic-Transformer group (RHDTG). By embedding the RHDTGs into an end-to-end trainable network, we show that our proposed method has fewer network parameters and lower computational costs while achieving competitive performance against state-of-the-art ones in terms of accuracy. More information is available at https://neonleexiang.github.io/DLGSANet/