Efficient Image Super-Resolution via Symmetric Visual Attention Network

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wu, Chengxu, Fan, Qinrui, Hu, Shu, Wu, Xi, Wang, Xin, Hu, Jing
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911759479078912
author Wu, Chengxu
Fan, Qinrui
Hu, Shu
Wu, Xi
Wang, Xin
Hu, Jing
author_facet Wu, Chengxu
Fan, Qinrui
Hu, Shu
Wu, Xi
Wang, Xin
Hu, Jing
contents An important development direction in the Single-Image Super-Resolution (SISR) algorithms is to improve the efficiency of the algorithms. Recently, efficient Super-Resolution (SR) research focuses on reducing model complexity and improving efficiency through improved deep small kernel convolution, leading to a small receptive field. The large receptive field obtained by large kernel convolution can significantly improve image quality, but the computational cost is too high. To improve the reconstruction details of efficient super-resolution reconstruction, we propose a Symmetric Visual Attention Network (SVAN) by applying large receptive fields. The SVAN decomposes a large kernel convolution into three different combinations of convolution operations and combines them with an attention mechanism to form a Symmetric Large Kernel Attention Block (SLKAB), which forms a symmetric attention block with a bottleneck structure by the size of the receptive field in the convolution combination to extract depth features effectively as the basic component of the SVAN. Our network gets a large receptive field while minimizing the number of parameters and improving the perceptual ability of the model. The experimental results show that the proposed SVAN can obtain high-quality super-resolution reconstruction results using only about 30% of the parameters of existing SOTA methods.
format Preprint
id arxiv_https___arxiv_org_abs_2401_08913
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Efficient Image Super-Resolution via Symmetric Visual Attention Network
Wu, Chengxu
Fan, Qinrui
Hu, Shu
Wu, Xi
Wang, Xin
Hu, Jing
Computer Vision and Pattern Recognition
Image and Video Processing
An important development direction in the Single-Image Super-Resolution (SISR) algorithms is to improve the efficiency of the algorithms. Recently, efficient Super-Resolution (SR) research focuses on reducing model complexity and improving efficiency through improved deep small kernel convolution, leading to a small receptive field. The large receptive field obtained by large kernel convolution can significantly improve image quality, but the computational cost is too high. To improve the reconstruction details of efficient super-resolution reconstruction, we propose a Symmetric Visual Attention Network (SVAN) by applying large receptive fields. The SVAN decomposes a large kernel convolution into three different combinations of convolution operations and combines them with an attention mechanism to form a Symmetric Large Kernel Attention Block (SLKAB), which forms a symmetric attention block with a bottleneck structure by the size of the receptive field in the convolution combination to extract depth features effectively as the basic component of the SVAN. Our network gets a large receptive field while minimizing the number of parameters and improving the perceptual ability of the model. The experimental results show that the proposed SVAN can obtain high-quality super-resolution reconstruction results using only about 30% of the parameters of existing SOTA methods.
title Efficient Image Super-Resolution via Symmetric Visual Attention Network
topic Computer Vision and Pattern Recognition
Image and Video Processing
url https://arxiv.org/abs/2401.08913