Separable feature complementary network with branch-wise and multi-scale spatial attention for lightweight image super-resolution

被引:0
作者
Wenming Zhang
Qiming Han
Yaqian Li
Haibin Li
机构
[1] Yanshan University,Engineering Research Center of the Ministry of Education for Intelligent Control System and Intelligent Equipment
[2] Yanshan University,Key Laboratory of Industrial Computer Control Engineering of Hebei Province
来源
Signal, Image and Video Processing | 2024年 / 18卷
关键词
Super-resolution; Feature complementary; Lightweight convolutional neural networks; Branch-wise attention; Feature fusion;
D O I
暂无
中图分类号
学科分类号
摘要
Recent advances in single image super-resolution (SISR) have shown promising results, but networks with optimal performance tend to have heavy computation, making them unsuitable for edge devices. How to achieve better results with fewer parameters is still a problem that requires further research. To overcome this issue, we propose a separable feature complementary network using branch-wise attention and multi-scale spatial attention (SFCN-BMSA). The network contains a feature complementary module, which utilizes a limited number of small-sized convolution kernels to combine long-range features from different positions on the feature map and utilizes them to enhance image reconstruction. In addition, we design a feature fusion module with branch-wise attention, which can fuse the features of different branches according to the importance of each branch. Finally, we also design a multi-scale spatial attention module, which utilizes three dilated convolutions with the size of 5×\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\times $$\end{document}5 to calculate attention from different spatial scales and combines them to obtain more refined attention while utilizing a larger receptive field. Experiments show that the proposed neural network achieves better reconstruction results with lower parameters.
引用
收藏
页码:1715 / 1724
页数:9
相关论文
共 31 条
[1]  
Yue L(2016)Image super-resolution: the techniques, applications, and future Signal Process 128 389-408
[2]  
Shen H(2014)Super-resolution: a comprehensive survey Machine Vis Appl 25 1423-1468
[3]  
Li J(2020)Deep learning for image super-resolution: a survey IEEE Trans. Patt. Anal. Mach. Intell. 43 3365-3387
[4]  
Yuan Q(2015)Image super-resolution using deep convolutional networks IEEE Trans Patt Anal Machine Intelligence 38 295-307
[5]  
Zhang H(2018)A new method for image super-resolution with multi-channel constraints Knowl-Based Syst 146 118-128
[6]  
Zhang L(2019)Super-resolution using neighbourhood regression with local structure prior Signal Process.: Image Commun. 72 58-68
[7]  
Nasrollahi K(2017)Sketch-based manga retrieval using manga109 dataset Multim. Tools Appl. 76 1811-21838
[8]  
Moeslund TB(2022)Exploring more diverse network architectures for single image super-resolution Knowl-Based Syst. 235 107648-undefined
[9]  
Wang Z(undefined)undefined undefined undefined undefined-undefined
[10]  
Chen J(undefined)undefined undefined undefined undefined-undefined