Wavelet attention-based implicit multi-granularity super-resolution network

Abstract Image super-resolution (SR) is a fundamental challenge in the field of computer vision. Recently, Convolutional Neural Network (CNN)-based methods for image SR have achieved significant progress across various SR tasks. However, most current research focuses on designing deeper and wider ar...

Full description

Saved in:
Bibliographic Details
Main Authors: Chen Boying, Shi Jie
Format: Article
Language:English
Published: Springer 2025-04-01
Series:Complex & Intelligent Systems
Subjects:
Online Access:https://doi.org/10.1007/s40747-025-01862-4
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850042265135742976
author Chen Boying
Shi Jie
author_facet Chen Boying
Shi Jie
author_sort Chen Boying
collection DOAJ
description Abstract Image super-resolution (SR) is a fundamental challenge in the field of computer vision. Recently, Convolutional Neural Network (CNN)-based methods for image SR have achieved significant progress across various SR tasks. However, most current research focuses on designing deeper and wider architectures, often sacrificing computational burden and speed in order to improve image SR quality. To achieve more efficient SR methods, this paper proposes a Wavelet Attention Network (WANet) for image SR. Firstly, a wavelet-based attention module is proposed. Compared to existing self-attention modules, the wavelet attention module decomposes image features into different frequency components using wavelet transforms. It then applies a self-attention mechanism to capture multi-scale features, enabling a more efficient and larger receptive field to help the network capture long-range feature dependencies. Secondly, local implicit features are introduced to enhance the encoder’s ability to aggregate local neighborhood features. Finally, coarse and fine-grained interwoven pixel features are collaboratively associated to improve the performance of the implicit feature decoder. Experimental comparisons with state-of-the-art SR methods demonstrate the effectiveness and superiority of WANet in the field of image SR.
format Article
id doaj-art-b78a5eb2df4346df8488518f29047f50
institution DOAJ
issn 2199-4536
2198-6053
language English
publishDate 2025-04-01
publisher Springer
record_format Article
series Complex & Intelligent Systems
spelling doaj-art-b78a5eb2df4346df8488518f29047f502025-08-20T02:55:36ZengSpringerComplex & Intelligent Systems2199-45362198-60532025-04-0111511110.1007/s40747-025-01862-4Wavelet attention-based implicit multi-granularity super-resolution networkChen Boying0Shi Jie1School of Electronics and Information Engineering, Suzhou University of Science and TechnologySchool of Electronics and Information Engineering, Suzhou University of Science and TechnologyAbstract Image super-resolution (SR) is a fundamental challenge in the field of computer vision. Recently, Convolutional Neural Network (CNN)-based methods for image SR have achieved significant progress across various SR tasks. However, most current research focuses on designing deeper and wider architectures, often sacrificing computational burden and speed in order to improve image SR quality. To achieve more efficient SR methods, this paper proposes a Wavelet Attention Network (WANet) for image SR. Firstly, a wavelet-based attention module is proposed. Compared to existing self-attention modules, the wavelet attention module decomposes image features into different frequency components using wavelet transforms. It then applies a self-attention mechanism to capture multi-scale features, enabling a more efficient and larger receptive field to help the network capture long-range feature dependencies. Secondly, local implicit features are introduced to enhance the encoder’s ability to aggregate local neighborhood features. Finally, coarse and fine-grained interwoven pixel features are collaboratively associated to improve the performance of the implicit feature decoder. Experimental comparisons with state-of-the-art SR methods demonstrate the effectiveness and superiority of WANet in the field of image SR.https://doi.org/10.1007/s40747-025-01862-4Continuous super-resolution reconstructionImplicit neural representationWavelet transformAttention mechanism
spellingShingle Chen Boying
Shi Jie
Wavelet attention-based implicit multi-granularity super-resolution network
Complex & Intelligent Systems
Continuous super-resolution reconstruction
Implicit neural representation
Wavelet transform
Attention mechanism
title Wavelet attention-based implicit multi-granularity super-resolution network
title_full Wavelet attention-based implicit multi-granularity super-resolution network
title_fullStr Wavelet attention-based implicit multi-granularity super-resolution network
title_full_unstemmed Wavelet attention-based implicit multi-granularity super-resolution network
title_short Wavelet attention-based implicit multi-granularity super-resolution network
title_sort wavelet attention based implicit multi granularity super resolution network
topic Continuous super-resolution reconstruction
Implicit neural representation
Wavelet transform
Attention mechanism
url https://doi.org/10.1007/s40747-025-01862-4
work_keys_str_mv AT chenboying waveletattentionbasedimplicitmultigranularitysuperresolutionnetwork
AT shijie waveletattentionbasedimplicitmultigranularitysuperresolutionnetwork