Cargando…

S2A: Scale-Attention-Aware Networks for Video Super-Resolution

Convolutional Neural Networks (CNNs) have been widely used in video super-resolution (VSR). Most existing VSR methods focus on how to utilize the information of multiple frames, while neglecting the feature correlations of the intermediate features, thus limiting the feature expression of the models...

Descripción completa

Detalles Bibliográficos
Autores principales: Guo, Taian, Dai, Tao, Liu, Ling, Zhu, Zexuan, Xia, Shu-Tao
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8619237/
https://www.ncbi.nlm.nih.gov/pubmed/34828096
http://dx.doi.org/10.3390/e23111398
_version_ 1784604941457817600
author Guo, Taian
Dai, Tao
Liu, Ling
Zhu, Zexuan
Xia, Shu-Tao
author_facet Guo, Taian
Dai, Tao
Liu, Ling
Zhu, Zexuan
Xia, Shu-Tao
author_sort Guo, Taian
collection PubMed
description Convolutional Neural Networks (CNNs) have been widely used in video super-resolution (VSR). Most existing VSR methods focus on how to utilize the information of multiple frames, while neglecting the feature correlations of the intermediate features, thus limiting the feature expression of the models. To address this problem, we propose a novel SAA network, that is, Scale-and-Attention-Aware Networks, to apply different attention to different temporal-length streams, while further exploring both spatial and channel attention on separate streams with a newly proposed Criss-Cross Channel Attention Module ([Formula: see text]). Experiments on public VSR datasets demonstrate the superiority of our method over other state-of-the-art methods in terms of both quantitative and qualitative metrics.
format Online
Article
Text
id pubmed-8619237
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-86192372021-11-27 S2A: Scale-Attention-Aware Networks for Video Super-Resolution Guo, Taian Dai, Tao Liu, Ling Zhu, Zexuan Xia, Shu-Tao Entropy (Basel) Article Convolutional Neural Networks (CNNs) have been widely used in video super-resolution (VSR). Most existing VSR methods focus on how to utilize the information of multiple frames, while neglecting the feature correlations of the intermediate features, thus limiting the feature expression of the models. To address this problem, we propose a novel SAA network, that is, Scale-and-Attention-Aware Networks, to apply different attention to different temporal-length streams, while further exploring both spatial and channel attention on separate streams with a newly proposed Criss-Cross Channel Attention Module ([Formula: see text]). Experiments on public VSR datasets demonstrate the superiority of our method over other state-of-the-art methods in terms of both quantitative and qualitative metrics. MDPI 2021-10-25 /pmc/articles/PMC8619237/ /pubmed/34828096 http://dx.doi.org/10.3390/e23111398 Text en © 2021 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Guo, Taian
Dai, Tao
Liu, Ling
Zhu, Zexuan
Xia, Shu-Tao
S2A: Scale-Attention-Aware Networks for Video Super-Resolution
title S2A: Scale-Attention-Aware Networks for Video Super-Resolution
title_full S2A: Scale-Attention-Aware Networks for Video Super-Resolution
title_fullStr S2A: Scale-Attention-Aware Networks for Video Super-Resolution
title_full_unstemmed S2A: Scale-Attention-Aware Networks for Video Super-Resolution
title_short S2A: Scale-Attention-Aware Networks for Video Super-Resolution
title_sort s2a: scale-attention-aware networks for video super-resolution
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8619237/
https://www.ncbi.nlm.nih.gov/pubmed/34828096
http://dx.doi.org/10.3390/e23111398
work_keys_str_mv AT guotaian s2ascaleattentionawarenetworksforvideosuperresolution
AT daitao s2ascaleattentionawarenetworksforvideosuperresolution
AT liuling s2ascaleattentionawarenetworksforvideosuperresolution
AT zhuzexuan s2ascaleattentionawarenetworksforvideosuperresolution
AT xiashutao s2ascaleattentionawarenetworksforvideosuperresolution