Cargando…

Multi-Stage Attentive Network for Motion Deblurring via Binary Cross-Entropy Loss

In this paper, we present the multi-stage attentive network (MSAN), an efficient and good generalization performance convolutional neural network (CNN) architecture for motion deblurring. We build a multi-stage encoder–decoder network with self-attention and use the binary cross-entropy loss to trai...

Descripción completa

Detalles Bibliográficos
Autores principales: Guo, Cai, Chen, Xinan, Chen, Yanhua, Yu, Chuying
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9601862/
https://www.ncbi.nlm.nih.gov/pubmed/37420434
http://dx.doi.org/10.3390/e24101414
_version_ 1784817169509384192
author Guo, Cai
Chen, Xinan
Chen, Yanhua
Yu, Chuying
author_facet Guo, Cai
Chen, Xinan
Chen, Yanhua
Yu, Chuying
author_sort Guo, Cai
collection PubMed
description In this paper, we present the multi-stage attentive network (MSAN), an efficient and good generalization performance convolutional neural network (CNN) architecture for motion deblurring. We build a multi-stage encoder–decoder network with self-attention and use the binary cross-entropy loss to train our model. In MSAN, there are two core designs. First, we introduce a new attention-based end-to-end method on top of multi-stage networks, which applies group convolution to the self-attention module, effectively reducing the computing cost and improving the model’s adaptability to different blurred images. Secondly, we propose using binary cross-entropy loss instead of pixel loss to optimize our model to minimize the over-smoothing impact of pixel loss while maintaining a good deblurring effect. We conduct extensive experiments on several deblurring datasets to evaluate the performance of our solution for deblurring. Our MSAN achieves superior performance while also generalizing and compares well with state-of-the-art methods.
format Online
Article
Text
id pubmed-9601862
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-96018622022-10-27 Multi-Stage Attentive Network for Motion Deblurring via Binary Cross-Entropy Loss Guo, Cai Chen, Xinan Chen, Yanhua Yu, Chuying Entropy (Basel) Article In this paper, we present the multi-stage attentive network (MSAN), an efficient and good generalization performance convolutional neural network (CNN) architecture for motion deblurring. We build a multi-stage encoder–decoder network with self-attention and use the binary cross-entropy loss to train our model. In MSAN, there are two core designs. First, we introduce a new attention-based end-to-end method on top of multi-stage networks, which applies group convolution to the self-attention module, effectively reducing the computing cost and improving the model’s adaptability to different blurred images. Secondly, we propose using binary cross-entropy loss instead of pixel loss to optimize our model to minimize the over-smoothing impact of pixel loss while maintaining a good deblurring effect. We conduct extensive experiments on several deblurring datasets to evaluate the performance of our solution for deblurring. Our MSAN achieves superior performance while also generalizing and compares well with state-of-the-art methods. MDPI 2022-10-03 /pmc/articles/PMC9601862/ /pubmed/37420434 http://dx.doi.org/10.3390/e24101414 Text en © 2022 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, Cai
Chen, Xinan
Chen, Yanhua
Yu, Chuying
Multi-Stage Attentive Network for Motion Deblurring via Binary Cross-Entropy Loss
title Multi-Stage Attentive Network for Motion Deblurring via Binary Cross-Entropy Loss
title_full Multi-Stage Attentive Network for Motion Deblurring via Binary Cross-Entropy Loss
title_fullStr Multi-Stage Attentive Network for Motion Deblurring via Binary Cross-Entropy Loss
title_full_unstemmed Multi-Stage Attentive Network for Motion Deblurring via Binary Cross-Entropy Loss
title_short Multi-Stage Attentive Network for Motion Deblurring via Binary Cross-Entropy Loss
title_sort multi-stage attentive network for motion deblurring via binary cross-entropy loss
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9601862/
https://www.ncbi.nlm.nih.gov/pubmed/37420434
http://dx.doi.org/10.3390/e24101414
work_keys_str_mv AT guocai multistageattentivenetworkformotiondeblurringviabinarycrossentropyloss
AT chenxinan multistageattentivenetworkformotiondeblurringviabinarycrossentropyloss
AT chenyanhua multistageattentivenetworkformotiondeblurringviabinarycrossentropyloss
AT yuchuying multistageattentivenetworkformotiondeblurringviabinarycrossentropyloss