Multi-Label Image Classification by Feature Attention Network

<p dir="ltr">Learning the correlation among labels is a standing-problem in the multi-label image recognition task. The label correlation is the key to solve the multi-label classification but it is too abstract to model. Most solutions try to learn image label dependencies to improv...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Zheng Yan (194231) (author)
مؤلفون آخرون: Weiwei Liu (341566) (author), Shiping Wen (7168688) (author), Yin Yang (35103) (author)
منشور في: 2019
الموضوعات:
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513505623801856
author Zheng Yan (194231)
author2 Weiwei Liu (341566)
Shiping Wen (7168688)
Yin Yang (35103)
author2_role author
author
author
author_facet Zheng Yan (194231)
Weiwei Liu (341566)
Shiping Wen (7168688)
Yin Yang (35103)
author_role author
dc.creator.none.fl_str_mv Zheng Yan (194231)
Weiwei Liu (341566)
Shiping Wen (7168688)
Yin Yang (35103)
dc.date.none.fl_str_mv 2019-07-18T06:00:00Z
dc.identifier.none.fl_str_mv 10.1109/access.2019.2929512
dc.relation.none.fl_str_mv https://figshare.com/articles/journal_contribution/Multi-Label_Image_Classification_by_Feature_Attention_Network/27003817
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Information and computing sciences
Machine learning
Correlation
Neural networks
Task analysis
Convolution
Semantics
Transforms
Object detection
Deep neural network
multi-label recognition
label correlation
attention
dc.title.none.fl_str_mv Multi-Label Image Classification by Feature Attention Network
dc.type.none.fl_str_mv Text
Journal contribution
info:eu-repo/semantics/publishedVersion
text
contribution to journal
description <p dir="ltr">Learning the correlation among labels is a standing-problem in the multi-label image recognition task. The label correlation is the key to solve the multi-label classification but it is too abstract to model. Most solutions try to learn image label dependencies to improve multi-label classification performance. However, they have ignored two more realistic problems: object scale inconsistent and label tail (category imbalance). These two problems will impact the bad influence on the classification model. To tackle these two problems and learn the label correlations, we propose feature attention network (FAN) which contains feature refinement network and correlation learning network. FAN builds top-down feature fusion mechanism to refine more important features and learn the correlations among convolutional features from FAN to indirect learn the label dependencies. Following our proposed solution, we achieve performed classification accuracy on MSCOCO 2014 and VOC 2007 dataset.</p><h2>Other Information</h2><p dir="ltr">Published in: IEEE Access<br>License: <a href="https://creativecommons.org/licenses/by/4.0/" rel="noreferrer" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1109/access.2019.2929512" target="_blank">https://dx.doi.org/10.1109/access.2019.2929512</a></p>
eu_rights_str_mv openAccess
id Manara2_d36992f4e9e87958efc520a89f1f3005
identifier_str_mv 10.1109/access.2019.2929512
network_acronym_str Manara2
network_name_str Manara2
oai_identifier_str oai:figshare.com:article/27003817
publishDate 2019
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling Multi-Label Image Classification by Feature Attention NetworkZheng Yan (194231)Weiwei Liu (341566)Shiping Wen (7168688)Yin Yang (35103)Information and computing sciencesMachine learningCorrelationNeural networksTask analysisConvolutionSemanticsTransformsObject detectionDeep neural networkmulti-label recognitionlabel correlationattention<p dir="ltr">Learning the correlation among labels is a standing-problem in the multi-label image recognition task. The label correlation is the key to solve the multi-label classification but it is too abstract to model. Most solutions try to learn image label dependencies to improve multi-label classification performance. However, they have ignored two more realistic problems: object scale inconsistent and label tail (category imbalance). These two problems will impact the bad influence on the classification model. To tackle these two problems and learn the label correlations, we propose feature attention network (FAN) which contains feature refinement network and correlation learning network. FAN builds top-down feature fusion mechanism to refine more important features and learn the correlations among convolutional features from FAN to indirect learn the label dependencies. Following our proposed solution, we achieve performed classification accuracy on MSCOCO 2014 and VOC 2007 dataset.</p><h2>Other Information</h2><p dir="ltr">Published in: IEEE Access<br>License: <a href="https://creativecommons.org/licenses/by/4.0/" rel="noreferrer" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1109/access.2019.2929512" target="_blank">https://dx.doi.org/10.1109/access.2019.2929512</a></p>2019-07-18T06:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1109/access.2019.2929512https://figshare.com/articles/journal_contribution/Multi-Label_Image_Classification_by_Feature_Attention_Network/27003817CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/270038172019-07-18T06:00:00Z
spellingShingle Multi-Label Image Classification by Feature Attention Network
Zheng Yan (194231)
Information and computing sciences
Machine learning
Correlation
Neural networks
Task analysis
Convolution
Semantics
Transforms
Object detection
Deep neural network
multi-label recognition
label correlation
attention
status_str publishedVersion
title Multi-Label Image Classification by Feature Attention Network
title_full Multi-Label Image Classification by Feature Attention Network
title_fullStr Multi-Label Image Classification by Feature Attention Network
title_full_unstemmed Multi-Label Image Classification by Feature Attention Network
title_short Multi-Label Image Classification by Feature Attention Network
title_sort Multi-Label Image Classification by Feature Attention Network
topic Information and computing sciences
Machine learning
Correlation
Neural networks
Task analysis
Convolution
Semantics
Transforms
Object detection
Deep neural network
multi-label recognition
label correlation
attention