A novel facial expression recognition framework using deep learning based dynamic cross-domain dual attention network

Variations in domain targets have recently posed significant challenges for facial expression recognition tasks, primarily due to domain shifts. Current methods focus largely on global feature adoption to achieve domain-invariant learning; however, transferring local features across diverse domains...

Full description

Saved in:
Bibliographic Details
Main Authors: Ahmed Omar Alzahrani, Ahmed Mohammed Alghamdi, M. Usman Ashraf, Iqra Ilyas, Nadeem Sarwar, Abdulrahman Alzahrani, Alaa Abdul Salam Alarood
Format: Article
Language:English
Published: PeerJ Inc. 2025-05-01
Series:PeerJ Computer Science
Subjects:
Online Access:https://peerj.com/articles/cs-2866.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849728515654549504
author Ahmed Omar Alzahrani
Ahmed Mohammed Alghamdi
M. Usman Ashraf
Iqra Ilyas
Nadeem Sarwar
Abdulrahman Alzahrani
Alaa Abdul Salam Alarood
author_facet Ahmed Omar Alzahrani
Ahmed Mohammed Alghamdi
M. Usman Ashraf
Iqra Ilyas
Nadeem Sarwar
Abdulrahman Alzahrani
Alaa Abdul Salam Alarood
author_sort Ahmed Omar Alzahrani
collection DOAJ
description Variations in domain targets have recently posed significant challenges for facial expression recognition tasks, primarily due to domain shifts. Current methods focus largely on global feature adoption to achieve domain-invariant learning; however, transferring local features across diverse domains remains an ongoing challenge. Additionally, during training on target datasets, these methods often suffer from reduced feature representation in the target domain due to insufficient discriminative supervision. To tackle these challenges, we propose a dynamic cross-domain dual attention network for facial expression recognition. Our model is specifically designed to learn domain-invariant features through separate modules for global and local adversarial learning. We also introduce a semantic-aware module to generate pseudo-labels, which computes semantic labels from both global and local features. We assess our model’s effectiveness through extensive experiments on the Real-world Affective Faces Database (RAF-DB), FER-PLUS, AffectNet, Expression in the Wild (ExpW), SFEW 2.0, and Japanese Female Facial Expression (JAFFE) datasets. The results demonstrate that our scheme outperforms the existing state-of-the-art methods by attaining recognition accuracies 93.18, 92.35, 82.13, 78.37, 72.47, 70.68 respectively.
format Article
id doaj-art-e2f71710501f4a6b817edd472cd75893
institution DOAJ
issn 2376-5992
language English
publishDate 2025-05-01
publisher PeerJ Inc.
record_format Article
series PeerJ Computer Science
spelling doaj-art-e2f71710501f4a6b817edd472cd758932025-08-20T03:09:32ZengPeerJ Inc.PeerJ Computer Science2376-59922025-05-0111e286610.7717/peerj-cs.2866A novel facial expression recognition framework using deep learning based dynamic cross-domain dual attention networkAhmed Omar Alzahrani0Ahmed Mohammed Alghamdi1M. Usman Ashraf2Iqra Ilyas3Nadeem Sarwar4Abdulrahman Alzahrani5Alaa Abdul Salam Alarood6Department of Information Systems and Technology, College of Computer Science and Engineering, University of Jeddah, Jeddah, Makkah, Saudi ArabiaDepartment of Software Engineering, College of Computer Science and Engineering, University of Jeddah, Jeddah, Makkah, Saudi ArabiaDepartment of Computer Science, Government College Women University Sialkot, Sialkot, Punjab, PakistanDepartment of Computer Science, Government College Women University Sialkot, Sialkot, Punjab, PakistanDepartment of Computer Science, Bahria University, Lahore, Punjab, PakistanDepartment of Information Systems and Technology, College of Computer Science and Engineering, University of Jeddah, Jeddah, Makkah, Saudi ArabiaDepartment of Information Systems and Technology, College of Computer Science and Engineering, University of Jeddah, Jeddah, Makkah, Saudi ArabiaVariations in domain targets have recently posed significant challenges for facial expression recognition tasks, primarily due to domain shifts. Current methods focus largely on global feature adoption to achieve domain-invariant learning; however, transferring local features across diverse domains remains an ongoing challenge. Additionally, during training on target datasets, these methods often suffer from reduced feature representation in the target domain due to insufficient discriminative supervision. To tackle these challenges, we propose a dynamic cross-domain dual attention network for facial expression recognition. Our model is specifically designed to learn domain-invariant features through separate modules for global and local adversarial learning. We also introduce a semantic-aware module to generate pseudo-labels, which computes semantic labels from both global and local features. We assess our model’s effectiveness through extensive experiments on the Real-world Affective Faces Database (RAF-DB), FER-PLUS, AffectNet, Expression in the Wild (ExpW), SFEW 2.0, and Japanese Female Facial Expression (JAFFE) datasets. The results demonstrate that our scheme outperforms the existing state-of-the-art methods by attaining recognition accuracies 93.18, 92.35, 82.13, 78.37, 72.47, 70.68 respectively.https://peerj.com/articles/cs-2866.pdfArtificial intelligenceFacial expression recognitionDeep learningCross-domains
spellingShingle Ahmed Omar Alzahrani
Ahmed Mohammed Alghamdi
M. Usman Ashraf
Iqra Ilyas
Nadeem Sarwar
Abdulrahman Alzahrani
Alaa Abdul Salam Alarood
A novel facial expression recognition framework using deep learning based dynamic cross-domain dual attention network
PeerJ Computer Science
Artificial intelligence
Facial expression recognition
Deep learning
Cross-domains
title A novel facial expression recognition framework using deep learning based dynamic cross-domain dual attention network
title_full A novel facial expression recognition framework using deep learning based dynamic cross-domain dual attention network
title_fullStr A novel facial expression recognition framework using deep learning based dynamic cross-domain dual attention network
title_full_unstemmed A novel facial expression recognition framework using deep learning based dynamic cross-domain dual attention network
title_short A novel facial expression recognition framework using deep learning based dynamic cross-domain dual attention network
title_sort novel facial expression recognition framework using deep learning based dynamic cross domain dual attention network
topic Artificial intelligence
Facial expression recognition
Deep learning
Cross-domains
url https://peerj.com/articles/cs-2866.pdf
work_keys_str_mv AT ahmedomaralzahrani anovelfacialexpressionrecognitionframeworkusingdeeplearningbaseddynamiccrossdomaindualattentionnetwork
AT ahmedmohammedalghamdi anovelfacialexpressionrecognitionframeworkusingdeeplearningbaseddynamiccrossdomaindualattentionnetwork
AT musmanashraf anovelfacialexpressionrecognitionframeworkusingdeeplearningbaseddynamiccrossdomaindualattentionnetwork
AT iqrailyas anovelfacialexpressionrecognitionframeworkusingdeeplearningbaseddynamiccrossdomaindualattentionnetwork
AT nadeemsarwar anovelfacialexpressionrecognitionframeworkusingdeeplearningbaseddynamiccrossdomaindualattentionnetwork
AT abdulrahmanalzahrani anovelfacialexpressionrecognitionframeworkusingdeeplearningbaseddynamiccrossdomaindualattentionnetwork
AT alaaabdulsalamalarood anovelfacialexpressionrecognitionframeworkusingdeeplearningbaseddynamiccrossdomaindualattentionnetwork
AT ahmedomaralzahrani novelfacialexpressionrecognitionframeworkusingdeeplearningbaseddynamiccrossdomaindualattentionnetwork
AT ahmedmohammedalghamdi novelfacialexpressionrecognitionframeworkusingdeeplearningbaseddynamiccrossdomaindualattentionnetwork
AT musmanashraf novelfacialexpressionrecognitionframeworkusingdeeplearningbaseddynamiccrossdomaindualattentionnetwork
AT iqrailyas novelfacialexpressionrecognitionframeworkusingdeeplearningbaseddynamiccrossdomaindualattentionnetwork
AT nadeemsarwar novelfacialexpressionrecognitionframeworkusingdeeplearningbaseddynamiccrossdomaindualattentionnetwork
AT abdulrahmanalzahrani novelfacialexpressionrecognitionframeworkusingdeeplearningbaseddynamiccrossdomaindualattentionnetwork
AT alaaabdulsalamalarood novelfacialexpressionrecognitionframeworkusingdeeplearningbaseddynamiccrossdomaindualattentionnetwork