Face anti‐spoofing with refined triplet loss and multi‐level attention constraint network

Abstract One critical issue for existing face recognition (FR) systems is to ensure its accuracy and robustness, which calls for the development of face anti‐spoofing (FAS) algorithms to work against presentation attacks (PA). This letter proposes a novel Multi‐level Attention Constraint Network wit...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Xingzhong Nong, Ying Zeng, Haifeng Hu
Formato:	article
Lenguaje:	EN
Publicado:	Wiley 2021
Materias:	Electrical engineering. Electronics. Nuclear engineering TK1-9971
Acceso en línea:	https://doaj.org/article/44b610422661466cac4a3d7878820c2b
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:44b610422661466cac4a3d7878820c2b
record_format	dspace
spelling	oai:doaj.org-article:44b610422661466cac4a3d7878820c2b2021-11-19T05:42:53ZFace anti‐spoofing with refined triplet loss and multi‐level attention constraint network1350-911X0013-519410.1049/ell2.12305https://doaj.org/article/44b610422661466cac4a3d7878820c2b2021-11-01T00:00:00Zhttps://doi.org/10.1049/ell2.12305https://doaj.org/toc/0013-5194https://doaj.org/toc/1350-911XAbstract One critical issue for existing face recognition (FR) systems is to ensure its accuracy and robustness, which calls for the development of face anti‐spoofing (FAS) algorithms to work against presentation attacks (PA). This letter proposes a novel Multi‐level Attention Constraint Network with a Refined Triplet Loss (MACN‐RTL) for the task of FAS. Specifically, an MACN which consists of two components is designed, that is, Multi‐level Attention Network (MAN) and Distribution Constraint (DC). MAN aims to exploit effective information from different levels, while DC helps to learn a more compact and discriminative feature embedding for classification. Besides, a Refined Triplet Loss for better model optimisation is devised. Compared with existing FAS works, MACN is designed for better feature extraction and a better solution for optimisation by RTL. Experimental results demonstrate the superiority of the proposed approach. i.To ensure the security of FR systems towards PA, a novel MACN‐RTL is proposed, which can generate a more informative and discriminative feature embedding for accurate classification. ii.The designed MACN leverages attention mechanisms to obtain effective representations and reduces the distributional discrepancy of cross‐domain samples. iii.An RTL is devised to refine the widely used triplet loss by adding a refinement term to achieve a better optimisation of the model.Xingzhong NongYing ZengHaifeng HuWileyarticleElectrical engineering. Electronics. Nuclear engineeringTK1-9971ENElectronics Letters, Vol 57, Iss 24, Pp 912-914 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Electrical engineering. Electronics. Nuclear engineering TK1-9971
spellingShingle	Electrical engineering. Electronics. Nuclear engineering TK1-9971 Xingzhong Nong Ying Zeng Haifeng Hu Face anti‐spoofing with refined triplet loss and multi‐level attention constraint network
description	Abstract One critical issue for existing face recognition (FR) systems is to ensure its accuracy and robustness, which calls for the development of face anti‐spoofing (FAS) algorithms to work against presentation attacks (PA). This letter proposes a novel Multi‐level Attention Constraint Network with a Refined Triplet Loss (MACN‐RTL) for the task of FAS. Specifically, an MACN which consists of two components is designed, that is, Multi‐level Attention Network (MAN) and Distribution Constraint (DC). MAN aims to exploit effective information from different levels, while DC helps to learn a more compact and discriminative feature embedding for classification. Besides, a Refined Triplet Loss for better model optimisation is devised. Compared with existing FAS works, MACN is designed for better feature extraction and a better solution for optimisation by RTL. Experimental results demonstrate the superiority of the proposed approach. i.To ensure the security of FR systems towards PA, a novel MACN‐RTL is proposed, which can generate a more informative and discriminative feature embedding for accurate classification. ii.The designed MACN leverages attention mechanisms to obtain effective representations and reduces the distributional discrepancy of cross‐domain samples. iii.An RTL is devised to refine the widely used triplet loss by adding a refinement term to achieve a better optimisation of the model.
format	article
author	Xingzhong Nong Ying Zeng Haifeng Hu
author_facet	Xingzhong Nong Ying Zeng Haifeng Hu
author_sort	Xingzhong Nong
title	Face anti‐spoofing with refined triplet loss and multi‐level attention constraint network
title_short	Face anti‐spoofing with refined triplet loss and multi‐level attention constraint network
title_full	Face anti‐spoofing with refined triplet loss and multi‐level attention constraint network
title_fullStr	Face anti‐spoofing with refined triplet loss and multi‐level attention constraint network
title_full_unstemmed	Face anti‐spoofing with refined triplet loss and multi‐level attention constraint network
title_sort	face anti‐spoofing with refined triplet loss and multi‐level attention constraint network
publisher	Wiley
publishDate	2021
url	https://doaj.org/article/44b610422661466cac4a3d7878820c2b
work_keys_str_mv	AT xingzhongnong faceantispoofingwithrefinedtripletlossandmultilevelattentionconstraintnetwork AT yingzeng faceantispoofingwithrefinedtripletlossandmultilevelattentionconstraintnetwork AT haifenghu faceantispoofingwithrefinedtripletlossandmultilevelattentionconstraintnetwork
_version_	1718420370449498112

Face anti‐spoofing with refined triplet loss and multi‐level attention constraint network

Ejemplares similares