Espace des ressources électronique en libre accès
Bibliothèque de la Faculté SNVST de L'UAMOB
Auteur Mahdi Mahdi Ghafourian
|
Documents disponibles écrits par cet auteur (1)
Faire une suggestion Affiner la recherche Interroger des sources externes
Titre : |
PAD-Phys: Exploiting Physiology for Presentation Attack Detection in Face Biometrics |
Type de document : |
document électronique |
Auteurs : |
Luis F. Gomez, Auteur ; Julian Fierrez, Auteur ; Mahdi Mahdi Ghafourian, Auteur ; Aythami Morales, Auteur |
Editeur : |
[S.l.] : IEEE (Institute of Electrical and Elctronic Engineers) |
Année de publication : |
2023 |
Note générale : |
Index Terms—Remote photoplethysmography, Presentation at-
tacks detection, Convolutional Attention Network |
Langues : |
Anglais (eng) |
Catégories : |
571 Physiologie générale et sujets connexes Physiology and related subjects
|
Index. décimale : |
571 |
Résumé : |
Presentation Attack Detection (PAD) is a crucial stage in facial recognition systems to avoid leakage of personal information or spoofing of identity to entities. Recently, pulse detection based on remote photoplethysmography (rPPG) has been shown to be effective in face presentation attack detection.
This work presents three different approaches to the presentation attack detection based on rPPG: (i) The physiological domain, a domain using rPPG-based models, (ii) the Deepfakes domain, a domain where models were retrained from the physiological domain to specific Deepfakes detection tasks; and (iii) a new Presentation Attack domain was trained by applying transfer learning from the two previous domains to improve the capability to differentiate between bona-fides and attacks.
The results show the efficiency of the rPPG-based models for presentation attack detection, evidencing a 21.70% decrease in average classification error rate (ACER) (from 41.03% to 19.32%) when the presentation attack domain is compared to the physiological and Deepfakes domains. Our experiments highlight the efficiency of transfer learning in rPPG-based models and perform well in presentation attack detection in instruments that do not allow copying of this physiological feature. |
En ligne : |
https://arxiv.org/abs/2310.02140 |
Format de la ressource électronique : |
PDF |
PAD-Phys: Exploiting Physiology for Presentation Attack Detection in Face Biometrics [document électronique] / Luis F. Gomez, Auteur ; Julian Fierrez, Auteur ; Mahdi Mahdi Ghafourian, Auteur ; Aythami Morales, Auteur . - [S.l.] : IEEE (Institute of Electrical and Elctronic Engineers), 2023. Index Terms—Remote photoplethysmography, Presentation at-
tacks detection, Convolutional Attention Network Langues : Anglais ( eng)
Catégories : |
571 Physiologie générale et sujets connexes Physiology and related subjects
|
Index. décimale : |
571 |
Résumé : |
Presentation Attack Detection (PAD) is a crucial stage in facial recognition systems to avoid leakage of personal information or spoofing of identity to entities. Recently, pulse detection based on remote photoplethysmography (rPPG) has been shown to be effective in face presentation attack detection.
This work presents three different approaches to the presentation attack detection based on rPPG: (i) The physiological domain, a domain using rPPG-based models, (ii) the Deepfakes domain, a domain where models were retrained from the physiological domain to specific Deepfakes detection tasks; and (iii) a new Presentation Attack domain was trained by applying transfer learning from the two previous domains to improve the capability to differentiate between bona-fides and attacks.
The results show the efficiency of the rPPG-based models for presentation attack detection, evidencing a 21.70% decrease in average classification error rate (ACER) (from 41.03% to 19.32%) when the presentation attack domain is compared to the physiological and Deepfakes domains. Our experiments highlight the efficiency of transfer learning in rPPG-based models and perform well in presentation attack detection in instruments that do not allow copying of this physiological feature. |
En ligne : |
https://arxiv.org/abs/2310.02140 |
Format de la ressource électronique : |
PDF |
|