TU Darmstadt / ULB / TUbiblio

OrthoMAD: Morphing Attack Detection Through Orthogonal Identity Disentanglement

Neto, Pedro C. ; Goncalves, Tiago ; Huber, Marco ; Damer, Naser ; Sequeira, Ana F. ; Cardoso, Jaime S. (2022)
OrthoMAD: Morphing Attack Detection Through Orthogonal Identity Disentanglement.
21st International Conference of the Biometrics Special Interest Group (BIOSIG'22). Darmstadt, Germany (14.-16.09.2022)
doi: 10.1109/BIOSIG55365.2022.9897057
Konferenzveröffentlichung, Bibliographie

Kurzbeschreibung (Abstract)

Morphing attacks are one of the many threats that are constantly affecting deep face recognition systems. It consists of selecting two faces from different individuals and fusing them into a final image that contains the identity information of both. In this work, we propose a novel regularisation term that takes into account the existent identity information in both and promotes the creation of two orthogonal latent vectors. We evaluate our proposed method (OrthoMAD) in five different types of morphing in the FRLL dataset and evaluate the performance of our model when trained on five distinct datasets. With a small ResNet-18 as the backbone, we achieve state-of-the-art results in the majority of the experiments, and competitive results in the others.

Typ des Eintrags: Konferenzveröffentlichung
Erschienen: 2022
Autor(en): Neto, Pedro C. ; Goncalves, Tiago ; Huber, Marco ; Damer, Naser ; Sequeira, Ana F. ; Cardoso, Jaime S.
Art des Eintrags: Bibliographie
Titel: OrthoMAD: Morphing Attack Detection Through Orthogonal Identity Disentanglement
Sprache: Englisch
Publikationsjahr: 27 September 2022
Verlag: Gesellschaft für Informatik e.V.
Buchtitel: BIOSIG 2022: Proceedings of the 21st International Conference of the Biometrics Special Interest Group
Reihe: Lecture Notes in Informatics
Band einer Reihe: 329
Veranstaltungstitel: 21st International Conference of the Biometrics Special Interest Group (BIOSIG'22)
Veranstaltungsort: Darmstadt, Germany
Veranstaltungsdatum: 14.-16.09.2022
DOI: 10.1109/BIOSIG55365.2022.9897057
Kurzbeschreibung (Abstract):

Morphing attacks are one of the many threats that are constantly affecting deep face recognition systems. It consists of selecting two faces from different individuals and fusing them into a final image that contains the identity information of both. In this work, we propose a novel regularisation term that takes into account the existent identity information in both and promotes the creation of two orthogonal latent vectors. We evaluate our proposed method (OrthoMAD) in five different types of morphing in the FRLL dataset and evaluate the performance of our model when trained on five distinct datasets. With a small ResNet-18 as the backbone, we achieve state-of-the-art results in the majority of the experiments, and competitive results in the others.

Freie Schlagworte: Biometrics, Face recognition, Morphing attack, Deep learning, Machine learning
Fachbereich(e)/-gebiet(e): 20 Fachbereich Informatik
20 Fachbereich Informatik > Graphisch-Interaktive Systeme
Hinterlegungsdatum: 23 Nov 2022 08:21
Letzte Änderung: 28 Mär 2023 11:58
PPN: 506362469
Export:
Suche nach Titel in: TUfind oder in Google
Frage zum Eintrag Frage zum Eintrag

Optionen (nur für Redakteure)
Redaktionelle Details anzeigen Redaktionelle Details anzeigen