Due to the importance of the Morphing Attack, the development of new and accurate Morphing Attack Detection (MAD) systems is urgently needed by private and public institutions. In this context, D-MAD methods, i.e. detectors fed with a trusted live image and a probe tend to show better performance with respect to S-MAD approaches, that are based on a single input image. However, D-MAD methods usually leverage the identity of the two input face images only, and then present two main drawbacks: they lose performance when the two subjects look alike, and they do not consider potential artifacts left by the morphing procedure (which are instead typically exploited by S-MAD approaches). Therefore, in this paper, we investigate the combined use of D-MAD and S-MAD to improve detection performance through the fusion of the features produced by these two MAD approaches.

Di Domenico, N., Borghi, G., Franco, A., Maltoni, D. (2023). Combining Identity Features and Artifact Analysis for Differential Morphing Attack Detection [10.1007/978-3-031-43148-7_9].

Combining Identity Features and Artifact Analysis for Differential Morphing Attack Detection

Di Domenico, Nicolò;Borghi, Guido;Franco, Annalisa;Maltoni, Davide
2023

Abstract

Due to the importance of the Morphing Attack, the development of new and accurate Morphing Attack Detection (MAD) systems is urgently needed by private and public institutions. In this context, D-MAD methods, i.e. detectors fed with a trusted live image and a probe tend to show better performance with respect to S-MAD approaches, that are based on a single input image. However, D-MAD methods usually leverage the identity of the two input face images only, and then present two main drawbacks: they lose performance when the two subjects look alike, and they do not consider potential artifacts left by the morphing procedure (which are instead typically exploited by S-MAD approaches). Therefore, in this paper, we investigate the combined use of D-MAD and S-MAD to improve detection performance through the fusion of the features produced by these two MAD approaches.
2023
Image Analysis and Processing – ICIAP 2023
100
111
Di Domenico, N., Borghi, G., Franco, A., Maltoni, D. (2023). Combining Identity Features and Artifact Analysis for Differential Morphing Attack Detection [10.1007/978-3-031-43148-7_9].
Di Domenico, Nicolò; Borghi, Guido; Franco, Annalisa; Maltoni, Davide
File in questo prodotto:
File Dimensione Formato  
ICIAP+2023.pdf

Open Access dal 06/09/2024

Tipo: Postprint
Licenza: Licenza per accesso libero gratuito
Dimensione 1.29 MB
Formato Adobe PDF
1.29 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11585/943323
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 5
  • ???jsp.display-item.citation.isi??? ND
social impact