Matrix-Regularized One-Class Multiple Kernel Learning for Unseen Face Presentation Attack Detection

• Published in: IEEE transactions on information forensics and security Vol. 16; pp. 4635 - 4647
• Main Author: Arashloo, Shervin Rahimzadeh
• Format: Journal Article
• Language: English
• Published: New York IEEE 2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Classification; Color imagery; Face recognition; Faces; Kernel; Kernels; Machine learning; matrix regularisation; multiple kernel learning; one-class Fisher null projection; Optimization; Regularization; Saddle points; Subsystems; Support vector machines; Task analysis; Training; Unseen face presentation attack detection; zero-shot learning
• ISSN: 1556-6013; 1556-6021
• DOI: 10.1109/TIFS.2021.3111766

The functionality of face biometric systems is severely challenged by presentation attacks (PA's), and especially those attacks that have not been available during the training phase of a PA detection (PAD) subsystem. Among other alternatives, the one-class classification (OCC) paradigm is an applicable strategy that has been observed to provide good generalisation against unseen attacks. Following an OCC approach for the unseen face PAD from RGB images, this work advocates a matrix-regularised multiple kernel learning algorithm to make use of several sources of information each constituting a different view of the face PAD problem. In particular, drawing on the one-class null Fisher classification principle, we characterise different deep CNN representations as kernels and propose a multiple kernel learning (MKL) algorithm subject to an (<inline-formula> <tex-math notation="LaTeX">r,p </tex-math></inline-formula>)-norm (<inline-formula> <tex-math notation="LaTeX">1\leq r,p </tex-math></inline-formula>) matrix regularisation constraint. The propose MKL algorithm is formulated as a saddle point Lagrangian optimisation task for which we present an effective optimisation algorithm with guaranteed convergence. An evaluation of the proposed one-class MKL algorithm on both general object images in an OCC setting as well as on different face PAD datasets in an unseen zero-shot attack detection setting illustrates the merits of the proposed method compared to other one-class multiple kernel and deep end-to-end CNN-based methods.