Bio-Inspired Presentation Attack Detection for Face Biometrics

Abstract

Today, face biometric systems are becoming widely accepted as a standard method for identity authentication in many security settings. For example, their deployment in automated border control gates plays a crucial role in accurate document authentication and reduced traveler flow rates in congested border zones. The proliferation of such systems is further spurred by the advent of portable devices. On the one hand, modern smartphone and tablet cameras have in-built user authentication applications while on the other hand, their displays are being consistently exploited for face spoofing. Similar to biometric systems of other physiological biometric identifiers, face biometric systems have their own unique set of potential vulnerabilities. In this work, these vulnerabilities (presentation attacks) are being explored via a biologically-inspired presentation attack detection model which is termed "BIOPAD." Our model employs Gabor features in a feedforward hierarchical structure of layers that progressively process and train from visual information of people's faces, along with their presentation attacks, in the visible and near-infrared spectral regions. BIOPAD's performance is directly compared with other popular biologically-inspired layered models such as the "Hierarchical Model And X" (HMAX) that applies similar handcrafted features, and Convolutional Neural Networks (CNN) that discover low-level features through stochastic descent training. BIOPAD shows superior performance to both HMAX and CNN in all of the three presentation attack databases examined and these results were consistent in two different classifiers (Support Vector Machine and k-nearest neighbor). In certain cases, our findings have shown that BIOPAD can produce authentication rates with 99% accuracy. Finally, we further introduce a new presentation attack database with visible and near-infrared information for direct comparisons. Overall, BIOPAD's operation, which is to fuse information from different spectral bands at both feature and score levels for the purpose of face presentation attack detection, has never been attempted before with a biologically-inspired algorithm. Obtained detection rates are promising and confirm that near-infrared visual information significantly assists in overcoming presentation attacks.

Keywords: anti-spoofing; biologically-inspired biometrics; face biometrics; multiple sensor fusion; presentation attack detection.

Figure 1

Examples of on-center and off-center receptive fields for color opponency channels. Plus sign indicates whether the particular color is on and the minus off.

Figure 2

The proposed model structure. Several layers L1 to L5 progressively process spatial and spectral facial features. All participants gave written informed consent for the publication of this manuscript.

Figure 3

A genuine access attempt vs. a photo-print attack. Top row shows the progressive process of a genuine photo attempt. Bottom row shows the printed photo attack. Column (A) shows the input layer images. Column (B) the L2a layer as processed from edge detection Gabor filters, column (C) the L2b layer processed from texture grating cells and column (D) the combined layers L2a and L2b after spatial summation. The richness and depth of edge-texture information in the original image (top row) is apparent. All participants gave written informed consent for the publication of this manuscript.

Figure 4

An example of a subject from the FRAV “attack” database. Top row left to right: Genuine access RGB photo, RGB Printed photo attack, RGB printed mask attack, RGB printed mask with eyes exposed attack, RGB tablet attack. Bottom row left to right: Genuine access NIR photo, NIR printed photo attack, NIR printed mask attack, NIR printed mask with eyes exposed attack, NIR tablet attack. All participants gave written informed consent for the publication of this manuscript.

Figure 5

L4 vectors visualized with t-SNE for the three datasets. (A) real vs. impostors–CASIA database, (B) presentation attacks—CASIA database, (C) real vs. impostors –MFSD database, (D) presentation attacks—MFSD database (E) real vs. impostors—FRAV “attack” database, and (F) presentation attacks—FRAV “attack” database.

Figure 6

HMAX vectors visualized with t-SNE for the three datasets in terms of real access attempts vs. impostors. (A) t-SNE for the CASIA dataset, (B) t-SNE for the MFSD dataset, and (C) t-SNE for the FRAV “attack” dataset.

Figure 7

L4 vectors visualized with t-SNE for the FRAV “attack” database and its NIR information. (A) real vs. impostors—FRAV “attack” database with NIR information only, and (B) presentation attacks—FRAV “attack” database with NIR information only, (C) real vs. impostors—FRAV “attack” database with RGB&NIR information fused at feature level, and (D) presentation attacks—FRAV “attack” database with RGB&NIR information fused at feature level.

Figure 8

BIOPAD Detection Error Tradeoff curves of SVM linear classifier for the FRAV “attack” database in NIR(red), RGB + NIR at feature level (blue) and RGB (green). Attack Presentation Error Rate—APER.