On the generalization of color texture-based face anti-spoofing

Abstract

Despite the significant attention given to the problem of face spoofing, we still lack generalized presentation attack detection (PAD) methods performing robustly in practical face recognition systems. The existing face anti-spoofing techniques have indeed achieved impressive results when trained and evaluated on the same database (i.e. intra-test protocols). Cross-database experiments have, however, revealed that the performance of the state-of-the-art methods drops drastically as they fail to cope with new attacks scenarios and other operating conditions that have not been seen during training and development phases. So far, even the popular convolutional neural networks (CNN) failed to derive well-generalizing features for face anti-spoofing. In this work, we explore the effects of different factors, such as acquisition condition and presentation attack instrument (PAI) variation, on the generalization of color texture-based face anti-spoofing. Our extensive cross-database evaluation of seven color texture-based methods demonstrate that most of the methods are unable to generalize to unseen spoofing attack scenarios. More importantly, the experiments show that some facial color texture representations are more robust to particular PAIs than others. From this observation, we propose a face PAD solution of attack-specific countermeasures based solely on color texture analysis and investigate how well it generalizes under display and print attacks in different conditions. The evaluation of the method combining attack-specific detectors on three benchmark face anti-spoofing databases showed remarkable generalization ability against display attacks while print attacks require still further attention.

Multi-Perspective Dynamic Features for Cross-Database Face Presentation Attack Detection

“…Second, a 2D discrete Fourier transform is applied to each frame of the residual noise video v NR , as formalized in Eq. (6). Then, the Fourier spectrum FS(u, w) can be obtained.…”

“…They can be categorized as methods based on static features and those based on dynamic features. Methods based on static features, such as color texture [6] and local shade [7], consider the features in a single frame. Methods based on dynamic information exploit unconscious facial motions, such as blinking [8] and micro-expressions [9].…”

Face liveness detection: fusing colour texture feature and deep feature

“…By learning facial microtextures of live faces, a method which expands the local binary pattern in time and space operator is applied to face anti-spoofing detection in [19]. Numerous recent works combine features of colour and texture to improve the accuracy of face liveness detection in [11,[20][21][22]. Recently, it is possible to use light field cameras for face spoofing detection.…”

Generalized Presentation Attack Detection: a face anti-spoofing evaluation proposal

“…Current face-PAD methods can be classified regarding the following standpoints: i) from the hardware used for data acquisition as rgb-only [11,21,26] or additional sensors [2,23] approaches; ii) from the required user interaction as active [14] or passive [16,26] methods; iii) from the input data type as single-frame [26] or video-based [1,22] approaches; iv) and, finally, depending on the feature extraction and classification strategy as hand-crafted [4,26] or deep learning [13,16]. Based on these classifications, we can depict that the most challenging scenario occurs when data is captured using rgb-only sensors using passive approaches that avoid any challenge-response interaction with the user (e.g.…”

“…Cross-FaceResolution. Building on top of some papers [4] that mention the influence of the distance between camera and subject, we propose a protocol to focus the analysis on the role of resolution of the region of the face in the task of detecting fake attempts. Two variants are possible for this protocol.…”