Using Visual Rhythms for Detecting Video-Based Facial Spoof Attacks

Abstract

Spoofing attacks or impersonation can be easily accomplished in a facial biometric system wherein users without access privileges attempt to authenticate themselves as valid users, in which an impostor needs only a photograph or a video with facial information of a legitimate user. Even with recent advances in biometrics, information forensics and security, vulnerability of facial biometric systems against spoofing attacks is still an open problem. Even though several methods have been proposed for photo-based spoofing attack detection, attacks performed with videos have been vastly overlooked, which hinders the use of the facial biometric systems in modern applications. In this paper, we present an algorithm for video-based spoofing attack detection through the analysis of global information which is invariant to content, since we discard video contents and analyze contentindependent noise signatures present in the video related to the unique acquisition processes. Our approach takes advantage of noise signatures generated by the recaptured video to distinguish between fake and valid access videos. For that, we use the Fourier spectrum followed by the computation of video visual rhythms and the extraction of different characterization methods. For evaluation, we consider the novel Unicamp Video-Attack Database (UVAD) which comprises 17,076 videos composed of real access and spoofing attack videos. In addition, we evaluate the proposed method using the Replay-Attack Database, which contains photo-based and video-based face spoofing attacks.

Using LDP-TOP in Video-Based Spoofing Detection

“…We validate the effectiveness of the proposed method on the Unicamp Video-Attack Database (UVAD) [11]. This dataset contains valid access and attempted attack videos of 404 different identities.…”

“…Since our method is dedicated to noise analysis, we compare our results mainly with the two recent and closely related works: Visual Rhythm (VR) [11] and Visual Codebooks (VC) [10]. VR based approach captures noise signatures in terms of 2D maps which are basically generated by traversing the spectrum video in horizontal, vertical and zigzag directions.…”

“…This can be considered as a baseline method in UVAD dataset. The authors in [11] report results in a different dataset configuration where the size of testing set is much smaller than our consideration which is described in the previous section. In order to make more thorough and coherent analysis, we run visual rhythm extraction on our dataset configuration by using the implementation provided by the authors.…”

“…A number of approaches have been proposed [14,9,8,2,6,11,10], and we classify them into two main categories: spatial domain and frequency domain analysis.…”

“…In [6], the authors extract Moiré patterns from still images using a bandpass filter and make analysis on frequency domain. A large-scale dataset dedicated to video-based attacks is introduced in [11]. This is a challenging dataset containing huge number of videos recorded under different environmental conditions.…”