Face spoofing detection ensemble via multistage optimisation and pruning

• Published in: Pattern recognition letters 2022-06, Vol.158, p.1-8
• Main Authors: Fatemifar, Soroush, Asadi, Shahrokh, Awais, Muhammad, Akbari, Ali, Kittler, Josef
• Format: Article
• Language: English
• Subjects: Anomalies; Anomaly detection; Client-specific information; Convolutional neural networks; Ensemble of one-class classifiers; Ensemble pruning; Face; Face recognition; Face spoofing detection; Genetic algorithms; Optimization; Pattern recognition; Pattern search; Spoofing
• ISSN: 0167-8655; 1872-7344
• DOI: 10.1016/j.patrec.2022.04.006

•We develop a solution for the face spoofing detection problem by fusing multiple anomaly experts using Weighted Averaging(WA).•We propose a novel three-stage optimisation approach to improve the generalisation capability and accuracy of the WA fusion.•We define a new score normalisation approach to support multiple anomaly detectors fusion.•We define an effective criterion to prune the WA to achieve better classification result and generalisation performance.•We experimentally demonstrate that the proposed anomaly-based WA achieves superior performance over state-of-theart methods. Despite the recent improvements in facial recognition, face spoofing attacks can still pose a serious security threat to biometric systems. As fraudsters are coming up with novel spoofing attacks, anomaly-based detectors, compared to the binary spoofing attack counterparts, have certain generalisation performance advantages. In this work, we investigate the merits of fusing multiple anomaly classifiers using weighted averaging (WA) fusion. The design of the entire system is based on genuine-access data only. To optimise the parameters of WA, we propose a novel three-stage optimisation method with the following contributions: (a) A new hybrid optimisation method using Genetic Algorithm (GA) and Pattern Search (PS) to explore the weight space more effectively (b) a novel two-sided score normalisation method to improve the anomaly detection performance (c) a new ensemble pruning method to improve the generalisation performance. To further boost the performance of the proposed anomaly detection ensemble, we incorporate client-specific information to train the proposed model. We evaluate the capability of the proposed model on publicly available face spoofing databases including Replay-Attack, Replay-Mobile and Rose-Youtu. The experimental results demonstrate that the proposed WA fusion outperforms the state-of-the-art anomaly-based and multiclass approaches.