Deep Face Representations for Differential Morphing Attack Detection

Abstract

The vulnerability of facial recognition systems to face morphing attacks is well known. Many different approaches for morphing attack detection (MAD) have been proposed in the scientific literature. However, the MAD algorithms proposed so far have mostly been trained and tested on datasets whose distributions of image characteristics are either very limited (e.g., only created with a single morphing tool) or rather unrealistic (e.g., no print-scan transformation). As a consequence, these methods easily overfit on certain image types and the results presented cannot be expected to apply to real-world scenarios. For example, the results of the latest NIST FRVT MORPH show that the majority of submitted MAD algorithms lacks robustness and performance when considering unseen and challenging datasets. In this work, subsets of the FERET and FRGCv2 face databases are used to create a realistic database for training and testing of MAD algorithms, containing a large number of ICAO-compliant bona fide facial images, corresponding unconstrained probe images, and morphed images created with four different face morphing tools. Furthermore, multiple post-processings are applied on the reference images, e.g., print-scan and JPEG2000 compression. On this database, previously proposed differential morphing algorithms are evaluated and compared. In addition, the application of deep face representations for differential MAD algorithms is investigated. It is shown that algorithms based on deep face representations can achieve very high detection performance (less than 3% D-EER) and robustness with respect to various post-processings. Finally, the limitations of the developed methods are analyzed.

Face Morphing Attack Generation and Detection: A Comprehensive Survey

“…This is a private database that consists of both digital and print-scan samples of morphed images and is composed of 964 + 964 + 529 morphed face images generated from subjects contained in the FRGCv2 and FERET databases. Another database by Scherhag et al [110] employs landmark-based morph generation techniques that include OpenCV, FaceMorphed, FaceFusion and the UBO morphing method. This database consists of approximately 791+3246 morphed face images from the FERET and FRGCv2 databases.…”

Makeup Presentation Attacks: Review and Detection Performance Benchmark

“…Differences which indicate M-PAs can be learned in a training stage employing a machine learningbased classifier. Differential attack detection systems have already been successfully proposed for detection of face morphing [50] and facial retouching [51]. Therefore, main focus is put on differential M-PAD.…”

“…In contrast, differential detection methods aiming to unveil digital image manipulation, e.g. [50], [51], consider the probe image to be a trusted live capture. In rare scenarios, a differential M-PAD might not be possible, i.e.…”

Improving Data Security, Privacy, and Interoperability for the IEEE Biometric Open Protocol Standard

“…Some of these works also address authentication, network security, privacy data, and processing power. Unfortunately, some of them also have a few drawbacks [22]- [27]. Moreover, even with the guidelines of the IEEE Biometric Open Protocol Standard (IEEE BOPS) [28], it is not possible to use each of these schemes in different biometric databases and make them communicate in a way that is reasonably security proof and does not compromise a person's privacy.…”