Among tangible threats facing current biometric systems are spoofing attacks. A spoofing attack occurs when a person tries to masquerade as someone else by falsifying data and thereby attempting to gain illegitimate access and advantages. Recently, an increasing attention has been given to this research problem, as can be attested by the growing number of articles and the various competitions that appear in major biometric forums. This study presents a comprehensive overview of the recent advances in face anti-spoofing state-of-the-art, discussing existing methodologies, available benchmarking databases, reported results and, more importantly, the open issues and future research directions. As a case study for illustration, a face anti-spoofing method is described, which employs a colour local binary pattern descriptor to jointly analyse colour and texture available from the luminance and chrominance channels. Two publicly available databases are used for the analysis, and the importance of inter-database evaluation to attest the generalisation capabilities of an anti-spoofing method is discussed.

The vulnerability of deep-learning-based face-recognition (FR) methods, to presentation attacks (PA), is studied in this study. Recently, proposed FR methods based on deep neural networks (DNN) have been shown to outperform most other methods by a significant margin. In a trustworthy face-verification system, however, maximising recognition-performance alone is not sufficient – the system should also be capable of resisting various kinds of attacks, including PA. Previous experience has shown that the PA vulnerability of FR systems tends to increase with face-verification accuracy. Using several publicly available PA datasets, the authors show that DNN-based FR systems compensate for variability between bona fide and PA samples, and tend to score them similarly, which makes such FR systems extremely vulnerable to PAs. Experiments show the vulnerability of the studied DNN-based FR systems to be consistently higher than 90%, and often higher than 98%.

Securing face recognition systems against spoofing attacks have been recognised as a real challenge. Spoofing attacks are conducted by printing or displaying a digital acquisition of a capture subject (target user) in front of the sensor. These extra reproduction stages generate colour distortions between face artefacts and real faces. In this work, the problem of spoof detection is addressed by modelling the radiometric distortions generated by the recapturing process. The spoof detection process takes advantage of enrolment data and occurs after face identification so that for each client the authors have at disposal at least one genuine face sample as a reference. Once identified, they compute the colour transformation between the observed face and its enrolment counterpart. A compact parametric representation is proposed to model those radiometric transforms and it is used as features for classification. They evaluate the proposed method on Replay-Attack, CASIA and MSU public databases and show its competitiveness with state-of-the-art countermeasures. Limitations of the proposed method are clearly identified and discussed through experiments in adversary evaluation conditions where colour distortions are not only generated by the recapturing process but also by natural illumination variations.

Face recognition systems are becoming ubiquitous, but they are vulnerable to spoofing attacks. The recently available light field cameras can be used for spoofing attack detection. In this study, the IST Lenslet Light Field Face Spoofing Database (IST LLFFSD) is proposed, consisting of 100 genuine images, from 50 subjects, captured with a Lytro ILLUM lenslet light field camera, and a set of 600 face spoofing attack images, captured using the same camera. The IST LLFFSD simulates six different types of presentation attacks, including printed paper, wrapped printed paper, laptop, tablet and two different mobile phones. This study also proposes a novel spoofing attack detection solution, based on a compact, yet effective, descriptor exploiting the colour and texture variations associated with the different directions of light captured in light field images. Extensive experiments show very effective results, with the proposed solution performing better than state-of-the-art alternatives for the face spoofing attack types considered.

Facial hyperspectral image analysis has become a popular topic since it provides additional spectral information on subjects unlike the 2D face imagery which only has spatial information, hence it has an opportunity to improve face recognition accuracy. Three new methods for feature extraction for facial hyperspectral image classification are proposed. The methods employ a three-dimensional discrete wavelet transform (3D-DWT) to extract features from facial hyperspectral images. One of the advantages of 3D-DWT for feature extraction in hyperspectral images is that the horizontal, vertical and spectral information are processed in parallel. The most important characteristic of 3D-DWT is decomposing hyperspectral images into a set of spatio-spectral frequency subbands. The study proposes three methods using 3D-DWT for feature extraction: 3D-subband energy, 3D-subband overlapping cube and 3D-global energy. The k-NN and collaborative representation-based classifier (CRC) are used to process extracted feature vector datasets, where classification accuracies are evaluated by four test scenarios. The results under different test scenarios revealed that accuracy of proposed 3D-DWT methods is superior to alternative methods using spatio-spectral classification.

The recent significant progress in face recognition is mainly achieved using learning-based (LE) techniques via an exhaustive training involving a huge number of face samples. However, in many applications, the number of face images available for training may be very limited. This makes LE techniques impractical for learning discriminative features and models. Thus, limited number of face samples (i.e. scarce data) degrades the recognition performance of most existing methods. To overcome this problem, the authors propose a novel approach based on two-layer collaborative representation to exploit the abundance of samples in some classes to enrich the scarce data in other classes. The first-layer collaborative representation uses the abundance of samples to construct representations for the scarce data. Then, a new face sample is recognised by computing residuals with the second-layer collaborative representation. Extensive experiments on four benchmark face databases demonstrate the effectiveness of their proposed approach which compares favourably against state-of-the-art methods.

Face verification is a problem approached in the literature mainly using non-linear class-specific subspace learning techniques. While it has been shown that kernel-based class-specific discriminant analysis is able to provide excellent performance in small- and medium-scale face verification problems, its application in today's large-scale problems is difficult due to its training space and computational requirements. In this study, generalising on kernel-based class-specific discriminant analysis, it is shown that class-specific subspace learning can be cast as a regression problem. This allows them to derive linear, (reduced) kernel and neural network-based class-specific discriminant analysis methods using efficient batch and/or iterative training schemes, suited for large-scale learning problems. The authors test the performance of these methods in two datasets describing medium- and large-scale face verification problems.

A biometric system uses the physical or behavioural attributes of a person, such as face, fingerprint, iris or voice, to recognise an individual. Many operational biometric systems store the biographic information of an individual, viz., name, gender, age and ethnicity, besides the biometric data itself. Thus, the biometric record pertaining to an individual consists of both biometric data and biographic data. We propose the use of a graph structure to model the relationship between the biometric records in a database. We show the benefits of such a graph in deducing biographic labels of incomplete records, i.e. records that may have missing biographic information. In particular, we use a label propagation scheme to deduce missing values for both binary-valued biographic attributes (e.g. gender) as well as multi-valued biographic attributes (e.g. age group). Experimental results using face-based biometric records consisting of name, age, gender and ethnicity convey the pros and cons of the proposed method.

Convolutional neural network (CNN) based approaches are the state of the art in various computer vision tasks including face recognition. Considerable research effort is currently being directed toward further improving CNNs by focusing on model architectures and training techniques. However, studies systematically exploring the strengths and weaknesses of existing deep models for face recognition are still relatively scarce. In this paper, we try to fill this gap and study the effects of different covariates on the verification performance of four recent CNN models using the Labelled Faces in the Wild dataset. Specifically, we investigate the influence of covariates related to image quality and model characteristics, and analyse their impact on the face verification performance of different deep CNN models. Based on comprehensive and rigorous experimentation, we identify the strengths and weaknesses of the deep learning models, and present key areas for potential future research. Our results indicate that high levels of noise, blur, missing pixels, and brightness have a detrimental effect on the verification performance of all models, whereas the impact of contrast changes and compression artefacts is limited. We find that the descriptor-computation strategy and colour information does not have a significant influence on performance.