Abstract Privacy regulations and ethical concerns have encouraged the use of privacy-friendly synthetic data for the training of facial analysis systems. However, the automated generation of images depicting the same synthetic subject in different environmental scenarios remains challenging, as identity-related features may not be accurately preserved. This is a severe issue for the training of differential morphing attack detection (MAD) algorithms, where subtle differences in facial features can indicate morphing attacks. This work introduces IDSwapMAD as a new way for generating privacy-friendly training data for differential MAD methods. In detail, a generative adversarial network is employed to generate synthetic facial images of which the faces are swapped with pairs of real reference and probe images containing variations that mimic a border control scenario. In this way, style-related properties of the reference and probe images are retained, while identity-related features are replaced. It is shown that the proposed IDSwapMAD technique is an effective and privacy-friendly strategy for training differential MAD methods, whose detection performance is on par with a state-of-the-art MAD method trained on real data.

Biometric recognition is widely used, making the privacy and security of extracted templates a critical concern. Biometric Template Protection schemes, especially those utilizing Homomorphic Encryption, introduce significant computational challenges due to increased workload. Recent advances in deep neural networks have enabled state-of-the-art feature extraction for face, fingerprint, and iris modalities. The ubiquity and affordability of biometric sensors further facilitate multi-modal fusion, which can enhance security by combining features from different modalities. This work investigates the biometric performance of reduced multi-biometric template sizes. Experiments are conducted on an in-house virtual multi-biometric database, derived from DNN-extracted features for face, fingerprint, and iris, using the FRGC, MCYT, and CASIA databases. The evaluated approaches are (i) explainable and straightforward to implement under encryption, (ii) training-free, and (iii) capable of generalization. Dimensionality reduction of feature vectors leads to fewer operations in the Homomorphic Encryption (HE) domain, enabling more efficient encrypted processing while maintaining biometric accuracy and security at a level equivalent to or exceeding single-biometric recognition. Our results demonstrate that, by fusing feature vectors from multiple modalities, template size can be reduced by 67 % with no loss in Equal Error Rate (EER) compared to the best-performing single modality.

This paper analyses and addresses the performance gap in the fuzzy vault-based Biometric Cryptosystem (BCS). We identify unstable error correction capabilities, which are caused by variable feature set sizes and their influence on similarity thresholds, as a key source of performance degradation. This issue is further magnified by information loss introduced through feature type transformations. To address both problems, we propose a novel feature quantization method based on equal frequent intervals. This method guarantees fixed feature set sizes and supports training-free adaptation to any number of intervals. The proposed approach significantly reduces the performance gap introduced by template protection. Additionally, it integrates seamlessly with existing systems to minimize the negative effects of feature transformation. Experiments on state-of-the-art face, fingerprint, and iris recognition systems confirm that only minimal performance degradation remains, demonstrating the effectiveness of the method across major biometric modalities.

The Fingerprint Liveness Detection Competition (LivDet) is a recurring benchmark series that evaluates the effectiveness of software-based Presentation Attack Detection (PAD) algorithms in fingerprint recognition. LivDet2025 presents three challenges: (1) "Liveness Detection in Action", requiring the integration of PAD with user-specific recognition; (2) "Fingerprint Representation", evaluating the compactness and discriminability of feature vectors; and (3) "Adversarial Robustness", assessing the resilience of PADs to adversarially-crafted presentation attack instruments. This edition marks a significant milestone with the inclusion of contactless fingerprint data, promoting interoperability and robustness across acquisition technologies. Furthermore, no training data was provided; participants must select and declare external datasets for model development. The competition was open to academic and industrial research groups, with all submitted algorithms evaluated on common datasets and under standardized protocols. LivDet2025 aims to provide a comprehensive assessment of PAD performance under realistic, multi-sensor, and multi-attack scenarios. Results reveal important trade-offs between PAD accuracy, usability, and computational efficiency. For instance, some systems achieved high presentation attack rejection at the cost of extremely high false rejection rates, while others optimised speed and generalizability but exhibited limited attack resilience.

Abstract Biometric cryptosystems enable privacy-preserving authentication using biometric data, such as fingerprints or iris scans. However, single modalities suffer from limited entropy, impacting both recognition performance and security. This work investigates the fusion of multiple biometric characteristics in a Deep Multi-biometric Fuzzy Commitment Scheme. In the experimental setup, we demonstrate how Deep Convolutional Neural Networks (DCNNs) are used to tackle the challenge of non-uniform representations by generating uniform embeddings. Uni-modal databases of iris and fingerprint embeddings, as well as the corresponding multi-biometric database, are employed for this purpose. Three fusion methods are proposed: concatenation, interleaving, and random shuffling within the fuzzy commitment scheme using error correction methods based on Hadamard and Reed-Solomon codes. The evaluation of performance and security reveals that random shuffling outperforms other methods like interleaving and concatenation in terms of recognition performance. Concatenation displayed the lowest performance. Finally, the findings are summarized and potential improvements are discussed.

Face recognition systems are widely deployed in high-security applications such as for biometric verification at border controls. Despite their high accuracy on pristine data, it is well-known that digital manipulations, such as face morphing, pose a security threat to face recognition systems. Malicious actors can exploit the facilities offered by the identity document issuance process to obtain identity documents containing morphed images. Thus, subjects who contributed to the creation of the morphed image can with high probability use the identity document to bypass automated face recognition systems. In recent years, no-reference (i.e., single image) and differential morphing attack detectors have been proposed to tackle this risk. These systems are typically evaluated in isolation from the face recognition system that they have to operate jointly with and do not consider the face recognition process. Contrary to most existing works, we present a novel method for adapting deep learning-based face recognition systems to be more robust against face morphing attacks. To this end, we introduce TetraLoss, a novel loss function that learns to separate morphed face images from its contributing subjects in the embedding space while still achieving high biometric verification performance. In a comprehensive evaluation, we show that the proposed method can significantly enhance the original system while also significantly outperforming other tested baseline methods.

The fairness of biometric systems, in particular facial recognition, is often analysed for larger demographic groups, e.g. female vs. male or black vs. white. In contrast to this, minority groups are commonly ignored. This paper investigates the performance of facial recognition algorithms on individuals with Down syndrome, a common chromosomal abnormality that affects approximately one in 1,000 births per year. To do so, a database of 98 individuals with Down syndrome, each represented by at least five facial images, is semi-automatically collected from YouTube. Subsequently, two facial image quality assessment algorithms and five recognition algorithms are evaluated on the newly collected database and on the public facial image databases CelebA and FRGCv2. The results show that the quality scores of facial images for individuals with Down syndrome are comparable to those of individuals without Down syndrome captured under similar conditions. Furthermore, it is observed that face recognition performance decreases significantly for individuals with Down syndrome, which is largely attributed to the increased likelihood of false matches.

Biometric Sample Quality Assessment (BSQA) estimates the usefulness of the captured image based on its utility for the recognition task. In this regard, the majority of studies have been proposed in the last decade for computing a sample quality score from facial images. In particular, methods that learn a regressor from pseudo-labels have obtained reliable results on various benchmarks. However, they fail to correctly estimate the quality of samples having both, low quality and low intra-class variability. This paper proposes a new BSQA approach, Perceptual Classifiability Ratio for Hand Image Quality Assessment (PCR-HIQA), which computes hand image quality by combining the relative classifiability of the sample with its fidelity-related properties. On the one hand, the classifiability ratio is calculated by mapping the feature representation of the training samples in the angular space with respect to its class centroid to the nearest negative class centroid. On the other hand, the fidelity properties encode the human perception of the input sample quality. Experimental results on the challenging HaGRID database, containing different hand gestures, underline the superiority of the proposed BSQA method which outperforms state-of-the-art techniques by up to 30%.1

Various face image datasets intended for facial biometrics research were created via web-scraping, i.e. the collection of images publicly available on the internet.This work presents an approach to detect both exactly and nearly identical face image duplicates, using file and image hashes.The approach is extended through the use of face image preprocessing.Additional steps based on face recognition and face image quality assessment models reduce false positives, and facilitate the deduplication of the face images both for intra-and inter-subject duplicate sets.The presented approach is applied to five datasets, namely LFW, TinyFace, Adience, CASIA-WebFace, and C-MS-Celeb (a cleaned MS-Celeb-1M variant).Duplicates are detected within every dataset, with hundreds to hundreds of thousands of duplicates for all except LFW.Face recognition and quality assessment experiments indicate a minor impact on the results through the duplicate removal.

It has been shown that digitally manipulated face images can pose a security threat to automated authentication systems (e.g., when such systems are used for border control). In such scenarios, a malicious actor can, in many countries, apply for an identity document using a manipulated face image, which can then be used to gain fraudulent access to a system. Research has shown that humans and algorithms struggle to detect digitally manipulated face images, especially when the type of manipulation is unknown or when evaluated across multiple types of manipulations. In this work, we consider the detection performance of algorithms and humans on datasets consisting of retouched, face swapped and morphed images. Specifically, we investigate the joint performance of algorithms and humans in a differential detection scenario where both a trusted and suspected image are presented simultaneously. To this end, we propose a conditional face image manipulation detection approach where the human decision is only considered when the algorithm is unsure about the decision outcome. The results show that the automated algorithm performs better than the human detectors and that combining the decisions of algorithms and humans, in general, leads to an increased detection performance. To our knowledge, this is the first study to explore the joint detection performance of algorithms and humans in a differential face manipulation detection scenario and when using a variety of face image manipulations.

Deep neural networks have become prevalent in human analysis, boosting the performance of applications, such as biometric recognition, action recognition, as well as person re-identification. However, the performance of such networks scales with the available training data. In human analysis, the demand for large-scale datasets poses a severe challenge, as data collection is tedious, time-expensive, costly and must comply with data protection laws. Current research investigates the generation of synthetic data as an efficient and privacy-ensuring alternative to collecting real data in the field. This survey introduces the basic definitions and methodologies, essential when generating and employing synthetic data for human analysis. We summarise current state-of-the-art methods and the main benefits of using synthetic data. We also provide an overview of publicly available synthetic datasets and generation models. Finally, we discuss limitations, as well as open research problems in this field. This survey is intended for researchers and practitioners in the field of human analysis.

The development of large-scale identification systems that ensure the privacy protection of enrolled subjects represents a major challenge. Biometric deployments that provide interoperability and usability by including efficient multi-biometric solutions are a recent requirement. In the context of privacy protection, several template protection schemes have been proposed in the past. However, these schemes seem inadequate for indexing (workload reduction) in biometric identification systems. More specifically, they have been used in identification systems that perform exhaustive searches, leading to a degradation of computational efficiency. To overcome these limitations, we present an efficient privacy-preserving multi-biometric identification system that retrieves protected deep cancelable templates and is agnostic with respect to biometric characteristics and biometric template protection schemes. To this end, a multi-biometric binning scheme is designed to exploit the low intra-class variation properties contained in the frequent binary patterns extracted from different types of biometric characteristics. Experimental results reported on publicly available databases using state-of-the-art Deep Neural Network (DNN)-based embedding extractors show that the protected multi-biometric identification system can reduce the computational workload to approximately 57% (indexing up to three types of biometric characteristics) and 53% (indexing up to two types of biometric characteristics), while simultaneously improving the biometric performance of the baseline biometric system at the high-security thresholds.

The accuracy of face recognition skyrocketed in past years and its robustness towards various covariates has been shown, such as variations in pose or age. In recent years, a considerable amount of research efforts has been devoted to cross-domain face recognition aiming at comparing facial images obtained from domains that are different in capture technologies, e.g. visible spectrum versus infrared, or signal representations, e.g. photographs versus sketches. Yet, various relevant domains have hardly been explored and a lack of public databases hampers the development of new algorithms.In this work, we introduce the HDA Cross-Domain (HDA-CD) face image database comprising 1,400 face images from three different domains including avatars, caricatures, and sketches. Said face images were manually generated using popular mobile apps. In a benchmark, we evaluate commercial and open-source state-of-the-art facial analysis methods on the HDA-CD database including face detection and recognition. For the latter task, generated facial images are compared against their original counter-parts. The HDA-CD database is made publicly available at: https://dasec.h-da.de/hda-cdfdb/

Lossy face image compression can degrade the image quality and the utility for the purpose of face recognition. This work investigates the effect of lossy image compression on a state-of-the-art face recognition model, and on multiple face image quality assessment models. The analysis is conducted over a range of specific image target sizes. Four compression types are considered, namely JPEG, JPEG 2000, downscaled PNG, and notably the new JPEG XL format. Frontal color images from the ColorFERET database were used in a Region Of Interest (ROI) variant and a portrait variant. We primarily conclude that JPEG XL allows for superior mean and worst case face recognition performance especially at lower target sizes, below approximately 5kB for the ROI variant, while there appears to be no critical advantage among the compression types at higher target sizes. Quality assessments from modern models correlate well overall with the compression effect on face recognition performance.

Biometric data stored in automated recognition systems are at risk of attacks. This is particularly true for large-scale biometric identification systems, where the reference database is often accessed remotely. A popular approach for the protection of the stored templates is homomorphic encryption, which grants privacy protection while maintaining the biometric performance of the unprotected system. However, it introduces a significant computational overhead that can render identification transactions infeasible. To reduce this workload, biometric indexing in the encrypted domain has become a recent research interest. In this work, we show that in such schemes, auxiliary indexing data can leak additional privacy-sensitive information that violate standardized requirements for biometric template protection. In response to this leakage, we propose a novel framework HEBI that protects biometric indexing approaches at a post-quantum security level while requiring a computational effort of only 0.12 milliseconds per cluster.

Accurate face recognition systems are increasingly important in sensitive applications like border control or migration management. Therefore, it becomes crucial to quantify the quality of facial images to ensure that lowquality images are not affecting recognition accuracy. In this context, the current draft of ISO/IEC 29794-5 introduces the concept of component quality to estimate how single factors of variation affect recognition outcomes. In this study, we propose a quality measure (NeutrEx) based on the accumulated distances of a 3D face reconstruction to a neutral expression anchor. Our evaluations demonstrate the superiority of our proposed method compared to baseline approaches obtained by training Support Vector Machines on face embeddings extracted from a pre-trained Convolutional Neural Network for facial expression classification. Furthermore, we highlight the explainable nature of our NeutrEx measures by computing per-vertex distances to unveil the most impactful face regions and allow operators to give actionable feedback to subjects1.

This study investigates the possibility of mitigating the demographic biases that affect face recognition technologies through the use of synthetic data. Demographic biases have the potential to impact individuals from specific demographic groups, and can be identified by observing disparate performance of face recognition systems across demographic groups. They primarily arise from the unequal representations of demographic groups in the training data. In recent times, synthetic data have emerged as a solution to some problems that affect face recognition systems. In particular, during the generation process it is possible to specify the desired demographic and facial attributes of images, in order to control the demographic distribution of the synthesized dataset, and fairly represent the different demographic groups. We propose to fine-tune with synthetic data existing face recognition systems that present some demographic biases. We use synthetic datasets generated with GANDiffFace, a novel framework able to synthesize datasets for face recognition with controllable demographic distribution and realistic intra-class variations. We consider multiple datasets representing different demographic groups for training and evaluation. Also, we fine-tune different face recognition systems, and evaluate their demographic fairness with different metrics. Our results support the proposed approach and the use of synthetic data to mitigate demographic biases in face recognition.

Contactless fingerprint recognition is an emerging biometric technology and Presentation Attack Detection (PAD) methods are crucial to preserve system security. Convolutional Neural Networks (CNNs) represent the state-of the-art of PAD algorithms for many contactless captured biometric characteristics and various research groups proposed specialized CNN-based PAD methods or used general purpose CNNs to detect Presentation Attacks (PAs). In this work, we compare nine CNN-based PAD methods for contactless fingerprint PAD: five general purpose algorithms, and four dedicated PAD methods designed for various biometric characteristics. To achieve this, we combine the COLFISPOOF database with three bona fide databases: the HDA database and both versions of the ISPFD database. We set up our experiments using a baseline evaluation protocol and four Leave-One-Out (LOO) protocols, to benchmark the generalization capabilities to unseen data. The results reported by using the Attack Presentation Classification Error Rate (APCER) vs. Bona fide Presentation Classification Error Rate (BPCER) and the Detection Equal Error Rate (D-EER). Further, we discuss the achieved results in detail and give recommendations for real-world implementations. Our results show that established PAD algorithms for other biometric characteristics can accurately detect PAs on contactless fingerprints. While strong deviations between the considered PAD algorithms are observed, the best performing method shows a D-EER between 0.01% and 0.08% (depending on the LOO partition) and a APCER of 0.00% at a BPCER of 1.00%.

The fuzzy vault scheme has been established as cryptographic primitive suitable for privacy-preserving biometric authentication. To improve accuracy and privacy protection, biometric information of multiple characteristics can be fused at feature level prior to locking it in a fuzzy vault. In this work, we provide a formalisation of feature-level fusion in multi-biometric fuzzy vaults, on the basis of which relevant security issues are elaborated. In a case study, we construct a multi-biometric fuzzy vault based on face and multiple fingerprints. On a multi-biometric database constructed from the FRGCv2 face and the MCYT-100 fingerprint databases, a perfect recognition accuracy is achieved at a false accept security above 30 bits. We define countermeasures for observed security issues, that are commonly ignored and may impair the overall system’s security. Finally, a method for extending the fuzzy vault scheme with a password is proposed.

Protecting minutiae-based fingerprint templates with fully homomorphic encryption has recently been recognised as a hard problem. In this work, we evaluate state-of-the-art fingerprint recognition based on minutiae templates using post-quantum secure fully homomorphic encryption that operates directly on floating point numbers, such that no simplification or quantisation of the comparison algorithm is necessary. In a practical evaluation on a publicly available dataset, we run a benchmark and provide directions for future work.

The efficiency of biometric systems, in particular efficient and accurate biometric identification, is one of the most challenging open problems in biometrics today. Adding to that, biometric data are sensitive data deserving adequate protection. As a solution, this work proposes an efficient privacy-preserving reduction of the computational workload of biometric identification systems using public-key encryption with keyword search (PEKS). For long-term protection of the biometric data, fully homomorphic encryption is applied for template protection. As all applied cryptographic schemes are lattice-based, they also offer post-quantum security. Throughout the system, the recognition accuracy of the unprotected system is preserved. In an evaluation on a public face database, the computational workload of an identification search in the encrypted domain is reduced down to 8.4% compared to an exhaustive search, achieving identification on 1062 subjects in 210 milliseconds. Based on these results, an identification search on 1 million subjects can be estimated at under 3 minutes using off-the-shelf hardware.

Synthetic fingerprint image generation is important for the development of biometric recognition systems at scale due to the lack of easy-to-distribute real data, e.g. due to legal restrictions. Based on a SFinGe ridge line pat-tern, the recently introduced SynCoLFinGer algorithm gen-erates synthetic contactless fingerprints. However, it neglects some of the most frequent imperfections shown in common fingerprint databases. This work aims to analyse the most frequently found imperfections on fingerprint images and implements additional imperfections to SynCoLFinGer, including ink stains, dermatological issues, wounds and scars. The resem-blance of the generated fingerprints is assessed with re-gard to their sample quality and visually in comparison to real samples. Moreover, we report the biometric performance and sample quality on a generated database. The source code of this work is made publicly available under: https://gitlab.com/jannispriesnitz/syncolfinger

Homomorphic Encryption (HE) has become a well-known tool for privacy-preserving recognition in biometric systems. Despite some important advantages of HE (such as preservation of recognition accuracy), there are two main drawbacks in the application of HE to biometric recognition systems: first, the security of the system solely depends on the secrecy of the private (decryption) key; second, the computational costs of the operations on the ciphertexts are expensive. To address these challenges, in this paper we propose a hybrid scheme for the protection of biometric templates, which combines cancelable biometrics (CB) methods and HE. Applying CB prior to HE enhances both the security and privacy of the overall system, since the protected templates remain irreversible even if the secret keys are leaked (commonly referred to as the full disclosure scenario). In addition, we can reduce the dimensionality of templates using CB before applying HE, which speeds up the computation over the ciphertexts. We use BioHashing, Multi-Layer Perceptron (MLP) hashing, and Index-of-Maximum (IoM) hashing as different CB methods, and for each of these schemes, we propose a method for computing scores between hybrid-protected templates in the encrypted domain. We evaluate our proposed hybrid scheme using different state-of-the-art face recognition models (Ar-cFace, ElasticFace, and FaceNet) on the MOBIO and LFW datasets. The source code of our experiments is publicly available, so our work can be fully reproduced.

In this work, we present a simple biometric indexing scheme which is binning and retrieving cancelable deep face templates based on frequent binary patterns. The simplicity of the proposed approach makes it applicable to unprotected as well as protected, i.e. cancelable, deep face templates. As such, this approach represents to the best of the authors' knowledge the first generic indexing scheme that can be applied to arbitrary cancelable face templates (o binary representation). In experiments, deep face templates are obtained from the Labelled Faces in the Wild (LFW) dataset using the ArcFace face recognition system for feature extraction. Protected templates are then generated by employing different cancelable biometric schemes, i.e. BioHashing and two variants of Index-of-Maximum Hashing. The proposed indexing scheme is evaluated on closed- and open-set identification scenarios. It is shown to maintain the recognition accuracy of the baseline system while reducing the penetration rate and hence the workload of identifications to approximately 40%.

We investigate the potential of fusing human examiner decisions for the task of digital face manipulation detection. To this end, various decision fusion methods are proposed incorporating the examiners’ decision confidence, experience level, and their time to take a decision. Conducted experiments are based on a psychophysical evaluation of digital face image manipulation detection capabilities of humans in which different manipulation techniques were applied, i.e. face morphing, face swapping and retouching. The decisions of 223 participants were fused to simulate crowds of up to seven human examiners. Experimental results reveal that (1) despite the moderate detection performance achieved by single human examiners, a high accuracy can be obtained through decision fusion and (2) a weighted fusion, which takes the examiners’ decision confidence into account, yields the most competitive detection performance.

We present the first method for synthetic generation of contactless fingerprint images, referred to as SynCoLFinGer. To this end, the constituent components of contactless fingerprint images regarding capturing, subject characteristics, and environmental influences are modeled and applied to a synthetically generated ridge pattern using the SFinGe algorithm. The proposed method is able to generate different synthetic samples corresponding to a single finger and it can be parameterized to generate contactless fingerprint images of various quality levels. The resemblance of the synthetically generated contactless fingerprints to real fingerprints is confirmed by evaluating biometric sample quality using an adapted NFIQ 2.0 algorithm and biometric utility using a state-of-the-art contactless fingerprint recognition system.

Recently, the use of remote education via electronic media, i.e. eLearning, has increased due to the COVID-19 pandemic. To achieve secure and reliable identity verification in eLearning exams, it has been suggested to employ face recognition technologies for remote student authentication in online examinations. However, novel attacks on face recognition in eLearning systems have rarely been considered in the scientific literature.In this work, we investigate the feasibility of so-called face morphing attacks in eLearning systems. Such attacks can be launched in scenarios where students are authenticated via identity documents containing face images that are remotely presented prior to online examinations. In this relevant scenario, students are able to fool human examination and automated face recognition by morphing their face image with that of an accomplice, e.g. fellow student. Resulting morphed face images contain biometric information of both subjects contributing to it. Consequentially, an accomplice could take part in an online examination for a student with high probability of passing an identity verification unnoticed. We assess the vulnerability of a commercial and an open-source face recognition system to said attack. To this end, a realistic dataset of morphing attacks is collected. It is shown that automated face recognition in eLearning systems can be tricked with alarmingly high success chance.

Face recognition can benefit from the utilization of depth data captured using low-cost cameras, in particular for presentation attack detection purposes. Depth video output from these capture devices can however contain defects such as holes or general depth inaccuracies. This work proposes a deep learning face depth enhancement method in this context of facial biometrics, which adds a security aspect to the topic. U-Net-like architectures are utilized, and the networks are compared against hand-crafted enhancer types, as well as a similar depth enhancer network from related work trained for an adjacent application scenario. All tested enhancer types exclusively use depth data as input, which differs from methods that enhance depth based on additional input data such as visible light color images. Synthetic face depth ground truth images and degraded forms thereof are created with help of PRNet, to train multiple deep learning enhancer models with different network sizes and training configurations. Evaluations are carried out on the synthetic data, on Kinect v1 images from the KinectFaceDB, and on in-house RealSense D435 images. These evaluations include an assessment of the falsification for occluded face depth input, which is relevant to biometric security. The proposed deep learning enhancers yield noticeably better results than the tested preexisting enhancers, without overly falsifying depth data when non-face input is provided, and are shown to reduce the error of a simple landmark-based PAD method.

Morphing attacks have posed a severe threat to Face Recognition System (FRS). Despite the number of advancements reported in recent works, we note serious open issues such as independent benchmarking, generalizability challenges and considerations to age, gender, ethnicity that are inadequately addressed. Morphing Attack Detection (MAD) algorithms often are prone to generalization challenges as they are database dependent. The existing databases, mostly of semi-public nature, lack in diversity in terms of ethnicity, various morphing process and post-processing pipelines. Further, they do not reflect a realistic operational scenario for Automated Border Control (ABC) and do not provide a basis to test MAD on unseen data, in order to benchmark the robustness of algorithms. In this work, we present a new sequestered dataset for facilitating the advancements of MAD where the algorithms can be tested on unseen data in an effort to better generalize. The newly constructed dataset consists of facial images from 150 subjects from various ethnicities, age-groups and both genders. In order to challenge the existing MAD algorithms, the morphed images are with careful subject pre-selection created from the contributing images, and further post-processed to remove morphing artifacts. The images are also printed and scanned to remove all digital cues and to simulate a realistic challenge for MAD algorithms. Further, we present a new online evaluation platform to test algorithms on sequestered data. With the platform we can benchmark the morph detection performance and study the generalization ability. This work also presents a detailed analysis on various subsets of sequestered data and outlines open challenges for future directions in MAD research.

The paper presents a summary of the 2020 Sclera Segmentation Benchmarking Competition (SSBC), the 7th in the series of group benchmarking efforts centred around the problem of sclera segmentation. Different from previous editions, the goal of SSBC 2020 was to evaluate the performance of sclera-segmentation models on images captured with mobile devices. The competition was used as a platform to assess the sensitivity of existing models to i) differences in mobile devices used for image capture and ii) changes in the ambient acquisition conditions. 26 research groups registered for SSBC 2020, out of which 13 took part in the final round and submitted a total of 16 segmentation models for scoring. These included a wide variety of deep-learning solutions as well as one approach based on standard image processing techniques. Experiments were conducted with three recent datasets. Most of the segmentation models achieved relatively consistent performance across images captured with different mobile devices (with slight differences across devices), but struggled most with low-quality images captured in challenging ambient conditions, i.e., in an indoor environment and with poor lighting.

Facial cosmetics have the ability to substantially alter the facial appearance, which can negatively affect the decisions of a face recognition. In addition, it was recently shown that the application of makeup can be abused to launch so-called makeup presentation attacks. In such attacks, the attacker might apply heavy makeup in order to achieve the facial appearance of a target subject for the purpose of impersonation. In this work, we assess the vulnerability of a COTS face recognition system to makeup presentation attacks employing the publicly available Makeup Induced Face Spoofing (MIFS) database. It is shown that makeup presentation attacks might seriously impact the security of the face recognition system. Further, we propose an attack detection scheme which distinguishes makeup presentation attacks from genuine authentication attempts by analysing differences in deep face representations obtained from potential makeup presentation attacks and corresponding target face images. The proposed detection system employs a machine learning-based classifier, which is trained with synthetically generated makeup presentation attacks utilizing a generative adversarial network for facial makeup transfer in conjunction with image warping. Experimental evaluations conducted using the MIFS database reveal a detection equal error rate of 0.7% for the task of separating genuine authentication attempts from makeup presentation attacks.

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.

In recent years, morphing of facial images has arisen as an important attack vector on biometric systems. Detection of morphed images has proven challenging for automated systems and human experts alike. Likewise, in recent years, the importance of efficient (fast) biometric identification has been emphasised by the rapid rise and growth of large-scale bio-metric systems around the world.In this paper, the aforementioned, hitherto unrelated, topics within the biometrics domain are combined: the properties of morphed images are exploited for the purpose of improving the transaction times of a biometric identification system. Specifically, morphs of two or more samples are used in the pre-selection step of a two-stage biometric identification system. In a proof-of-concept experimental evaluation using two state-of-the-art open-source facial recognition frameworks it is shown, that the proposed system achieves hit rates comparable to that of an exhaustive search-based baseline, while significantly reducing the penetration rate (and thus the computational workload) associated with the biometric identification transactions.

The vulnerability of face recognition systems to attacks based on morphed biometric samples has been established in the recent past. Such attacks pose a severe security threat to a biometric recognition system in particular within the widely deployed border control applications. However, so far a reliable detection of morphed images has remained an unsolved research challenge. In this work, automated morph detection algorithms based on general purpose pattern recognition algorithms are benchmarked for two scenarios relevant in the context of fraud detection for electronic travel documents, i.e. single image (no-reference) and image pair (differential) morph detection. In the latter scenario a trusted live capture from an authentication attempt serves as additional source of information and, hence, the difference between features obtained from this face image and a potential morph can be estimated. A dataset of 2,206 ICAO compliant bona fide face images of the FRGCv2 face database is used to automatically generate 4,808 morphs. It is shown that in a differential scenario morph detectors which utilize a score level-based fusion of detection scores obtained from a single image and differences between image pairs generally outperform no-reference morph detectors with regard to the employed algorithms and used parameters. On average a relative improvement of more than 25% in terms of detection equal error rate is achieved.

Feature vectors extracted from biometric characteristics are often represented using floating point values. It is, however, more appealing to store and compare feature vectors in a binary representation, since it generally requires less storage and facilitates efficient comparators which utilise intrinsic bit operations. Furthermore, the binary representations are very often necessary for some specific application scenarios, e.g. template protection and indexing. In recent years, usage of deep neural networks for facial recognition has vastly improved the biometric performance of said systems. In this paper, various binarisation schemes are applied to such feature vectors and benchmarked for biometric performance. It is shown that with only a negligible drop in biometric performance, the storage space and computational requirements can be vastly decreased.

The wide deployment of biometric recognition systems in the last two decades has raised privacy concerns regarding the storage and use of biometric data. As a consequence, the ISO/IEC 24745 international standard on biometric information protection has established two main requirements for protecting biometric templates: irreversibility and unlinkability. Numerous efforts have been directed to the development and analysis of irreversible templates. However, there is still no systematic quantitative manner to analyze the unlinkability of such templates. In this paper, we address this shortcoming by proposing a new general framework for the evaluation of biometric templates' unlinkability. To illustrate the potential of the approach, it is applied to assess the unlinkability of the four state-of-the-art techniques for biometric template protection: biometric salting, bloom filters, homomorphic encryption, and block re-mapping. For the last technique, the proposed framework is compared with other existing metrics to show its advantages.

In contrast to electronic travel documents (e.g. ePassports), the standardisation of breeder documents (e.g. birth certificates), regarding harmonisation of content and contained security features is in statu nascendi. Due to the fact that breeder documents can be used as an evidence of identity and enable the application for electronic travel documents, they pose the weakest link in the identity life cycle and represent a security gap for identity management. In this work, we present a cost efficient way to enhance the long-term security of breeder documents by utilizing blockchain technology. A conceptual architecture to enhance breeder document long-term security and an introduction of the concept's constituting system components is presented. Our investigations provide evidence that the Bitcoin blockchain is most suitable for breeder document long-term security.

We present a multi-iris indexing system for efficient and accurate large-scale identification. The system is based on Bloom filters and binary search trees. We describe and empirically evaluate several possible information fusion strategies for the system. Those experiments are performed using a combination of several publicly available datasets; the proposed system is tested in an open-set identification scenario consisting of 6,000 genuine and 100,000 impostor transactions. The system maintains the near-optimal biometric performance of an iris-code, score fusion based baseline system, while reducing the required lookup workload to less than 1% thereof.

The necessity of biometric template alignment imposes a significant computational load and increases the probability of false positive occurrences in biometric systems. While for some modalities, automatic pre-alignment of biometric samples is utilised, this topic has not yet been explored for systems based on the iris. This paper presents a method for pre-alignment of iris images based on the positions ofautomatically detected eye corners. Existing work in the area of automatic eye corner detection has hitherto only involved visible wavelength images; for the near-infrared images, used in the vast majority of current iris recognition systems, this task is significantly more challenging and as of yet unexplored. A comparative study of two methods for solving this problem is presented in this paper. The eye corners detected by the two methods are then used for the pre-alignment and biometric performance evaluation experiments. The system utilising image pre-alignment is benchmarked against a baseline iris recognition system on the iris subset of the BioSecure database. In the benchmark, the workload associated with alignment compensation is significantly reduced, while the biometric performance remains unchanged or even improves slightly.

Morphing techniques can be used to create artificial biometric samples, which resemble the biometric information of two (or more) individuals in image and feature domain. If morphed biometric images or templates are infiltrated to a biometric recognition system the subjects contributing to the morphed image will both (or all) be successfully verified against a single enrolled template. Hence, the unique link between individuals and their biometric reference data is annulled. The vulnerability of face and fingerprint recognition systems to such morphing attacks has been assessed in the recent past. In this paper we investigate the feasibility of morphing iris-codes. Two relevant attack scenarios are discussed and a scheme for morphing pairs of iris-codes depending on the expected stability of their bits is proposed. Different iris recognition systems, which accept comparison scores at a recommended Hamming distance of 0.32, are shown to be vulnerable to attacks based on the presented morphing technique.

18th Annual Conference of the International Speech Communication Association, INTERSPEECH 2017, Stockholm, Sweden, 20-24 August 2017

The vast majority of text-independent speaker recognition systems rely on intermediate-sized vectors (i-vectors), which are compared by probabilistic linear discriminant analysis (PLDA). This paper proposes a PLDA-alike approach with restricted Boltzmann machines for i-vector based speaker recognition: two deep architectures are presented and examined, which aim at suppressing channel effects and recovering speaker-discriminative information on back-ends trained on a small dataset. Experiments are carried out on the MOBIO SRE'13 database, which is a challenging and publicly available dataset for mobile speaker recognition with limited amounts of training data. The experiments show that the proposed system outperforms the baseline i-vector/PLDA approach by relative gains of 31% on female and 9% on male speakers in terms of half total error rate.

Iris recognition technologies are deployed in numerous large-scale nation-wide projects in order to provide robust and reliable biometric recognition of individuals. Moreover, the iris has been found to be rather stable over time, i.e. iris biometric reference data provides a strong and permanent link between individuals and their biometric traits. Hence, unprotected storage of (iris) biometric data provokes serious privacy threats, e.g. identity theft, limited re-newability, or cross-matching. Biometric cryptosystems grant a significant improvement in data privacy and increase the likelihood that individuals will effectively consent in the biometric system usage. However, the vast majority of proposed biometric cryptosystems do not guarantee desired properties of irreversibility, unlinkability, and re-newability without significantly degrading the biometric performance. In this work, we propose an unlinkable multi-instance iris biometric cryptosystem based on the improved fuzzy vault scheme. The proposed system locks biometric feature sets extracted from binary iris biometric reference data, i.e. iris-codes, of the left and right irises in a single fuzzy vault. In order to retain the size of the protected template and authentication speed, the proposed fusion step combines the most discriminative parts of two iris-codes at feature level. It is shown that the proposed key-binding process enables the generation of irreversible protected templates which prevents from previously proposed cross-matching attacks. Further, we investigate the optimal choice among potential decoding strategies with respect to biometric performance and time of key retrieval. The fully reproducible system is integrated to two different publicly available iris recognition systems and evaluated on the CASIAv3-Interval and the IITDv1 iris databases. Compared to the corresponding unprotected recognition schemes, genuine match rates of approximately 95 and 97 % at which no false accepts are observed and maintained in a single- and multi-instance scenario, respectively. Moreover, the multi-iris system is shown to significantly improve privacy protection achieving security levels of approximately 70 bits at practical biometric performance.

The vast majority of speaker recognition cross-entropy evaluations are focused on score domain. By examining the generalized relative distance between genuine and impostor sub-spaces, biometric characteristics become comparable to other authentication approaches. In this paper we demonstrate that the i-vector feature space's biometric information measured by relative entropy is comparable to e.g., knowledge-based mechanisms or face recognition. Examining NIST SRE 2004-2010 corpora, short samples of e.g, 5 seconds duration, comprise already 127 bits in a text-independent scenario. Further, the vast majority of short samples does not fall below 50% of the biometric information of samples having a duration of more than 40 seconds. The generalized i-vector feature space entropy of long samples corresponds to 182.1 bits, and the highest lower entropy bound of a subject was observed at 471.6 bits.

The International Organization for Standardization (ISO) specifies iris biometric data to be recorded and stored in (raw) image form (ISO/IEC 19794-6), rather than in extracted templates, i.e. iris-codes. Existing literature confirms the applicability of lossy image compression in iris biometric systems, however, so far investigations on the impact of image compression on iris segmentation algorithms have remained elusive. In this work we examine the impact of severe image compression algorithms in particular, JPEG, JPEG 2000, and JPEG-XR, on the performance of different iris segmentation approaches. Experiments are carried out on an uncompressed iris database and, based on a manually annotated ground truth, effects of image compression on iris segmentation are quantified. It is found that surprisingly, JPEG causes the least segmentation errors over a wide range of high to medium bitrates (down to 0.3 bpp) despite of its weak performance in terms of PSNR rate-distortion behaviour.

Biometric template protection schemes in particular, biometric cryptosystems bind secret keys to biometric data, i.e. complex key retrieval processes are performed at each authentication attempt. Focusing on biometric identification exhaustive 1: N comparisons are required for identifying a biometric probe. As a consequence comparison time frequently dominates the overall computational workload, preventing biometric cryptosystems from being operated in identification mode. In this paper we propose a computational efficient two-stage identification system for fingerprint-biometric cryptosystems. Employing the concept of adaptive Bloom filter-based cancelable biometrics, pseudonymous binary prescreeners are extracted based on which top-candidates are returned from a database. Thereby the number of required key-retrieval processes is reduced to a fraction of the total. Experimental evaluations confirm that, by employing the proposed technique, biometric cryptosystems, e.g. fuzzy vault scheme, can be enhanced in order to enable a real-time privacy preserving identification, while at the same time biometric performance is maintained.

Biometric data are considered sensitive personal data and any privacy leakage poses severe security risks. Biometric templates should hence be protected, obscuring the biometric signal in a non-reversible manner, while preserving the unprotected system's performance. In the present work, irreversible face templates based on adaptive Bloom filters are proposed. Experiments are carried out on the publicly available Bio Secure DB utilizing the free Bob image processing toolbox, so that research is fully reproducible. The performance and security evaluations proof the irreversibility of the protected templates, while preserving the verification performance. Furthermore, template size is considerably reduced.

Fuzzy commitment schemes have been established as a reliable means of binding cryptographic keys to binary feature vectors extracted from diverse biometric modalities. In addition, attempts have been made to extend fuzzy commitment schemes to incorporate multiple biometric feature vectors. Within these schemes potential improvements through feature level fusion are commonly neglected. In this paper a feature level fusion technique for fuzzy commitment schemes is presented. The proposed reliability- balanced feature level fusion is designed to re-arrange and combine two binary biometric templates in a way that error correction capacities are exploited more effectively within a fuzzy commitment scheme yielding improvement with respect to key-retrieval rates. In experiments, which are carried out on iris-biometric data, reliability-balanced feature level fusion significantly outperforms conventional approaches to multi-biometric fuzzy commitment schemes confirming the soundness of the proposed technique.

Form a privacy perspective most concerns against the common use of biometrics arise from the storage and misuse of biometric data. Biometric cryptosystems and cancelable biometrics represent emerging technologies of biometric template protection addressing these concerns and improving public confidence and acceptance of biometrics. In addition, biometric cryptosystems provide mechanisms for biometric-dependent key-release. In the last years a significant amount of approaches to both technologies have been published. A comprehensive survey of biometric cryptosystems and cancelable biometrics is presented. State-of-the-art approaches are reviewed based on which an in-depth discussion and an outlook to future prospects are given.

In this paper we propose a modified hill-climbing attack to iris biometric systems. Applying our technique we are able to effectively gain access to iris biometric systems at very low effort. Furthermore, we demonstrate that reconstructing approximations of original iris images is highly non-trivial.

Performing identification on large-scale biometric databases requires an exhaustive linear search. Since biometric data does not have any natural sorting order, indexing databases, in order to minimize the response time of the system, represents a great challenge. In this work we propose a biometric hash generation technique for the purpose of biometric database indexing, applied to iris biometrics. Experimental results demonstrate that the presented approach highly accelerates biometric identification.