Philipp Terhörst

Marco Huber, Anh Thi Luu, Philipp Terhörst et al.

Explainable Face Recognition is gaining growing attention as the use of the technology is gaining ground in security-critical applications. Understanding why two faces images are matched or not matched by a given face recognition system is important to operators, users, anddevelopers to increase trust, accountability, develop better systems, and highlight unfair behavior. In this work, we propose xSSAB, an approach to back-propagate similarity score-based arguments that support or oppose the face matching decision to visualize spatial maps that indicate similar and dissimilar areas as interpreted by the underlying FR model. Furthermore, we present Patch-LFW, a new explainable face verification benchmark that enables along with a novel evaluation protocol, the first quantitative evaluation of the validity of similarity and dissimilarity maps in explainable face recognition approaches. We compare our efficient approach to state-of-the-art approaches demonstrating a superior trade-off between efficiency and performance. The code as well as the proposed Patch-LFW is publicly available at: https://github.com/marcohuber/xSSAB.

Marco Huber, Philipp Terhörst, Florian Kirchbuchner et al.

Face Recognition (FR) is increasingly used in critical verification decisions and thus, there is a need for assessing the trustworthiness of such decisions. The confidence of a decision is often based on the overall performance of the model or on the image quality. We propose to propagate model uncertainties to scores and decisions in an effort to increase the transparency of verification decisions. This work presents two contributions. First, we propose an approach to estimate the uncertainty of face comparison scores. Second, we introduce a confidence measure of the system's decision to provide insights into the verification decision. The suitability of the comparison scores uncertainties and the verification decision confidences have been experimentally proven on three face recognition models on two datasets.

Ying Xu, Philipp Terhörst, Kiran Raja et al.

In recent years, image and video manipulations with Deepfake have become a severe concern for security and society. Many detection models and datasets have been proposed to detect Deepfake data reliably. However, there is an increased concern that these models and training databases might be biased and, thus, cause Deepfake detectors to fail. In this work, we investigate factors causing biased detection in public Deepfake datasets by (a) creating large-scale demographic and non-demographic attribute annotations with 47 different attributes for five popular Deepfake datasets and (b) comprehensively analysing attributes resulting in AI-bias of three state-of-the-art Deepfake detection backbone models on these datasets. The analysis shows how various attributes influence a large variety of distinctive attributes (from over 65M labels) on the detection performance which includes demographic (age, gender, ethnicity) and non-demographic (hair, skin, accessories, etc.) attributes. The results examined datasets show limited diversity and, more importantly, show that the utilised Deepfake detection backbone models are strongly affected by investigated attributes making them not fair across attributes. The Deepfake detection backbone methods trained on such imbalanced/biased datasets result in incorrect detection results leading to generalisability, fairness, and security issues. Our findings and annotated datasets will guide future research to evaluate and mitigate bias in Deepfake detection techniques. The annotated datasets and the corresponding code are publicly available.

Philipp Terhörst, Florian Bierbaum, Marco Huber et al.

A MasterFace is a face image that can successfully match against a large portion of the population. Since their generation does not require access to the information of the enrolled subjects, MasterFace attacks represent a potential security risk for widely-used face recognition systems. Previous works proposed methods for generating such images and demonstrated that these attacks can strongly compromise face recognition. However, previous works followed evaluation settings consisting of older recognition models, limited cross-dataset and cross-model evaluations, and the use of low-scale testing data. This makes it hard to state the generalizability of these attacks. In this work, we comprehensively analyse the generalizability of MasterFace attacks in empirical and theoretical investigations. The empirical investigations include the use of six state-of-the-art FR models, cross-dataset and cross-model evaluation protocols, and utilizing testing datasets of significantly higher size and variance. The results indicate a low generalizability when MasterFaces are training on a different face recognition model than the one used for testing. In these cases, the attack performance is similar to zero-effort imposter attacks. In the theoretical investigations, we define and estimate the face capacity and the maximum MasterFace coverage under the assumption that identities in the face space are well separated. The current trend of increasing the fairness and generalizability in face recognition indicates that the vulnerability of future systems might further decrease. Future works might analyse the utility of MasterFaces for understanding and enhancing the robustness of face recognition models.

Ivan DeAndres-Tame, Muhammad Faisal, Ruben Tolosana et al.

Face Recognition (FR) has advanced significantly with the development of deep learning, achieving high accuracy in several applications. However, the lack of interpretability of these systems raises concerns about their accountability, fairness, and reliability. In the present study, we propose an interactive framework to enhance the explainability of FR models by combining model-agnostic Explainable Artificial Intelligence (XAI) and Natural Language Processing (NLP) techniques. The proposed framework is able to accurately answer various questions of the user through an interactive chatbot. In particular, the explanations generated by our proposed method are in the form of natural language text and visual representations, which for example can describe how different facial regions contribute to the similarity measure between two faces. This is achieved through the automatic analysis of the output's saliency heatmaps of the face images and a BERT question-answering model, providing users with an interface that facilitates a comprehensive understanding of the FR decisions. The proposed approach is interactive, allowing the users to ask questions to get more precise information based on the user's background knowledge. More importantly, in contrast to previous studies, our solution does not decrease the face recognition performance. We demonstrate the effectiveness of the method through different experiments, highlighting its potential to make FR systems more interpretable and user-friendly, especially in sensitive applications where decision-making transparency is crucial.

Pedro C. Neto, Ana F. Sequeira, Jaime S. Cardoso et al.

In the context of biometrics, matching confidence refers to the confidence that a given matching decision is correct. Since many biometric systems operate in critical decision-making processes, such as in forensics investigations, accurately and reliably stating the matching confidence becomes of high importance. Previous works on biometric confidence estimation can well differentiate between high and low confidence, but lack interpretability. Therefore, they do not provide accurate probabilistic estimates of the correctness of a decision. In this work, we propose a probabilistic interpretable comparison (PIC) score that accurately reflects the probability that the score originates from samples of the same identity. We prove that the proposed approach provides optimal matching confidence. Contrary to other approaches, it can also optimally combine multiple samples in a joint PIC score which further increases the recognition and confidence estimation performance. In the experiments, the proposed PIC approach is compared against all biometric confidence estimation methods available on four publicly available databases and five state-of-the-art face recognition systems. The results demonstrate that PIC has a significantly more accurate probabilistic interpretation than similar approaches and is highly effective for multi-biometric recognition. The code is publicly-available.

Paul Jonas Kurz, Haiyu Wu, Kevin W. Bowyer et al.

Face recognition systems (FRS) exhibit significant accuracy differences based on the user's gender. Since such a gender gap reduces the trustworthiness of FRS, more recent efforts have tried to find the causes. However, these studies make use of manually selected, correlated, and small-sized sets of facial features to support their claims. In this work, we analyse gender bias in face recognition by successfully extending the search domain to decorrelated combinations of 40 non-demographic facial characteristics. First, we propose a toolchain to effectively decorrelate and aggregate facial attributes to enable a less-biased gender analysis on large-scale data. Second, we introduce two new fairness metrics to measure fairness with and without context. Based on these grounds, we thirdly present a novel unsupervised algorithm able to reliably identify attribute combinations that lead to vanishing bias when used as filter predicates for balanced testing datasets. The experiments show that the gender gap vanishes when images of male and female subjects share specific attributes, clearly indicating that the issue is not a question of biology but of the social definition of appearance. These findings could reshape our understanding of fairness in face biometrics and provide insights into FRS, helping to address gender bias issues.

Rouqaiah Al-Refai, Pankaja Priya Ramasamy, Ragini Ramesh et al.

The rapid advancement of authentication systems and their increasing reliance on biometrics for faster and more accurate user verification experience, highlight the critical need for a reliable framework to evaluate the suitability of biometric modalities for specific applications. Currently, the most widely known evaluation framework is a comparative table from 1998, which no longer adequately captures recent technological developments or emerging vulnerabilities in biometric systems. To address these challenges, this work revisits the evaluation of biometric modalities through an expert survey involving 24 biometric specialists. The findings indicate substantial shifts in property ratings across modalities. For example, face recognition, shows improved ratings due to technological progress, while fingerprint, shows decreased reliability because of emerging vulnerabilities and attacks. Further analysis of expert agreement levels across rated properties highlighted the consistency of the provided evaluations and ensured the reliability of the ratings. Finally, expert assessments are compared with dataset-level uncertainty across 55 biometric datasets, revealing strong alignment in most modalities and underscoring the importance of integrating empirical evidence with expert insight. Moreover, the identified expert disagreements reveal key open challenges and help guide future research toward resolving them.

Philipp Terhörst, Malte Ihlefeld, Marco Huber et al.

Face recognition systems have to deal with large variabilities (such as different poses, illuminations, and expressions) that might lead to incorrect matching decisions. These variabilities can be measured in terms of face image quality which is defined over the utility of a sample for recognition. Previous works on face recognition either do not employ this valuable information or make use of non-inherently fit quality estimates. In this work, we propose a simple and effective face recognition solution (QMagFace) that combines a quality-aware comparison score with a recognition model based on a magnitude-aware angular margin loss. The proposed approach includes model-specific face image qualities in the comparison process to enhance the recognition performance under unconstrained circumstances. Exploiting the linearity between the qualities and their comparison scores induced by the utilized loss, our quality-aware comparison function is simple and highly generalizable. The experiments conducted on several face recognition databases and benchmarks demonstrate that the introduced quality-awareness leads to consistent improvements in the recognition performance. Moreover, the proposed QMagFace approach performs especially well under challenging circumstances, such as cross-pose, cross-age, or cross-quality. Consequently, it leads to state-of-the-art performances on several face recognition benchmarks, such as 98.50% on AgeDB, 83.95% on XQLFQ, and 98.74% on CFP-FP. The code for QMagFace is publicly available

Dailé Osorio-Roig, Christian Rathgeb, Pawel Drozdowski et al.

In the recent past, different researchers have proposed privacy-enhancing face recognition systems designed to conceal soft-biometric attributes at feature level. These works have reported impressive results, but generally did not consider specific attacks in their analysis of privacy protection. We introduce an attack on said schemes based on two observations: (1) highly similar facial representations usually originate from face images with similar soft-biometric attributes; (2) to achieve high recognition accuracy, robustness against intra-class variations within facial representations has to be retained in their privacy-enhanced versions. The presented attack only requires the privacy-enhancing algorithm as a black-box and a relatively small database of face images with annotated soft-biometric attributes. Firstly, an intercepted privacy-enhanced face representation is compared against the attacker's database. Subsequently, the unknown attribute is inferred from the attributes associated with the highest obtained similarity scores. In the experiments, the attack is applied against two state-of-the-art approaches. The attack is shown to circumvent the privacy enhancement to a considerable degree and is able to correctly classify gender with an accuracy of up to approximately 90%. Future works on privacy-enhancing face recognition are encouraged to include the proposed attack in evaluations on the privacy protection.

Philipp Terhörst, Marco Huber, Naser Damer et al.

An essential factor to achieve high performance in face recognition systems is the quality of its samples. Since these systems are involved in daily life there is a strong need of making face recognition processes understandable for humans. In this work, we introduce the concept of pixel-level face image quality that determines the utility of pixels in a face image for recognition. We propose a training-free approach to assess the pixel-level qualities of a face image given an arbitrary face recognition network. To achieve this, a model-specific quality value of the input image is estimated and used to build a sample-specific quality regression model. Based on this model, quality-based gradients are back-propagated and converted into pixel-level quality estimates. In the experiments, we qualitatively and quantitatively investigated the meaningfulness of our proposed pixel-level qualities based on real and artificial disturbances and by comparing the explanation maps on faces incompliant with the ICAO standards. In all scenarios, the results demonstrate that the proposed solution produces meaningful pixel-level qualities enhancing the interpretability of the complete face image quality. The code is publicly available

Philipp Terhörst, André Boller, Naser Damer et al.

An essential factor to achieve high accuracies in fingerprint recognition systems is the quality of its samples. Previous works mainly proposed supervised solutions based on image properties that neglects the minutiae extraction process, despite that most fingerprint recognition techniques are based on detected minutiae. Consequently, a fingerprint image might be assigned a high quality even if the utilized minutia extractor produces unreliable information. In this work, we propose a novel concept of assessing minutia and fingerprint quality based on minutia detection confidence (MiDeCon). MiDeCon can be applied to an arbitrary deep learning based minutia extractor and does not require quality labels for learning. We propose using the detection reliability of the extracted minutia as its quality indicator. By combining the highest minutia qualities, MiDeCon also accurately determines the quality of a full fingerprint. Experiments are conducted on the publicly available databases of the FVC 2006 and compared against several baselines, such as NIST's widely-used fingerprint image quality software NFIQ1 and NFIQ2. The results demonstrate a significantly stronger quality assessment performance of the proposed MiDeCon-qualities as related works on both, minutia- and fingerprint-level. The implementation is publicly available.

Philipp Terhörst, Jan Niklas Kolf, Marco Huber et al.

Face recognition (FR) systems have a growing effect on critical decision-making processes. Recent works have shown that FR solutions show strong performance differences based on the user's demographics. However, to enable a trustworthy FR technology, it is essential to know the influence of an extended range of facial attributes on FR beyond demographics. Therefore, in this work, we analyse FR bias over a wide range of attributes. We investigate the influence of 47 attributes on the verification performance of two popular FR models. The experiments were performed on the publicly available MAADFace attribute database with over 120M high-quality attribute annotations. To prevent misleading statements about biased performances, we introduced control group based validity values to decide if unbalanced test data causes the performance differences. The results demonstrate that also many non-demographic attributes strongly affect the recognition performance, such as accessories, hair-styles and colors, face shapes, or facial anomalies. The observations of this work show the strong need for further advances in making FR system more robust, explainable, and fair. Moreover, our findings might help to a better understanding of how FR networks work, to enhance the robustness of these networks, and to develop more generalized bias-mitigating face recognition solutions.

Philipp Terhörst, Daniel Fährmann, Jan Niklas Kolf et al.

Soft-biometrics play an important role in face biometrics and related fields since these might lead to biased performances, threatens the user's privacy, or are valuable for commercial aspects. Current face databases are specifically constructed for the development of face recognition applications. Consequently, these databases contain large amount of face images but lack in the number of attribute annotations and the overall annotation correctness. In this work, we propose MAADFace, a new face annotations database that is characterized by the large number of its high-quality attribute annotations. MAADFace is build on the VGGFace2 database and thus, consists of 3.3M faces of over 9k individuals. Using a novel annotation transfer-pipeline that allows an accurate label-transfer from multiple source-datasets to a target-dataset, MAAD-Face consists of 123.9M attribute annotations of 47 different binary attributes. Consequently, it provides 15 and 137 times more attribute labels than CelebA and LFW. Our investigation on the annotation quality by three human evaluators demonstrated the superiority of the MAAD-Face annotations over existing databases. Additionally, we make use of the large amount of high-quality annotations from MAAD-Face to study the viability of soft-biometrics for recognition, providing insights about which attributes support genuine and imposter decisions. The MAAD-Face annotations dataset is publicly available.

Philipp Terhörst, Daniel Fährmann, Naser Damer et al.

Deeply-learned face representations enable the success of current face recognition systems. Despite the ability of these representations to encode the identity of an individual, recent works have shown that more information is stored within, such as demographics, image characteristics, and social traits. This threatens the user's privacy, since for many applications these templates are expected to be solely used for recognition purposes. Knowing the encoded information in face templates helps to develop bias-mitigating and privacy-preserving face recognition technologies. This work aims to support the development of these two branches by analysing face templates regarding 113 attributes. Experiments were conducted on two publicly available face embeddings. For evaluating the predictability of the attributes, we trained a massive attribute classifier that is additionally able to accurately state its prediction confidence. This allows us to make more sophisticated statements about the attribute predictability. The results demonstrate that up to 74 attributes can be accurately predicted from face templates. Especially non-permanent attributes, such as age, hairstyles, haircolors, beards, and various accessories, found to be easily-predictable. Since face recognition systems aim to be robust against these variations, future research might build on this work to develop more understandable privacy preserving solutions and build robust and fair face templates.

Philipp Terhörst, Jan Niklas Kolf, Naser Damer et al.

Face quality assessment aims at estimating the utility of a face image for the purpose of recognition. It is a key factor to achieve high face recognition performances. Currently, the high performance of these face recognition systems come with the cost of a strong bias against demographic and non-demographic sub-groups. Recent work has shown that face quality assessment algorithms should adapt to the deployed face recognition system, in order to achieve highly accurate and robust quality estimations. However, this could lead to a bias transfer towards the face quality assessment leading to discriminatory effects e.g. during enrolment. In this work, we present an in-depth analysis of the correlation between bias in face recognition and face quality assessment. Experiments were conducted on two publicly available datasets captured under controlled and uncontrolled circumstances with two popular face embeddings. We evaluated four state-of-the-art solutions for face quality assessment towards biases to pose, ethnicity, and age. The experiments showed that the face quality assessment solutions assign significantly lower quality values towards subgroups affected by the recognition bias demonstrating that these approaches are biased as well. This raises ethical questions towards fairness and discrimination which future works have to address.

Philipp Terhörst, Jan Niklas Kolf, Naser Damer et al.

Face image quality is an important factor to enable high performance face recognition systems. Face quality assessment aims at estimating the suitability of a face image for recognition. Previous work proposed supervised solutions that require artificially or human labelled quality values. However, both labelling mechanisms are error-prone as they do not rely on a clear definition of quality and may not know the best characteristics for the utilized face recognition system. Avoiding the use of inaccurate quality labels, we proposed a novel concept to measure face quality based on an arbitrary face recognition model. By determining the embedding variations generated from random subnetworks of a face model, the robustness of a sample representation and thus, its quality is estimated. The experiments are conducted in a cross-database evaluation setting on three publicly available databases. We compare our proposed solution on two face embeddings against six state-of-the-art approaches from academia and industry. The results show that our unsupervised solution outperforms all other approaches in the majority of the investigated scenarios. In contrast to previous works, the proposed solution shows a stable performance over all scenarios. Utilizing the deployed face recognition model for our face quality assessment methodology avoids the training phase completely and further outperforms all baseline approaches by a large margin. Our solution can be easily integrated into current face recognition systems and can be modified to other tasks beyond face recognition.

Philipp Terhörst, Marco Huber, Naser Damer et al.

Current research on soft-biometrics showed that privacy-sensitive information can be deduced from biometric templates of an individual. Since for many applications, these templates are expected to be used for recognition purposes only, this raises major privacy issues. Previous works focused on supervised privacy-enhancing solutions that require privacy-sensitive information about individuals and limit their application to the suppression of single and pre-defined attributes. Consequently, they do not take into account attributes that are not considered in the training. In this work, we present Negative Face Recognition (NFR), a novel face recognition approach that enhances the soft-biometric privacy on the template-level by representing face templates in a complementary (negative) domain. While ordinary templates characterize facial properties of an individual, negative templates describe facial properties that does not exist for this individual. This suppresses privacy-sensitive information from stored templates. Experiments are conducted on two publicly available datasets captured under controlled and uncontrolled scenarios on three privacy-sensitive attributes. The experiments demonstrate that our proposed approach reaches higher suppression rates than previous work, while maintaining higher recognition performances as well. Unlike previous works, our approach does not require privacy-sensitive labels and offers a more comprehensive privacy-protection not limited to pre-defined attributes.

Philipp Terhörst, Jan Niklas Kolf, Naser Damer et al.

Current face recognition systems achieve high progress on several benchmark tests. Despite this progress, recent works showed that these systems are strongly biased against demographic sub-groups. Consequently, an easily integrable solution is needed to reduce the discriminatory effect of these biased systems. Previous work mainly focused on learning less biased face representations, which comes at the cost of a strongly degraded overall recognition performance. In this work, we propose a novel unsupervised fair score normalization approach that is specifically designed to reduce the effect of bias in face recognition and subsequently lead to a significant overall performance boost. Our hypothesis is built on the notation of individual fairness by designing a normalization approach that leads to treating similar individuals similarly. Experiments were conducted on three publicly available datasets captured under controlled and in-the-wild circumstances. Results demonstrate that our solution reduces demographic biases, e.g. by up to 82.7% in the case when gender is considered. Moreover, it mitigates the bias more consistently than existing works. In contrast to previous works, our fair normalization approach enhances the overall performance by up to 53.2% at false match rate of 0.001 and up to 82.9% at a false match rate of 0.00001. Additionally, it is easily integrable into existing recognition systems and not limited to face biometrics.