In the current era of pervasive short video content, the exposure of passersby’s data frequently raises privacy concerns. Traditional anonymization techniques for passersby, like blurring and mosaicing, are often used before uploading such videos. However, these methods tend to degrade the informational richness of the visual content, markedly reducing the quality of the anonymized videos. Recent advancements of diffusion models have paved the way for text-guided image and video synthesis, yet applying these models to the anonymization of passersby poses three main challenges: i) bridging the domain gap between specific passersby data and high-quality image/video datasets that are used for pre-training diffusion models, ii) ensuring temporal consistency in the anonymized videos, and iii) preserving the integrity of video subjects’ content while exclusively anonymizing passersby-related information. To address these challenges, we propose the Passersby-Anonymizer, a novel diffusion-based framework for anonymizing identity-specific attributes in video content. At its core, our model introduces a spatial content adapter (SCA) to adapt to the visual patterns of passersby image datasets. We introduce a Temporal Content Stabilizer (TCS) to maintain the temporal consistency of the anonymized videos. Furthermore, we design a mask-aware training strategy that specifically targets the anonymization of the mask region while preserving the integrity of other contents. Our experimental evaluations demonstrate that our model effectively addresses the challenge of anonymizing passersby without compromising the informational integrity of the social videos. The source code is available at https://github.com/HappyDeepLearning/Passersby-Anonymizer.

Alzheimer's disease (AD) is a neurodegenerative disease caused by the progressive death of brain cells over time. It represents the most frequent cause of dementia in the western world, and affects an individual's cognitive ability and psychological capacity. While clinical diagnoses of AD are made primarily on the basis of clinical evaluation and mental health tests, diagnostic certainty is only possible through necropsy. One non-invasive approach to investigating AD is to use electroencephalograms (eEGs), which reflect brain electrical activity and so can be used to detect electrical abnormalities in brain signals with non-invasive cranial surface electrodes. Generally EEGs in AD patients show a shift to lower frequencies in spectral analysis and display less complexity and contain more regular patterns compared to those of control subjects. Here we present a method for differentiating AD patients from healthy ones based on their EEG signals using Benford's law and support vector machines (SVMs) with a radial basis function (RBF) kernel. EEG signals from eleven AD and eleven age-matched controls were divided into artefact-free 5-sec epochs and used to train an SVM. 10 fold cross validation was performed at both the epoch- and subject-level to evaluate the importance of each electrode in discriminating between AD and healthy subjects. Substantive variability was seen across the different electrodes, with electrodes O1, O2 and C4 particularly being important. Performance across the electrodes was reduced when subject-level cross validation was performed, but relative performance across the electrodes was consistent with that found using epoch-level cross validation.

Biometric authentication systems have certain limitations. Recent studies have shown that biometric features may change over time, which can entail a decrease in recognition performance of the biometric system. An adaptive biometric system addresses this problem by adapting the biometric reference/template over time, thereby tracking the changes automatically. However, the use of these systems usually requires the adoption of a high threshold value to avoid the inclusion of impostor patterns into the genuine biometric reference. In this study, we hypothesize that score normalization procedures, which have been used to improve the recognition performance of biometric systems through a better refinement of their decision, can also improve the overall performance of adaptive systems. With such a normalization, a better threshold choice could also be made, which would then increase the number of genuine samples used for adaptation. To the best of our knowledge, this is the first investigation towards the use of score normalization to enhance adaptive biometric systems dealing with the change of user features over time. Through a systematic experimental design tested on two behavioral biometric traits, the obtained results indeed support our conjecture. Moreover, the experimental results show that the performance gain brought by adaptation can have a higher overall impact than score normalization alone. (C) 2017 Elsevier Ltd. All rights reserved.

Images of the kidneys using dynamic contrast-enhanced magnetic resonance renography (DCE-MRR) contains unwanted complex organ motion due to respiration. This gives rise to motion artefacts that hinder the clinical assessment of kidney function. However, due to the rapid change in contrast agent within the DCE-MR image sequence, commonly used intensity-based image registration techniques are likely to fail. While semi-automated approaches involving human experts are a possible alternative, they pose significant drawbacks including inter-observer variability, and the bottleneck introduced through manual inspection of the multiplicity of images produced during a DCE-MRR study. To address this issue, we present a novel automated, registration-free movement correction approach based on windowed and reconstruction variants of dynamic mode decomposition (WR-DMD). Our proposed method is validated on ten different healthy volunteers’ kidney DCE-MRI data sets. The results, using block-matching-block evaluation on the image sequence produced by WR-DMD, show the elimination of $$99\%$$ of mean motion magnitude when compared to the original data sets, thereby demonstrating the viability of automatic movement correction using WR-DMD.

Biometric systems use personal biological or behavioural traits that can uniquely characterise an individual but this uniqueness property also becomes its potential weakness when the template characterising a biometric trait is stolen or compromised. To this end, we consider two strategies to improving biometric template protection and performance, namely, (1) using multiple privacy schemes and (2) using multiple matching algorithms. While multiple privacy schemes can improve the security of a biometric system by protecting its template; using multiple matching algorithms or similarly, multiple biometric traits along with their respective matching algorithms, can improve the system performance due to reduced intra-class variability. The above two strategies lead to a novel, ensemble system that is derived from multiple privacy schemes. Our findings suggest that, under the worst-case scenario evaluation where the key or keys protecting the template are stolen, multi-privacy protection scheme can outperform a single protection scheme as well as the baseline biometric system without template protection.

Assessing biometric performance is challenging because an experimental outcome depends on the choice of demographics and the chosen application scenario of an experiment. If one can quantify biometric samples into good, bad, and ugly categories for one application, the proportion of these categories is likely to be different for another application. As a result, a typical performance curve of a biometric experiment cannot generalize to another different application setting, even though the same system is used. We propose an algorithm that is capable of generalizing a biometric performance curve in terms of detection error tradeoff or equivalently receiver's operating characteristics, by allowing the user (system operator, policy-maker, and biometric researcher) to explicitly set the proportion of data differently. This offers the possibility for the user to simulate different operating conditions that can better match the setting of a target application. We demonstrated the utility of the algorithm in three scenarios, namely: 1) estimating the system performance under varying quality; 2) spoof and zero-effort attacks; and 3) cross-device matching. Based on the results of 1300 use-case experiments, we found that the quality of prediction on unseen (test) data, measured in terms of coverage, is typically between 60% and 80%, which is significantly better than random, that is, 50%.

Rendering a face recognition system robust is vital in order to safeguard it against spoof attacks carried out using printed pictures of a victim (also known as print attack) or a replayed video of the person (replay attack). A key property in distinguishing a live, valid access from printed media or replayed videos is by exploiting the information dynamics of the video content, such as blinking eyes, moving lips, and facial dynamics. We advance the state of the art in facial antispoofing by applying a recently developed algorithm called dynamic mode decomposition (DMD) as a general purpose, entirely data-driven approach to capture the above liveness cues. We propose a classification pipeline consisting of DMD, local binary patterns (LBPs), and support vector machines (SVMs) with a histogram intersection kernel. A unique property of DMD is its ability to conveniently represent the temporal information of the entire video as a single image with the same dimensions as those images contained in the video. The pipeline of DMD + LBP + SVM proves to be efficient, convenient to use, and effective. In fact only the spatial configuration for LBP needs to be tuned. The effectiveness of the methodology was demonstrated using three publicly available databases: (1) print-attack; (2) replay-attack; and (3) CASIA-FASD, attaining comparable results with the state of the art, following the respective published experimental protocols.

Student performance depends upon factors other than intrinsic ability, such as environment, socio-economic status, personality and familial-context. Capturing these patterns of influence may enable an educator to ameliorate some of these factors, or for governments to adjust social policy accordingly. In order to understand these factors, we have undertaken the exercise of predicting student performance, using a cohort of approximately 8,000 South African college students. They all took a number of tests in English and Maths. We show that it is possible to predict English comprehension test results from (1) other test results; (2) from covariates about self-efficacy, social economic status, and specific learning difficulties there are 100 survey questions altogether; (3) from other test results + covariates (combination of (1) and (2)); and from (4) a more advanced model similar to (3) except that the covariates are subject to dimensionality reduction (via PCA). Models 1-4 can predict student performance up to a standard error of 13-15%. In comparison, a random guess would have a standard error of 17%. In short, it is possible to conditionally predict student performance based on self-efficacy, socio-economic background, learning difficulties, and related academic test results.

Survey questionnaires are often heterogeneous because they contain both quantitative (numeric) and qualitative (text) responses, as well as missing values. While traditional, model-based methods are commonly used by clinicians, we deploy Self Organizing Maps (SOM) as a means to visualise the data. In a survey study aiming at understanding the self-care behaviour of 611 patients with Type-1 Diabetes, we show that SOM can be used to (1) identify co-morbidities; (2) to link self-care factors that are dependent on each other; and (3) to visualise individual patient profiles; In evaluation with clinicians and experts in Type-1 Diabetes, the knowledge and insights extracted using SOM correspond well to clinical expectation. Furthermore, the output of SOM in the form of a U-matrix is found to offer an interesting alternative means of visualising patient profiles instead of a usual tabular form.

Facial imaging has been largely addressed for automatic personal identification, in a variety of different environments. However, automatic face recognition becomes very challenging whenever the acquisition conditions are unconstrained. In this paper, a picture-specific cohort normalization approach, based on polynomial regression, is proposed to enhance the robustness of face matching under challenging conditions. A careful analysis is presented to better understand the actual discriminative power of a given cohort set. In particular, it is shown that the cohort polynomial regression alone conveys some discriminative information on the matching face pair, which is just marginally worse than the raw matching score. The influence of the cohort set size in the matching accuracy is also investigated. Further, tests performed on the Face Recognition Grand Challenge ver 2 database and the labeled faces in the wild database allowed to determine the relation between the quality of the cohort samples and cohort normalization performance. Experimental results obtained from the LFW data set demonstrate the effectiveness of the proposed approach to improve the recognition accuracy in unconstrained face acquisition scenarios.

Recent research has shown that collaborative representation-based classifier (CRC) can lead to promising results for the classification of face images. However, CRC is conducted in the original image space rather than the nonlinear high dimensional feature space in which features belonging to the same class are better grouped together and thus can be easily separable. To address this problem, this paper presents a novel classifier, Kernel Collaborative Representation-based Classifier (KCRC), by incorporating the kernel trick into the framework of CRC. Extensive experiments on both the AT&T and the FERET face databases demonstrate the priority of KCRC to CRC and several state-of-the-art methods.

We address reducing the mismatch between training and testing conditions for hands-free in-car speech recognition. It is well known that the distortions caused by background noise, channel effects, etc., are highly nonlinear in the log-spectral or cepstral domain. This letter introduces a joint sparse representation (JSR) to estimate the underlying clean feature vector from a noisy feature vector. Performing a joint dictionary learning by sharing the same representation coefficients, the proposed method intends to capture the complex relationships (or mapping functions) between clean and noisy speech. Speech recognition experiments on realistic in-car data demonstrate that the proposed method shows excellent recognition performance with a relative improvement of 39.4% compared with the “baseline” frontends.

There is mounting evidence about the benefit of tailoring a biometric authentication system to each user by postprocessing the system output at the score level, also known as client-specific score normalisation. Examples of these procedures are Z-norm and F-norm. These procedures can calibrate the uneven hypothesis space such that the dispropotionate false acceptance and false rejection errors are reduced after the calibration. The interest in studying these schemes is that they are applicable to any biometric authentication system regardless of the underlying biometric modality, and furthermore, potentially be extended to object recognition framed as a verification problem. We propose to further improve these procedures by adding additional client-specific terms that cannot be incorporated easily in their respective existing form. Experiments carried out on 13 face and speech systems show that both variants systematically outperform their respective score normalisation scheme (Z-norm or F-norm).

This paper proposes a unified framework for quality-based fusion of multimodal biometrics. Quality-dependent fusion algorithms aim to dynamically combine several classifier (biometric expert) outputs as a function of automatically derived (biometric) sample quality. Quality measures used for this purpose quantify the degree of conformance of biometric samples to some predefined criteria known to influence the system performance. Designing a fusion classifier to take quality into consideration is difficult because quality measures cannot be used to distinguish genuine users from impostors, i.e., they are nondiscriminative yet still useful for classification. We propose a general Bayesian framework that can utilize the quality information effectively. We show that this framework encompasses several recently proposed quality-based fusion algorithms in the literature--Nandakumar et al., 2006; Poh et al., 2007; Kryszczuk and Drygajo, 2007; Kittler et al., 2007; Alonso-Fernandez, 2008; Maurer and Baker, 2007; Poh et al., 2010. Furthermore, thanks to the systematic study concluded herein, we also develop two alternative formulations of the problem, leading to more efficient implementation (with fewer parameters) and achieving performance comparable to, or better than, the state of the art. Last but not least, the framework also improves the understanding of the role of quality in multiple classifier combination.

An increasing body of evidence suggests that cohort-based score normalization can improve the performance of biometric authentication. This approach relies on the use of N cohort biometric templates, which can be computationally expensive. We contribute to the advancement of cohort score normalization in two ways. First, we show both theoretically and empirically that the most similar and the most dissimilar cohort templates to a target user contain discriminative information. We then investigate the extraction of this information using polynomial regression. Extensive evaluation on the face and fingerprint modalities in the Biosecure DS2 dataset indicates that the proposed method outperforms the state-of-the-art cohort score normalization methods, while reducing the computation cost by as much as half.

One of the most promising ways to improve biometric person recognition is indisputably via information fusion, that is, to combine different sources of information. This pa per proposes a novel fusion paradigm that combines heterogeneous sources of information such as user-specific, cohort and quality information. Two formulations of this problem are proposed, differing in the assumption on the independence of the information sources. Unlike the more common multimodal/multi-algorithmic fusion, the novel paradigm has to deal with information that is not necessarily discriminative but still it is relevant. The methodology can be applied to any biometric system. Furthermore, extensive experiments based on 30 face and fingerprint experiments indicate that the performance gain with respect to the baseline system is about 30%. In contrast, solving this problem using conventional fusion paradigm leads to degraded results.

The performance of biometric systems can be significantly affected by changes in signal quality. In this paper, two types of changes are considered: change in acquisition environment and in sensing devices. We investigated three solutions: (i) model-level adaptation, (ii) score-level adaptation (normalisation), and (iii) the combination of the two, called “compound” adaptation. In order to cope with the above changing conditions, the model-level adaptation attempts to update the parameters of the expert systems (classifiers). This approach requires the authenticity of the candidate samples used for adaptation be known (corresponding to supervised adaptation), or can be estimated (unsupervised adaptation). In comparison, the score-level adaptation merely involves post processing the expert output, with the objective of rendering the associated decision threshold to be dependent only on the class priors despite the changing acquisition conditions. Since the above adaptation strategies treat the underlying biometric experts/classifiers as a black-box, they can be applied to any unimodal or multimodal biometric system, thus facilitating system-level integration and performance optimisation. Our contributions are: (i) proposal of compound adaptation; (ii) investigation and comparison of two different quality-dependent score normalisation strategies; and, (iii) empirical comparison of the merit of the above three solutions on the BANCA face (video) and speech database.

A new multimodal biometric database designed and acquired within the framework of the European BioSecure Network of Excellence is presented. It is comprised of more than 600 individuals acquired simultaneously in three scenarios: 1) over the Internet, 2) in an office environment with desktop PC, and 3) in indoor/outdoor environments with mobile portable hardware. The three scenarios include a common part of audio/video data. Also, signature and fingerprint data have been acquired both with desktop PC and mobile portable hardware. Additionally, hand and iris data were acquired in the second scenario using desktop PC. Acquisition has been conducted by 11 European institutions. Additional features of the BioSecure Multimodal Database (BMDB) are: two acquisition sessions, several sensors in certain modalities, balanced gender and age distributions, multimodal realistic scenarios with simple and quick tasks per modality, cross-European diversity, availability of demographic data, and compatibility with other multimodal databases. The novel acquisition conditions of the BMDB allow us to perform new challenging research and evaluation of either monomodal or multimodal biometric systems, as in the recent BioSecure Multimodal Evaluation campaign. A description of this campaign including baseline results of individual modalities from the new database is also given. The database is expected to be available for research purposes through the BioSecure Association during 2008.

Person recognition using facial features, e.g., mug-shot images, has long been used in identity documents. However, due to the widespread use of web-cams and mobile devices embedded with a camera, it is now possible to realize facial video recognition, rather than resorting to just still images. In fact, facial video recognition offers many advantages over still image recognition; these include the potential of boosting the system accuracy and deterring spoof attacks. This paper presents an evaluation of person identity verification using facial video data, organized in conjunction with the International Conference on Biometrics (ICB 2009). It involves 18 systems submitted by seven academic institutes. These systems provide for a diverse set of assumptions, including feature representation and preprocessing variations, allowing us to assess the effect of adverse conditions, usage of quality information, query selection, and template construction for video-to-video face authentication.

In multimodal biometric information fusion, it is common to encounter missing modalities in which matching cannot be performed. As a result, at the match score level, this implies that scores will be missing. We address the multimodal fusion problem involving missing modalities (scores) using support vector machines (SVMs) with the neutral point substitution (NPS) method. The approach starts by processing each modality using a kernel. When a modality is missing, at the kernel level, the missing modality is substituted by one that is unbiased with regards to the classification, called a neutral point. Critically, unlike conventional missing-data substitution methods, explicit calculation of neutral points may be omitted by virtue of their implicit incorporation within the SVM training framework. Experiments based on the publicly available Biosecure DS2 multimodal (scores) data set show that the SVM-NPS approach achieves very good generalization performance compared to the sum rule fusion, especially with severe missing modalities.

As biometric technology is rolled out on a larger scale, it will be a common scenario (known as cross-device matching) to have a template acquired by one biometric device used by another during testing. This requires a biometric system to work with different acquisition devices, an issue known as device interoperability. We further distinguish two subproblems, depending on whether the device identity is known or unknown. In the latter case, we show that the device information can be probabilistically inferred given quality measures (e.g., image resolution) derived from the raw biometric data. By keeping the template unchanged, cross-device matching can result in significant degradation in performance. We propose to minimize this degradation by using device-specific quality-dependent score normalization. In the context of fusion, after having normalized each device output independently, these outputs can be combined using the naive Bayes principal. We have compared and categorized several state-of-the-art quality-based score normalization procedures, depending on how the relationship between quality measures and score is modeled, as follows: 1) direct modeling; 2) modeling via the cluster index of quality measures; and 3) extending 2) to further include the device information (device-specific cluster index). Experimental results carried out on the Biosecure DS2 data set show that the last approach can reduce both false acceptance and false rejection rates simultaneously. Furthermore, the compounded effect of normalizing each system individually in multimodal fusion is a significant improvement in performance over the baseline fusion (without using any quality information) when the device information is given.

Automatically verifying the identity of a person by means of biometrics (e.g., face and fingerprint) is an important application in our day-to-day activities such as accessing banking services and security control in airports. To increase the system reliability, several biometric devices are often used. Such a combined system is known as a multimodal biometric system. This paper reports a benchmarking study carried out within the framework of the BioSecure DS2 (Access Control) evaluation campaign organized by the University of Surrey, involving face, fingerprint, and iris biometrics for person authentication, targeting the application of physical access control in a medium-size establishment with some 500 persons. While multimodal biometrics is a well-investigated subject in the literature, there exists no benchmark for a fusion algorithm comparison. Working towards this goal, we designed two sets of experiments: quality-dependent and cost-sensitive evaluation. The quality-dependent evaluation aims at assessing how well fusion algorithms can perform under changing quality of raw biometric images principally due to change of devices. The cost-sensitive evaluation, on the other hand, investigates how well a fusion algorithm can perform given restricted computation and in the presence of software and hardware failures, resulting in errors such as failure-to-acquire and failure-to-match. Since multiple capturing devices are available, a fusion algorithm should be able to handle this nonideal but nevertheless realistic scenario. In both evaluations, each fusion algorithm is provided with scores from each biometric comparison subsystem as well as the quality measures of both the template and the query data. The response to the call of the evaluation campaign proved very encouraging, with the submission of 22 fusion systems. To the best of our knowledge, this campaign is the first attempt to benchmark quality-based multimodal fusion algorithms. In the presence of changing image quality which may be due to a change of acquisition devices and/or device capturing configurations, we observe that the top performing fusion algorithms are those that exploit automatically derived quality measurements. Our evaluation also suggests that while using all the available biometric sensors can definitely increase the fusion performance, this comes at the expense of increased cost in terms of acquisition time, computation time, the physical cost of hardware, and its maintenance cost. As demonstrated in our experiments, a promising solution which minimizes the composite cost is sequential fusion, where a fusion algorithm sequentially uses match scores until a desired confidence is reached, or until all the match scores are exhausted, before outputting the final combined score.

Automatically verifying the identity of a person by means of biometrics (e.g., face and fingerprint) is an important application in our day-to-day activities such as accessing banking services and security control in airports. To increase the system reliability, several biometric devices are often used. This paper considers how auxiliary information such as the quality associated with a biometric sample and the device information can be used when combining the output of several biometric devices. Since both these sources of information are not discriminative in distinguishing genuine users from impostors, combining them is indeed a challenging problem. We advance the state of the art of multimodal biometric fusion in two ways: first, we unify several existing generative classifiers using Bayesian networks. Second, we propose a novel fusion classifier incorporating both the quality and device information simultaneously. Our experiments based on the Biosecure DS2 dataset suggests that the proposed classifier can systematically achieve the best generalization performance compared to currently available state-of-the-art classifiers.

While using more biometric traits in multimodal biometric fusion can effectively increase the system robustness, often, the cost associated to adding additional systems is not considered. In this paper, we propose an algorithm that can efficiently bound the biometric system error. This helps not only to speed up the search for the optimal system configuration by an order of magnitude but also unexpectedly to enhance the robustness to population mismatch. This suggests that bounding the error of biometric system from above can possibly be better than directly estimating it from the data. The latter strategy can be susceptible to spurious biometric samples and the particular choice of users. The efficiency of the proposal is achieved thanks to the use of Chernoff bound in estimating the authentication error. Unfortunately, such a bound assumes that the match scores are normally distributed, which is not necessarily the correct distribution model. We propose to transform simultaneously the class conditional match scores (genuine user or impostor scores) into ones that are more conforming to normal distributions using a modified criterion of the Box-Cox transform.

It has been shown that the authentication performance of a biometric system is dependent on the models/templates specific to a user. As a result, some users may be more easily recognized or impersonated than others. The various categories of users have been characterized by Doddington(1988). We refer to this unbalanced performance across users as the Doddington's zoo effect. In the context of fusion, we argue that this effect is system-dependent, i.e., a user model that is easily impersonated (a lamb) in one system may be easily recognized in another system (a sheep). While in principle, a fusion system could be trained to cope with the changing animal behavior of users from system to system, the lack of training data makes it impossible. We believe that one major cause of the Doddington's zoo effect is the variation of class conditional scores from one speaker model to another. We propose a two-level fusion framework that effectively realizes a fusion classifier adapted to each user. First, one applies aclient-specific(or model-specific) score normalization procedure to each of the system outputs to be combined. Then, one feeds the resulting normalized outputs to a fusion classifier (common to all users) as input to obtain a final combined score. Two existing model-specific score normalization procedures are considered in this framework, i.e., F- and Z-norms. In addition to them, a novel score normalization method called model-specific log-likelihood ratio (MS-LLR) is also proposed. While Z-norm is impostor-centric, i.e., it makes use of only the impostor score statistics, F-norm and the proposed MS-LLR are client-impostor centric, i.e., they consider both the client and impostor score statistics simultaneously. Our findings based on the XM2VTS and the NIST2005 databases show that when client-impostor centric normalization procedures are used to implement the proposed two-level fusion framework, the res

Evaluating biometric authentication performance is a complex task because the performance depends on the user set size, composition and the choice of samples. We propose to reduce the performance dependency of these three factors by deriving appropriate confidence intervals. In this study, we focus on deriving a confidence region based on the recently proposed expected performance curve (EPC). An EPC is different from the conventional DET or ROC curve because an EPC assumes that the test class-conditional (client and impostor) score distributions are unknown and this includes the choice of the decision threshold for various operating points. Instead, an EPC selects thresholds based on the training set and applies them on the test set. The proposed technique is useful, for example, to quote realistic upper and lower bounds of the decision cost function used in the NIST annual speaker evaluation. Our findings, based on the 24 systems submitted to the NIST2005 evaluation, show that the confidence region obtained from our proposed algorithm can correctly predict the performance of an unseen database with two times more users with an average coverage of 95% (over all the 24 systems). A coverage is the proportion of the unseen EPC covered by the derived confidence interval.

Biometric authentication performance is often depicted by a detection error trade-off (DET) curve. We show that this curve is dependent on the choice of samples available, the demographic composition and the number of users specific to a database. We propose a two-step bootstrap procedure to take into account the three mentioned sources of variability. This is an extension to the Bolle et al.'s bootstrap subset technique. Preliminary experiments on the NIST2005 and XM2VTS benchmark databases are encouraging, e.g., the average result across all 24 systems evaluated on NIST2005 indicates that one can predict, with more than 75 percent of DET coverage, an unseen DET curve with eight times more users. Furthermore, our finding suggests that with more data available, the confidence intervals become smaller and, hence, more useful.

Chimeric users have recently been proposed in the field of biometric person authentication as a way to overcome the problem of lack of real multimodal biometric databases as well as an important privacy issue - the fact that too many biometric modalities of a same person stored in a single location can present a higher risk of identity theft. While the privacy problem is indeed solved using chimeric users, it is still an open question of how such chimeric database can be efficiently used. For instance, the following two questions arise: i) is the performance measured on a chimeric database a good predictor of that measured on a real-user database, and, ii) can a chimeric database be exploited to improve the generalization performance of a fusion operator on a real-user database. Based on a considerable amount of empirical biometric person authentication experiments (21 real-user data sets and up to 21 times 1000 chimeric data sets and two fusion operators), our previous study (N. Poh and S. Bengio, 2005) answers no to the first question. The current study aims to answer the second question. Having tested on four classifiers and as many as 3380 face and speech bimodal fusion tasks (over 4 different protocols) on the BANCA database and four different fusion operators, this study shows that generating multiple chimeric databases does not degrade nor improve the performance of a fusion operator when tested on a real-user database with respect to using only a real-user database. Considering the possibly expensive cost involved in collecting the real-user multimodal data, our proposed approach is thus useful to construct a trainable fusion classifier while at the same time being able to overcome the problem of small size training data

The paper investigates a new client-dependent normalisation to improve biometric authentication systems. There exist many client-dependent score normalisation techniques, such as Z-Norm, D-Norm and T-Norm, that are applied to speaker authentication. Such normalisation is intended to adjust the variation across different client models. We propose "F-ratio" normalisation, or F-Norm, applied to face and speaker authentication systems. This normalisation requires only that as few as two client-dependent accesses are available (the more the better). Different from previous normalisation techniques, F-Norm considers the client and impostor distributions simultaneously. We show that F-ratio is a natural choice because it is directly associated to equal error rate. It has the effect of centering the client and impostor distributions such that a global threshold can be easily found. Another difference is that F-Norm actually "interpolates" between client-independent and client-dependent information by introducing a mixture parameter. This parameter can be optimised to maximise the class dispersion (the degree of separability between client and impostor distributions) while the aforementioned normalisation techniques cannot. The results of 13 unimodal experiments carried out on the XM2VTS multimodal database show that such normalisation is advantageous over Z-Norm, client-dependent threshold normalisation or no normalisation.

Multi-band, multi-stream and multi-modal approaches have proven to be very successful both in experiments and in real-life applications, among which speech recognition and biometric authentication are of particular interest here. However, there is a lack of a theoretical study to justify why and how they work, when one combines the streams at the feature or classifier score levels. In this paper, we attempt to cast a light onto the latter subject. While there exists literature discussing this aspect, a study on the relationship between correlation, variance reduction and equal error rate (often used in biometric authentication) has not been treated theoretically as done here, using the mean operator. Our findings suggest that combining several experts using the mean operator, multi-layer-perceptrons and support vector machines always perform better than the average performance of the underlying experts. Furthermore, in practice, most combined experts using the methods mentioned above perform better than the best underlying expert.

We present a simple yet effective way of improving a face verification system by generating multiple virtual samples from the unique image corresponding to an access request. These images are generated using simple geometric transformations. This method is often used during training to improve the accuracy of a neural network model by making it robust against minor translation, scale and orientation changes. Our main contribution is to introduce such a method during testing. By generating N images from one single image and propagating them to a trained network model, one obtains N scores. By merging these scores using a simple mean operator, we show that the variance of merged scores is decreased by a factor between 1 and N. An experiment is carried out on the XM2VTS database which achieves new state-of-the-art performances.