With the advent of deep generative models, there has been some recent interest in the manipulation of people’s facial features. This has many potential applications in fashion and biometrics. However, it is a complex task. Indeed, a modification of a given attribute should not have any effect on the others, identity should be preserved, and image quality should not be altered. So far, the evaluation of the proposed methods has been mostly qualitative, which is insufficient to demonstrate progress and performance. We propose a comprehensive evaluation framework to estimate the quality of facial attribute editing methods with respect to several criteria: image quality, effective modification of the targeted attribute, level of entanglement between attributes and identity preservation. Three generative models are used to demonstrate the proposed evaluation framework over three datasets and three editing methods, resulting in the analysis of over 29k generated images.

The detection of sexual explicit content in multimedia plays a critical role in digital forensic investigations, offering substantial benefits in both criminal justice and Cybersecurity contexts. Automated and manual identification of such content can aid in uncovering illegal activities, including the possession and distribution of child sexual abuse material (CSAM), human trafficking, and sexual harassment. As digital environments continue to evolve, integrating advanced technologies such as artificial intelligence and machine learning is essential to improve detection capabilities, ensure legal compliance, and uphold ethical standards. In order to contribute to this issue, we propose in this paper an original method for sexual explicit content detection in videos with light deep learning models for allowing a fast analysis and deployment. The proposed systems provide very good results ($>96 \%$of accuracy) compared to the state of the art on a significant dataset (LSPD). The proposed solution keeps a low processing time that is an important property for the digital investigation application.

We propose SARMA (Saliency Analysis with a Residual Mesh-based Architecture), a graph neural network designed to predict visual saliency on 3D surface meshes. Unlike traditional methods that rely on handcrafted geometric features, SARMA learns saliency patterns in an end-to-end fashion using residual Feature-Steered Graph Convolution (FeaStConv) layers. The network takes a mesh as input, represented as a graph of vertices and edges, and outputs per-vertex saliency values. To capture perceptually relevant geometry, we enrich the input features with discrete mean curvature alongside 3D coordinates. The model consists of three FeaStConv layers, each followed by a residual connection that stabilizes training and mitigates oversmoothing. We evaluate SARMA on the Schelling dataset of 3D models annotated with human saliency points, using PLCC and AUC as evaluation metrics. Our approach outperforms prior handcrafted and deep learning-based methods in terms of correlation with ground truth saliency, demonstrating the effectiveness of residual graph-based architectures for perceptual analysis of 3D shapes.

The advent of 3D data has revolutionized various industries, from architecture and engineering to healthcare and entertainment, enabling more precise simulations and realistic visualizations. However, 3D data is susceptible to noise and loss during generation and transmission, making quality assessment crucial for ensuring accuracy and usability. While existing literature addresses quality assessment for 3D point clouds and meshes separately, a gap exists in assessing the quality of 3D colored meshes due to the lack of reference datasets. This paper proposes an approach for No Reference 3D Colored Mesh Visual Quality Assessment (CMVQA), based on previous work related to quality assessment of 3D non colored meshes quality assessment. Our approach combines geometric and color features with spatial domain features extracted from mesh projections. Through extensive experiments and comparisons with full-reference metrics, including image quality metrics, our proposed approach demonstrates superior performance.

Fingerprint recognition is a common solution for user authentication in Cybersecurity. This paper deals with the context of the certification of fingerprint biometric systems. The increasing use of biometric systems makes their certification a mandatory step in their development to assess their behavior in a real situation use. It has been shown that certain parameters such as environmental conditions can have a significant impact on the performance of biometric systems. However, there are also non-controlled parameters that depend on the user’s state such as the quality of his biometric samples. In this paper, we propose a study that explores the performance of fingerprint systems across these parameters.

Networks have provided a representation for a wide range of real systems, including communication networks, money transfer networks and biological systems. Communities repre-sent fundamental structures for understanding the organization of real-world networks. Uncovering coherent groups in these networks is the goal of community detection. A community is a mesoscopic structure with nodes heavily connected in their groups by comparison to the nodes in other groups. Commu-nities might also overlap as they may share one or multiple nodes. This paper lays the foundation for an application on transactional multigraphs (networks of financial transactions in which nodes can be linked with multiple edges), through the discovery of communities. Due to their complexity, our goal is to find the most effective way of simplifying multigraphs to weighted graphs, while preserving properties of the network. We tested five weights' calculation function and community detection algorithms were applied. A comparison of the outputs based on extrinsic and intrinsic evaluation metrics is then held.

Keystroke dynamics is one solution to enhance the security of password authentication without adding any disruptive handling for users. Industries are looking for more security without impacting too much user experience. Considered as a friction-less solution, keystroke dynamics is a powerful solution to increase trust during user authentication without adding charge to the user. In this paper, we address the problem of user authentication considering the keystroke dynamics modality. We proposed a new approach based on the conversion of behavioral biometrics data (time series) into a 3D image. This transformation process keeps all the characteristics of the behavioral signal. The time series do not receive any filtering operation with this transformation and the method is bijective. This transformation allows us to train images based on convolutional neural networks. We evaluate the performance of the authentication system in terms of Equal Error Rate (EER) on a significant dataset and we show the efficiency of the proposed approach on a multi-instance system.

Ensuring security on biometric systems has always been a high priority concern. Certification of biometric systems involves the testing of the system’s performance and its resistance to spoof attacks. The anti-spoofing test implies the creation and scan of multiples physical spoofs. This requests laboratory expertise and high amount of time for spoofs creation. In this paper, we propose a new solution based on deep learning to translate genuine fingerprint images and transform them into what they would look like if they were created from known spoof materials usually involved in fingerprint spoofing tests. Digitally Synthetized Fingerprint Spoofs (DSFS) help to cover a larger number of spoofs materials than it would be possible to physically fabricate in a given time. Validation method shows that synthetized images are as good as real spoofs considering their quality.

Networks have provided a representation for a wide range of real systems, including communication flow, money transfer or biological systems, to mention just a few. Communities represent fundamental structures for understanding the organization of real-world networks. Uncovering coherent groups in these networks is the goal of community detection. A community is a mesoscopic structure with nodes heavily connected within their groups by comparison to the nodes in other groups. Communities might also overlap as they may share one or multiple nodes. Evaluating the results of a community detection algorithm is an equally important task. This paper introduces metrics for evaluating overlapping community detection. The idea of introducing new metrics comes from the lack of efficiency and adequacy of state-of-the-art metrics for overlapping communities. The new metrics are tested both on simulated data and standard datasets and are compared with existing metrics.

Abstract In this paper, we present a full-reference quality assessment metric based on the information of visual saliency. The saliency information is provided under the form of degrees associated to each vertex of the surface mesh. From these degrees, statistical attributes reflecting the structures of the reference and distorted meshes are computed. These are used by four comparisons functions genetically optimized that quantify the structure differences between a reference and a distorted mesh. We also present a statistical comparison study of six full-reference quality assessment metrics for 3D meshes. We compare the objective metrics results with humans subjective scores of quality considering the 3D meshes in one hand and the distorsion types in the other hand. Also, we show which metrics are statistically superior to their counterparts. For these comparisons we use the Spearman Rank Ordered Correlation Coefficient and the hypothetic test of Student (ttest). To attest the pertinence of the proposed approach, a comparison with a ground truth saliency and an application associated to the assessment of the visual rendering of smoothing algorithms are presented. Experimental results show that the proposed metric is very competitive with the state-of-the-art.

Nowadays, users interact with computer systems. Behavioral biometrics consists of analyzing user's interactions for identification and verification applications. This approach could be very useful for enhancing security and improving user experience and many privacy concerns are also related. In this paper, we address the problem of user identification considering their behaviors. How efficient are classical machine learning methods on such data? What about deep learning approaches? We illustrate this work on two behavioral modalities namely human activity using smartphones and keystroke dynamics on a laptop. Since the accuracy rates of most behavioral biometrics modalities are lower than morphological ones, we consider two approaches for these modalities that can be represented as time series: classical machine learning and deep learning techniques. We intend to show that many algorithms can obtain very good performance for different modalities without any specific tuning to the considered modality. This comparative analysis allows us to show that behavioral biometrics can be used for security applications (i.e. who is accessing the company information system) but could be a privacy concern as a user could be identified while navigating on the Internet.

Biometrics testing has for objective to determine the performance of a biometric system in order to guarantee security and user experience requirements. Providing trust in biometric systems is a key for many manufacturers. The performance is usually measured through the computation of matching scores between legitimate and impostor samples from a given database. Different bias in particular those linked to the environmental conditions can modify the performance of a biometric system. In this paper, we study the impact of acquisition conditions on fingerprint systems considering at the same time the quality and accuracy. We defined an own-made database controlling the acquisition conditions and we observe the behavior of three different matchers on these biometric data. Experimental results allow us to quantity their impact on performance and draw conclusions for testing biometric systems.

Due to the advantages in privacy and efficiency requirements, minutiae template based matching is the dominant technique among the authentication approaches of fingerprint image and its performance fully relies on the quality of the input fingerprint image. In this case, it is reasonable to consider qualifying fingerprint with fingerprint minutiae template information extracted from fingerprint image, particularly when using for embedded applications due the limited memory. In fact, the speed of fingerprint recognition increases with the decrease of the size of database. For these reasons, a new confidence measure called Minutia Confidence Index (MiCI) for each minutia of the template is proposed. This index predicts the importance and the usefulness of each minutia with respect to the others in the template. It takes into account only minutiae template information (i.e., x and y coordinates, the type and the orientation). MiCI score is a value between 0 and 1, where highest values are for the mostly relevant minutiae in the template whereas lowest values are for less important ones. This measure has been applied in the template reduction use case on Fingerprint Verification Competition (FVC) and SFINGEO databases and demonstrated its capability to reach high performance.

The quality assessment of biometric samples is a crucial issue in biometrics, indeed, many studies showed its significant impact on the subsequent performance of the biometric system. Many metrics have been proposed and studied in the literature in order to quantify their usefulness. In this paper, we propose to merge different metrics in order to improve the utility estimation of the quality assessment. We use the enrollment selection validation approach in order to compute the utility estimation of the fused metrics. We show the efficiency of the proposed approach comparing with 7 well known metrics on the 12 FVC datasets and 5 synthesized SFinGE-based databases with two matching algorithms. Experimental results show a good improvement on the fused metric to better qualify the quality of digital fingerprints. Those results demonstrate the effectiveness of the approach.

As a trending method for the authentication, biometrics tends to be integrated in various devices, and in particular in smartphones. If the evaluation is performed on operational device, the biometric sample and algorithm are not reachable by the assessors. So, these latter have to perform an evaluation on a system considered as a black box. This kind of evaluation implies numerous manual comparison. This paper proposes a methodology to perform an evaluation of biometric black boxes. Two preliminary experiments were performed in order to determine an optimized conduct. This paper describes the used methodology to perform evaluation on black boxes systems, and the results obtained on the systems under test.

The overall performance of iris recognition systems is affected by the quality of acquired iris sample images. Due to the development of imaging technologies, visible wavelength iris recognition gained a lot of attention in the past few years. However, iris sample quality of unconstrained imaging conditions is a more challenging issue compared to the traditional near infrared iris biometrics. Therefore, measuring the quality of such iris images is essential in order to have good quality samples for iris recognition. In this paper, we investigate whether general purpose no-reference image quality metrics can assess visible wavelength iris sample quality.

Fingerprint analysis is a very important issue in biometry. The minutiae representation of a fingerprint is the most used modality to identify people or authorize access when using a biometric system. In this paper, we propose some features based on triangle parameters from the Delaunay triangulation of minutiae. We show the benefit of these features to recognize the type of a fingerprint without any access to the associated fingerprint image.

We propose in this paper a novel perceptual viewpoint-independent metric for the quality assessment of 3D meshes. This full-reference objective metric relies on the method proposed by Wang et al. [1] that compares the structural informations between an original signal and a distorted one. In order to extract the structural informations of a 3D mesh, we use a multi-scale visual saliency map on which we compute the local statistics. The experimental results attest the strong correlation between the objective scores provided by our metric and the human judgments. Also, comparisons with the state-of-the-art prove that our metric is very competitive.

Image quality is an important factor for automated fingerprint identification systems (AFIS) because the matching performance could be significantly affected by poor quality samples. Most of the existing studies mainly focus on calculating a quality index via either a single feature or a combination of multiple features, and some others achieve this purpose with learning approaches which may depend on a prior-knowledge of matching performance. In this paper, a general framework for estimating fingerprint image quality is proposed by fusing features in segmentation phase. The quality index is indicated by a ratio of the pixel number of the integrated foreground area to the size (pixel number) of the fingerprint image. The potential advantage of this framework is that it could be improved by integrating other segmentation approaches or quality features rather than fusing them in a more complicated manner. The experiment is performed with several fingerprint datasets created via different sensors. Experimental results obtained from a dual evaluation approach demonstrate the validity of the proposed method in improving the overall performance.

Mesh surface saliency detection is an important preprocessing step for many 3D applications. This paper proposes a novel saliency computation method by the use of a local vertex descriptor in the form an adaptive patch. This descriptor is used as a basis for similarity measurement and integrated into a weighted multi-scale saliency computation. Experimental results demonstrate that the proposed approach achieves competitive and innovative results, especially for 3d colored meshes.

Nowadays, when someone wants to make a payment with a smartcard, the user has to enter a pin code to be identified. Only biometrics is able to authenticate a user; yet biometric information is sensitive. To ensure the security and privacy of biometric data, OCC (On-Card-Comparison) has been proposed. This approach consists in storing biometric data in a secure zone on a smartcard and computing the verification decision in a Secure Element (SE). The purpose of this paper is to propose an evaluation platform for testing biometric systems such as the analysis of performance and security on biometric OCC. Based on two examples, we illustrate its different uses in an operationnal context. The first example focus on the ”Quality module” which allows to choose the enrollment by considering the fingerprint quality with one proposed metric. The second one addresses the minutiae reduction of the fingerprint template when the number of minutiae is higher than expected by the OCC.

This article proposes a new quality assessment method of fingerprint, represented by only a set of minutiae points. The proposed quality metric is modeled with the convex-hull and Delaunay triangulation of the minutiae points. The validity of this quality metric is verified on several Fingerprint Verification Competition (FVC) databases by referring to an image-based metric from the state of the art (considered as the reference). The experiments of the utility-based evaluation approach demonstrate that the proposed quality metric is able to generate a desired result. We reveal the possibility of assessing fingerprint quality when only the minutiae template is available.

Biometric sample quality assessment approaches are generally designed in terms of utility property due to the potential difference between human perception of quality and the biometric quality requirements for a recognition system. This study proposes a utility based quality assessment method of fingerprints by considering several complementary aspects: 1) Image quality assessment without any reference which is consistent with human conception of inspecting quality, 2) Textural features related to the fingerprint image and 3) minutiae features which correspond to the most used information for matching. The proposed quality metric is obtained by a linear combination of these features and is validated with a reference metric using different approaches. Experiments performed on several trial databases show the benefit of the proposed fingerprint quality metric.

We propose a blind (no reference or NR) video quality evaluation model that is nondistortion specific. The approach relies on a spatio-temporal model of video scenes in the discrete cosine transform domain, and on a model that characterizes the type of motion occurring in the scenes, to predict video quality. We use the models to define video statistics and perceptual features that are the basis of a video quality assessment (VQA) algorithm that does not require the presence of a pristine video to compare against in order to predict a perceptual quality score. The contributions of this paper are threefold. 1) We propose a spatio-temporal natural scene statistics (NSS) model for videos. 2) We propose a motion model that quantifies motion coherency in video scenes. 3) We show that the proposed NSS and motion coherency models are appropriate for quality assessment of videos, and we utilize them to design a blind VQA algorithm that correlates highly with human judgments of quality. The proposed algorithm, called video BLIINDS, is tested on the LIVE VQA database and on the EPFL-PoliMi video database and shown to perform close to the level of top performing reduced and full reference VQA algorithms.

A crucial step in the assessment of an image compression method is the evaluation of the perceived quality of the compressed images. Typically, researchers ask observers to rate perceived image quality directly and use these rating measures, averaged across observers and images, to assess how image quality degrades with increasing compression. These ratings in turn are used to calibrate and compare image quality assessment algorithms intended to predict human perception of image degradation. There are several drawbacks to using such omnibus measures. First, the interpretation of the rating scale is subjective and may differ from one observer to the next. Second, it is easy to overlook compression artifacts that are only present in particular kinds of images. In this paper, we use a recently developed method for assessing perceived image quality, maximum likelihood difference scaling (MLDS), and use it to assess the performance of a widely-used image quality assessment algorithm, multiscale structural similarity (MS-SSIM). MLDS allows us to quantify supra-threshold perceptual differences between pairs of images and to examine how perceived image quality, estimated through MLDS, changes as the compression rate is increased. We apply the method to a wide range of images and also analyze results for specific images. This approach circumvents the limitations inherent in the use of rating methods, and allows us also to evaluate MS-SSIM for different classes of visual image. We show how the data collected by MLDS allow us to recalibrate MS-SSIM to improve its performance.

We develop an efficient, general-purpose, blind/noreference image quality assessment (NR-IQA) algorithm using a natural scene statistics (NSS) model of discrete cosine transform (DCT) coefficients. The algorithm is computationally appealing, given the availability of platforms optimized for DCT computation. The approach relies on a simple Bayesian inference model to predict image quality scores given certain extracted features. The features are based on an NSS model of the image DCT coefficients. The estimated parameters of the model are utilized to form features that are indicative of perceptual quality. These features are used in a simple Bayesian inference approach to predict quality scores. The resulting algorithm, which we name BLIINDS-II, requires minimal training and adopts a simple probabilistic model for score prediction. Given the extracted features from a test image, the quality score that maximizes the probability of the empirically determined inference model is chosen as the predicted quality score of that image. When tested on the LIVE IQA database, BLIINDS-II is shown to correlate highly with human judgments of quality, at a level that is competitive with the popular SSIM index.

In this paper, we describe a recently developed method for assessing perceived image quality, Maximum Likelihood Difference Scaling (MLDS), and use it to assess the performance of MS-SSIM on compression distored images. MLDS allows us to quantify supra-threshold perceptual differences between pairs of images and to examine how perceived image quality, estimated through MLDS, changes as the compression rate is increased. We show how the data collected by MLDS allows us to recalibrate MS-SSIM to improve its performance.

We propose an efficient, general-purpose, distortion-agnostic, blind/no-reference image quality assessment (NR-IQA) algorithm based on a natural scene statistics model of discrete cosine transform (DCT) coefficients. The algorithm is computationally appealing, given the availability of platforms optimized for DCT computation. We propose a generalized parametric model of the extracted DCT coefficients. The parameters of the model are utilized to predict image quality scores. The resulting algorithm, which we name BLIINDS-II, requires minimal training and adopts a simple probabilistic model for score prediction. When tested on the LIVE IQA database, BLIINDS-II is shown to correlate highly with human visual perception of quality, at a level that is even competitive with the powerful full-reference SSIM index.

General-purpose no-reference image quality assessment approaches still lag the advances in full-reference methods. Most no-reference methods are either distortion specific (i.e. they quantify one or more distortions such as blur, blockiness, or ringing), or they train a learning machine based on a large number of features. In this approach, we propose a discrete cosine transform (DCT) statistics-based support vector machine (SVM) approach based on only 3 features in the DCT domain. The approach extracts a very small number of features and is entirely in the DCT domain, making it computationally convenient. The results are shown to correlate highly with human visual perception of quality.

In this paper, we present a method for estimating the amount of Gaussian noise and Gaussian blur in a distorted image. Our method is based on the MS-SSIM framework which, although designed to measure image quality, is used to estimate the amount of blur and noise in a degraded image given a reference image. Various search strategies such as Newton, Simplex, and brute force search are presented and rigorously compared. Based on quantitative results, we show that the amount of blur and noise in a distorted image can be recovered with an accuracy up to 0.95% and 5.40%, respectively. To our knowledge, such precision has never been achieved before.

Abstract—The development of general-purpose no-reference approaches to image quality assessment still lags recent advances in full-reference methods. Additionally, most no-reference or blind approaches are distortion-specific, meaning they assess only a specific type of distortion assumed present in the test image (such as blockiness, blur, or ringing). This limits their application domain. Other approaches rely on training a machine learning algorithm. These methods however, are only as effective as the features used to train their learning machines. Towards ameliorating this we introduce the BLIINDS index (BLind Image Integrity Notator using DCT Statistics) which is a no-reference approach to image quality assessment that does not assume a specific type of distortion of the image. It is based on predicting image quality based on observing the statistics of local discrete cosine transform coefficients, and it requires only minimal training. The method is shown to correlate highly with human perception of quality. Index Terms—Anisotropy, discrete cosine transform, kurtosis, natural scene statistics, no-reference quality assessment. I.

A model selection method based on tabu search is proposed to build support vector machines (binary decision functions) of reduced complexity and efficient generalization. The aim is to build a fast and efficient support vector machines classifier. A criterion is defined to evaluate the decision function quality which blends recognition rate and the complexity of a binary decision functions together. The selection of the simplification level by vector quantization, of a feature subset and of support vector machines hyperparameters are performed by tabu search method to optimize the defined decision function quality criterion in order to find a good sub-optimal model on tractable times.

A tool for diagnosis assistance by automatic segmentation of microscopic cellular images is introduced. This method is based on an automatic segmentation technique combining (with the Dempster-Shafer rule) the results obtained by support vector machines (SVM) applied within different color spaces. This combination is performed by integrating uncertainties and redundancies for each color space. Those uncertainties are computed as a posteriori probabilities according to the SVM obtained results. An improvement of the final segmentation quality is performed by taking into account the inconsistencies of several pixel classifications

Summary form only given. A nonuniform quantization scheme, based on the perceptual relevance of each channel signal component, has been developed to exploit HVS properties. This strategy is based on the exploitation of the visual masking effect by performing a subband decomposition. An evaluation of the quality was performed using psychophysical measures for validation. The obtained results denote an improvement of the image quality when images are compressed combining intra channel masking visual effect. Furthermore, a compression gain of 10% can be reached depending on the image content.

In order to optimize the codebook used by the vector quantization compression scheme, we have developed a process based on the max-min algorithm. This process optimizes color space partitioning from vector blocks selected iteratively within the training set according to three algorithms. The partitioning algorithm is based on the nearest neighbor query. The selection algorithm searches the furthest color of the nearest vector block of the training set already computed. A centroid process generates the codebook in refining the vector block selection. In order to counterbalance cases of study for which the centroid process modifies the vector block selection, we have introduced three tests. These tests restrict the training set from which representative colors can be selected.