Roberto Tronci

Competition on counter measures to 2-D facial spoofing attacks

Spoofing identities using photographs is one of the most common techniques to attack 2-D face recognition systems. There seems to exist no comparative studies of different techniques using the same protocols and data. The motivation behind this competition is to compare the performance of different state-of-the-art algorithms on the same database using a unique evaluation method. Six different teams from universities around the world have participated in the contest. Use of one or multiple techniques from motion, texture analysis and liveness detection appears to be the common trend in this competition. Most of the algorithms are able to clearly separate spoof attempts from real accesses. The results suggest the investigation of more complex attacks.

HMMPayl: An intrusion detection system based on Hidden Markov Models

Nowadays the security of Web applications is one of the key topics in Computer Security. Among all the solutions that have been proposed so far, the analysis of the HTTP payload at the byte level has proven to be effective as it does not require the detailed knowledge of the applications running on the Web server. The solutions proposed in the literature actually achieved good results for the detection rate, while there is still room for reducing the false positive rate. To this end, in this paper we propose HMMPayl, an IDS where the payload is represented as a sequence of bytes, and the analysis is performed using Hidden Markov Models (HMM). The algorithm we propose for feature extraction and the joint use of HMM guarantee the same expressive power of n -gram analysis, while allowing to overcome its computational complexity. In addition, we designed HMMPayl following the Multiple Classifiers System paradigm to provide for a better classification accuracy, to increase the difficulty of evading the IDS, and to mitigate the weaknesses due to a non optimal choice of HMM parameters. Experimental results, obtained both on public and private datasets, show that the analysis performed by HMMPayl is particularly effective against the most frequent attacks toward Web applications (such as XSS and SQL-Injection). In particular, for a fixed false positive rate, HMMPayl achieves a higher detection rate respect to previously proposed approaches it has been compared with.

Fusion of multiple clues for photo-attack detection in face recognition systems

We faced the problem of detecting 2-D face spoofing attacks performed by placing a printed photo of a real user in front of the camera. For this type of attack it is not possible to relay just on the face movements as a clue of vitality because the attacker can easily simulate such a case, and also because real users often show a “low vitality” during the authentication session. In this paper, we perform both video and static analysis in order to employ complementary information about motion, texture and liveness and consequently to obtain a more robust classification.

The 2nd competition on counter measures to 2D face spoofing attacks

As a crucial security problem, anti-spoofing in biometrics, and particularly for the face modality, has achieved great progress in the recent years. Still, new threats arrive inform of better, more realistic and more sophisticated spoofing attacks. The objective of the 2nd Competition on Counter Measures to 2D Face Spoofing Attacks is to challenge researchers to create counter measures effectively detecting a variety of attacks. The submitted propositions are evaluated on the Replay-Attack database and the achieved results are presented in this paper.

ImageHunter: A Novel Tool for Relevance Feedback in Content Based Image Retrieval

Nowadays, a very large number of digital image archives is easily produced thanks to the wide diffusion of personal digital cameras and mobile devices with embedded cameras. Thus, personal computers, personal storage units, as well as photo-sharing and social-network websites, are rapidly becoming the repository for thousands, or even billions of images (i.e., more than 100 million photos are uploaded every day on the social site Facebook). As a consequence, there is an increasing need for tools enabling the semantic search, classification, and retrieval of images. The use of meta-data associated to images solves the problems only partially, as the process of assigning reliable meta-data to images is not trivial, is slow, and closely related to whom performed the task. One solution for effective image search and retrieval is to combine content-based analysis with feedbacks from the users. In this chapter we present Image Hunter, a tool that implements a Content Based Image Retrieval (CBIR) engine with a Relevance Feedback mechanism. Thanks to a user friendly interface the tool is especially suited to unskilled users. In addition, the modular structure permits the use of the same core both in web-based and stand alone applications.

Diversity in Ensembles of Codebooks for Visual Concept Detection

Visual codebooks generated by the quantization of local descriptors allows building effective feature vectors for image archives. Codebooks are usually constructed by clustering a subset of image descriptors from a set of training images. In this paper we investigate the effect of the combination of an ensemble of different codebooks, each codebook being created by using different pseudo-random techniques for subsampling the set of local descriptors. Despite the claims in the literature on the gain attained by combining different codebook representations, reported results on different visual detection tasks show that the diversity is quite small, thus allowing for modest improvement in performance w.r.t. the standard random subsampling procedure, and calling for further investigation on the use of ensemble approaches in this context.

Dynamic score combination of binary experts

The combination of experts is used to improve the performance of a classification system. In this paper we propose three dynamic score combination techniques that embed the selection and the fusion approach for combining experts. The proposed techniques are designed to combine binary experts that output a score measuring the degree of similarity to the positive class. Reported results on two biometric dataset show the effectiveness of the proposed techniques in terms of AUC and EER.

Dynamic linear combination of two-class classifiers

In two-class problems, the linear combination of the outputs (scores) of an ensemble of classifiers is widely used to attain high performance. In this paper we investigate some techniques aimed at dynamically estimate the coefficients of the linear combination on a pattern per pattern basis. We will show that such a technique allows providing better performance than those of static combination techniques, whose parameters are computed beforehand. The coefficients of the linear combination are dynamically computed according to the Wilcoxon-Mann-Whitney statistic. Reported results on a multi-modal biometric dataset show that the proposed dynamic mechanism allows attaining very low error rates when high level of precision are required.

A Study on the Evaluation of Relevance Feedback in Multi-tagged Image Datasets

This paper proposes a study on the evaluation of relevance feedback approaches when a multi-tagged dataset is available. The aim of this study is to verify how the relevance feedback works in a real-word scenario, i.e. by taking into account the multiple concepts represented by the query image. To this end, we first assessed how relevance feedback mechanisms adapt the search when the same image is used for retrieving different concepts. Then, we investigated the scenarios in which the same image is used for retrieving multiple concepts. The experimental results shows that relevance feedback can effectively focus the search according to the users feedback even if the query image provides a rough example of the target concept. We also propose two performance measures aimed at comparing the accuracy of retrieval results when the same image is used as a prototype for a number of different concepts.

Index driven combination of multiple biometric experts for AUC maximisation

A biometric system produces a matching score representing the degree of similarity of the input with the set of templates for that user. If the score is greater than a prefixed threshold, then the user is accepted, otherwise the user is rejected. Typically, the performance is evaluated in terms of the Receiver Operating Characteristic (ROC) curve, where the correct acceptance rate is plotted against the false authentication rate. A measure used to characterise a ROC curve is the Area Under the Curve (AUC), the larger the AUC, the better the ROC. In order to increase the reliability of authentication through biometrics, the combination of different biometric systems is currently investigated by researchers. In this paper two open problems are addressed: the selection of the experts to be combined and their related performance improvements. To this end we propose an index to be used for the experts selection to be combined, with the aim of the AUC maximisation. Reported results on FVC2004 dataset show the effectiveness of the proposed index.