Maria De Marsico

Evaluating web sites: exploiting user's expectations

A new goal-based approach to measure usability of web sites is presented, strongly taking into account the customers expectations, which are often hardly foreseeable as a whole. After a general discussion on web site design issues, we present a short survey of evaluation methods currently used for web sites. We next introduce a new taxonomy of site categories in a three-dimensional space, derived from Aristotles rhetorical triangle, including different aspects of the site designers goals. In our approach, we use this taxonomy to identify a number of sites belonging to the same category, in order to carry out a comparative analysis of their features. This analysis is the basis for a two-shot generation of a form for the evaluation of that category of sites. In the first shot, the users fill a generic evaluation form, acquainting them with sites characteristics. They are next asked to perform specific tasks of their choice, according to what they expect from a site of the given category. They note their impressions and list those features they found useful; the analysis of their comments is exploited to formulate statements specific to the given category, to be added to the initial form (second shot). We found that the responses to the second, expanded form, provide more comprehensive criteria for site evaluation, and turn helpful to precisely locate flaws in site functionalities. After testing, our methodology has proved very promising and may be applied for the evaluation of any other site category, most of all those providing a set of special services.

FIRME: Face and Iris Recognition for Mobile Engagement☆

Abstract Mobile devices, namely phones and tablets, have long gone “smart”. Their growing use is both a cause and an effect of their technological advancement. Among the others, their increasing ability to store and exchange sensitive information, has caused interest in exploiting their vulnerabilities, and the opposite need to protect users and their data through secure protocols for access and identification on mobile platforms. Face and iris recognition are especially attractive, since they are sufficiently reliable, and just require the webcam normally equipping the involved devices. On the contrary, the alternative use of fingerprints requires a dedicated sensor. Moreover, some kinds of biometrics lend themselves to uses that go beyond security. Ambient intelligence services bound to the recognition of a user, as well as social applications, such as automatic photo tagging on social networks, can especially exploit face recognition. This paper describes FIRME (Face and Iris Recognition for Mobile Engagement) as a biometric application based on a multimodal recognition of face and iris, which is designed to be embedded in mobile devices. Both design and implementation of FIRME rely on a modular architecture, whose workflow includes separate and replaceable packages. The starting one handles image acquisition. From this point, different branches perform detection, segmentation, feature extraction, and matching for face and iris separately. As for face, an antispoofing step is also performed after segmentation. Finally, results from the two branches are fused. In order to address also security-critical applications, FIRME can perform continuous reidentification and best sample selection. To further address the possible limited resources of mobile devices, all algorithms are optimized to be low-demanding and computation-light.

Mobile Iris Challenge Evaluation (MICHE)-I, biometric iris dataset and protocols

A new dataset of iris images acquired by mobile devices can support researchers.MICHE-I will assist with developing continuous authentication to counter spoofing.The dataset includes images from different mobile devices, sessions and conditions. We introduce and describe here MICHE-I, a new iris biometric dataset captured under uncontrolled settings using mobile devices. The key features of the MICHE-I dataset are a wide and diverse population of subjects, the use of different mobile devices for iris acquisition, realistic simulation of the acquisition process (including noise), several data capture sessions separated in time, and image annotation using metadata. The aim of MICHE-I dataset is to make up the starting core of a wider dataset that we plan to collect, with the further aim to address interoperability, both in the sense of matching samples acquired with different devices and of assessing the robustness of algorithms to the use of devices with different characteristics. We discuss throughout the merits of MICHE-I with regard to biometric dimensions of interest including uncontrolled settings, demographics, interoperability, and real-world applications. We also consider the potential for MICHE-I to assist with developing continuous authentication aimed to counter adversarial spoofing and impersonation, when the bar for uncontrolled settings raises even higher for proper and effective defensive measures.

Moving face spoofing detection via 3D projective invariants

Face recognition provides many advantages compared with other available biometrics, but it is particularly subject to spoofing. The most accurate methods in literature addressing this problem, rely on the estimation of the three-dimensionality of faces, which heavily increase the whole cost of the system. This paper proposes an effective and efficient solution to problem of face spoofing. Starting from a set of automatically located facial points, we exploit geometric invariants for detecting replay attacks. The presented results demonstrate the effectiveness and efficiency of the proposed indices.

Robust face recognition after plastic surgery using local region analysis

Face recognition in real-world applications is often hindered by uncontrolled settings including pose, expression, and illumination changes, and/or ageing. Additional challenges related to changes in facial appearance due to plastic surgery have become apparent recently. We exploit the fact that plastic surgery bears on appearance in a non-uniform fashion using a recognition approach that integrates information derived from local region analysis. We implemented and evaluated the performance of two new integrative methods, FARO and FACE, which are based on fractals and a localized version of a correlation index, respectively Experimental results confirm the expectation that face recognition is indeed challenged by the effects of plastic surgery. The same experimental results also show that both FARO and FACE compare favourably against standard face recognition methods such as PCA and LDA.

Noisy Iris Recognition Integrated Scheme

One of the most challenging issues in iris recognition is the design of techniques able to ensure high accuracy even in adverse conditions. This paper deals with an approach to iris matching based on the combination of local features: Linear Binary Patterns (LBP) and discriminable textons (BLOBs) are presently exploited. The techniques have been refined ad hoc, to allow the extraction of significant discriminative features, even with images captured in variable visible light conditions, and affected by noise due to distance/resolution or to scarce user collaboration (blurring, off-axis iris, occlusion by eyelashes and eyelids). The obtained results strongly motivate further investigations along this line, most of all the addition of more local features.

IS_IS: Iris Segmentation for Identification Systems

Advances in processing procedures make the iris a realistic candidate to the role of biometry of the future. Precise detection and segmentation for such biometry are a crucial ongoing research area. We propose an iris segmentation technique and show that it is more reliable than existent ones.

Face: face analysis for Commercial Entities

Though face recognition gained significant attention and credibility in the last decade, quite few commercial applications were able to benefit from this. In this paper we propose FACE (Face analysis for Commercial Entities), a robust framework to address face analysis, aiming at supporting the activities of various Commercial Entities. In particular, we present two case studies, with the related experimental results which sustain the presented approach.

A proposal toward the development of accessible e-learning content by human involvement

Most of the existing efforts for supporting the design, preparation, and deployment of accessible e-learning applications propose guidelines that primarily address technical accessibility issues. Little, if any, consideration is given to the real actors involved in the learning experience, such as didactical experts and disabled learners. Moreover, implementing artifacts addressed to the e-learning world requires a wide range of particular skills which are related not only to technical but also to didactical, pedagogical, usability, and accessibility aspects of the produced material. This paper argues that the know-how of a number of stakeholders should be blended into a joint design activity, and that it should be possible to determine the role of each participant in the successive phases of the development lifecycle of e-learning applications. The paper sketches the methodological guidelines of a design framework based on involving the users with disabilities, as well as pedagogical experts, in the development process. The novelty of this proposal mainly stems from being built up around the core of strategies and choices specifically bound to accessibility requirements. Characteristic elements of learner-centered design are then further integrated into processes and methodologies which are typical of participatory and contextual design approaches. Following such guidelines, it will be possible to gain a deeper understanding of the requirements and of the operational context of people needing accessible material, either as learners or educators. The underlying objective is to increase the potential to realize learning systems that better meet different user needs and that provide a more satisfying learning experience. Moreover, when people get involved in the development process, they gain a sense of ownership of the system and are therefore more likely to accept and “promote” it.

A Framework to Support Social-Collaborative Personalized e-Learning

We propose a comprehensive framework to support the personalization and adaptivity of courses in e-learning environments where the traditional activity of individual study is augmented by social-collaborative and group based educational activities. The framework aims to get its pedagogical significance from the Vygotskij Theory; it points out a minimal set of requirements to meet, in order to allow its implementations based on modules possibly constituted by independent e-learning software systems, all collaborating under a common interface.