Abstract
With new trends like 3D and deep learning alternatives for face recognition becoming more popular, it becomes essential to establish a complete benchmark for the evaluation of such algorithms, in a wide variety of data sources and non-ideal scenarios. We propose a new RGB-depth-infrared (RGB-D-IR) dataset, RealFace, acquired with the novel Intel® RealSense TM collection of sensors, and characterized by multiple variations in pose, lighting and disguise. As baseline for future works, we assess the performance of multiple deep and “shallow” feature descriptors. We conclude that our dataset presents some relevant challenges and that deep feature descriptors present both higher robustness in RGB images, as well as an interesting margin for improvement in alternative sources, such as depth and IR. © Springer International Publishing AG 2016.

Abstract
In many applications, false positives (type I error) and false negatives (type II) have different impact. In medicine, it is not considered as bad to falsely diagnosticate someone healthy as sick (false positive) as it is to diagnosticate someone sick as healthy (false negative). But we are also willing to accept some rate of false negatives errors in order to make the classification task possible at all. Where the line is drawn is subjective and prone to controversy. Usually, this compromise is given by a cost matrix where an exchange rate between errors is defined. For many reasons, however, it might not be natural to think of this trade-off in terms of relative costs. We explore novel learning paradigms where this trade-off can be given in the form of the amount of false negatives we are willing to tolerate. The classifier then tries to minimize false positives while keeping false negatives within the acceptable bound. Here we consider classifiers based on kernel density estimation, gradient descent modifications and applying a threshold to classifying and ranking scores.

Abstract
Despite the existence of patterns able to discriminate between consensual and non-consensual intercourse, the relevance of genital lesions in the corroboration of a legal rape complaint is currently under debate in many countries. The testimony of the physicians when assessing these lesions has been questioned in court due to several factors (e.g. a lack of comprehensive knowledge of lesions, wide spectrum of background area, among others). Thereby, it is relevant to provide automated tools to support the decision process in an objective manner. In this work, we compare traditional handcrafted features and deep learning techniques in the automated processing of colposcopic images for genital injury detection. Positive results where achieved by both paradigms in segmentation and classification subtasks, being traditional and deep models the best strategy for each subtask type respectively.

Abstract
Reject option is a technique used to improve classifier's reliability in decision support systems. It consists in withholding the automatic classification of an item, if the decision is considered not sufficiently reliable. The rejected item is then handled by a different classifier or by a human expert. The vast majority of the works on this issue has been concerned with the development of reject option mechanisms to be used by supervised learning architectures (e.g., MLP, LVQ or SVM). In this paper, however, we aim at proposing alternatives to this view, which are based on the self-organizing map (SOM), originally an unsupervised learning scheme, but that has also been successfully used in the design of prototype-based classifiers. The basic hypothesis we defend is that it is possible to design SOM-based classifiers endowed with reject option mechanisms whose performances are comparable to or better than those achieved by standard supervised classifiers. For this purpose, we carried out a comprehensively evaluation of the proposed SOM-based classifiers on two synthetic and three real-world datasets. The obtained results suggest that the proposed SOM-based classifiers consistently outperform standard supervised classifiers.

Abstract
Generalized additive models are well-known as a powerful and palatable predictive modelling technique. Scorecards, the discretized version of generalized additive models, are a long-established method in the industry, due to its balance between simplicity and performance. Scorecards are easy to apply and easy to understand. Moreover, in spite of their simplicity, scorecards can model nonlinear relationships between the inputs and the value to be predicted. In the scientific community, scorecards have been largely overlooked in favor of more recent models such as neural networks or support vector machines. In this paper, we address scorecard development, introducing a new formulation more suitable to support regularization. We tackle both the binary and the ordinal data classification problems. In both settings, the proposed methodology shows advantages when evaluated using real datasets.

Abstract
Biometric systems are vulnerable to different kinds of attacks. Particularly, the systems based on iris are vulnerable to direct attacks consisting on the presentation of a fake iris to the sensor trying to access the system as it was from a legitimate user. The analysis of some countermeasures against this type of attacking scheme is the problem addressed in the present paper. Several state-of-the-art methods were implemented and included in a feature selection framework so as to determine the best cardinality and the best subset that conducts to the highest classification rate. Three different classifiers were used: Discriminant analysis, K nearest neighbours and Support Vector Machines. The implemented methods were tested in existing databases for iris liveness purposes (Biosec and Clarkson) and in a new fake database which was constructed for evaluation of iris liveness detection methods in the mobile scenario. The results suggest that this new database is more challenging than the others. Therefore, improvements are required in this line of research to achieve good performance in real world mobile applications.