Abstract: We address the problem of calibration of workers whose task is to label patterns with continuous variables, which arises for instance in labeling images of videos of humans with continuous traits. Worker bias is particularly dicult to evaluate and correct when many workers contribute just a few labels, a situation arising typically when labeling is crowd-sourced. In the scenario of labeling short videos of people facing a camera with personality traits, we evaluate the feasibility of the pairwise ranking method to alleviate bias problems. Workers are exposed to pairs of videos at a time and must order by preference. The variable levels are reconstructed by fitting a Bradley-Terry-Luce model with maximum likelihood. This method may at first sight, seem prohibitively expensive because for N videos, p = N (N-1)/2 pairs must be potentially processed by workers rather that N videos. However, by performing extensive simulations, we determine an empirical law for the scaling of the number of pairs needed as a function of the number of videos in order to achieve a given accuracy of score reconstruction and show that the pairwise method is a ordable. We apply the method to the labeling of a large scale dataset of 10,000 videos used in the ChaLearn Apparent Personality Trait challenge.

Abstract: The representation (or feature) learning has been an emerging concept in the last years, since it collects a set of techniques that are present in any theoretical or practical methodology referring to artificial intelligence. In computer vision, a very common representation has adopted the form of the well-known Bag of Visual Words. This representation appears implicitly in most approaches where images are described, and is also present in a huge number of areas and domains: image content retrieval, pedestrian detection, human-computer interaction, surveillance, e-health, and social computing, amongst others. The early stages of this dissertation provide an approach for learning visual representations inside evolutionary algorithms, which consists of evolving weighting schemes to improve the BoVW representations for the task of recognizing categories of videos and images. Thus, we demonstrate the applicability of the most common weighting schemes, which are often used in text mining but are less frequently found in computer vision tasks. Beyond learning these visual representations, we provide an approach based on fusion strategies for learning spatiotemporal representations, from multimodal data obtained by depth sensors. Besides, we specially aim at the evolutionary and dynamic modelling, where the temporal factor is present in the nature of the data, such as video sequences of gestures and actions. Indeed, we explore the effects of probabilistic modelling for those approaches based on dynamic programming, so as to handle the temporal deformation and variance amongst video sequences of different categories. Finally, we integrate dynamic programming and generative models into an evolutionary computation framework, with the aim of learning Bags of SubGestures (BoSG) representations and hence to improve the generalization capability of standard gesture recognition approaches. The results obtained in the experimentation demonstrate, first, that evolutionary algorithms are useful for improving the representation of BoVW approaches in several datasets for recognizing categories in still images and video sequences. On the other hand, our experimentation reveals that both, the use of dynamic programming and generative models to align video sequences, and the representations obtained from applying fusion strategies in multimodal data, entail an enhancement on the performance when recognizing some gesture categories. Furthermore, the combination of evolutionary algorithms with models based on dynamic programming and generative approaches results, when aiming at the classification of video categories on large video datasets, in a considerable improvement over standard gesture and action recognition approaches. Finally, we demonstrate the applications of these representations in several domains for human analysis: classification of images where humans may be present, action and gesture recognition for general applications, and in particular for conversational settings within the field of restorative justice

Abstract: The amount of multimedia data collected in museum databases is growing fast, while the capacity of museums to display information to visitors is acutely limited by physical space. Museums must seek the perfect balance of information given on individual pieces in order to provide sufficient information to aid visitor understanding while maintaining sparse usage of the walls and guaranteeing high appreciation of the exhibit. Moreover, museums often target the interests of average visitors instead of the entire spectrum of different interests each individual visitor might have. Finally, visiting a museum should not be an experience contained in the physical space of the museum but a door opened onto a broader context of related artworks, authors, artistic trends, etc. In this paper we describe the MNEMOSYNE system that attempts to address these issues through a new multimedia museum experience. Based on passive observation, the system builds a profile of the artworks of interest for each visitor. These profiles of interest are then used to drive an interactive table that personalizes multimedia content delivery. The natural user interface on the interactive table uses the visitor’s profile, an ontology of museum content and a recommendation system to personalize exploration of multimedia content. At the end of their visit, the visitor can take home a personalized summary of their visit on a custom mobile application. In this article we describe in detail each component of our approach as well as the first field trials of our prototype system built and deployed at our permanent exhibition space at LeMurate (http://www.lemurate.comune.fi.it/lemurate/) in Florence together with the first results of the evaluation process during the official installation in the National Museum of Bargello (http://www.uffizi.firenze.it/musei/?m=bargello).

Abstract: In this paper we propose a contextual rescoring method for predicting the position of body parts in a human pose estimation framework. A set of poselets is incorporated in the model, and their detections are used to extract spatial and score-related features relative to other body part hypotheses. A method is proposed for the automatic discovery of a compact subset of poselets that covers the different poses in a set of validation images while maximizing precision. A rescoring mechanism is defined as a set-based boosting classifier that computes a new score for each body joint detection, given its relationship to detections of other body joints and mid-level parts in the image. This new score is incorporated in the pictorial structure model as an additional unary potential, following the recent work of Pishchulin et al. Experiments on two benchmarks show comparable results to Pishchulin et al. while reducing the size of the mid-level representation by an order of magnitude, reducing the execution time by 68 % accordingly.

Abstract: The extraction of relevant information from historical document collections is one of the key steps in order to make these documents available for access and searches. The usual approach combines transcription and grammars in order to extract semantically meaningful entities. In this paper, we describe a new method to obtain word categories directly from non-preprocessed handwritten word images. The method can be used to directly extract information, being an alternative to the transcription. Thus it can be used as a first step in any kind of syntactical analysis. The approach is based on Convolutional Neural Networks with a Spatial Pyramid Pooling layer to deal with the different shapes of the input images. We performed the experiments on a historical marriage record dataset, obtaining promising results.

Abstract: A key topic in classification is the accuracy loss produced when the data distribution in the training (source) domain differs from that in the testing (target) domain. This is being recognized as a very relevant problem for many
computer vision tasks such as image classification, object detection, and object category recognition. In this paper, we present a novel domain adaptation method that leverages multiple target domains (or sub-domains) in a hierarchical adaptation tree. The core idea is to exploit the commonalities and differences of the jointly considered target domains.
Given the relevance of structural SVM (SSVM) classifiers, we apply our idea to the adaptive SSVM (A-SSVM), which only requires the target domain samples together with the existing source-domain classifier for performing the desired adaptation. Altogether, we term our proposal as hierarchical A-SSVM (HA-SSVM).
As proof of concept we use HA-SSVM for pedestrian detection, object category recognition and face recognition. In the former we apply HA-SSVM to the deformable partbased model (DPM) while in the rest HA-SSVM is applied to multi-category classifiers. We will show how HA-SSVM is effective in increasing the detection/recognition accuracy with respect to adaptation strategies that ignore the structure of the target data. Since, the sub-domains of the target data are not always known a priori, we shown how HA-SSVM can incorporate sub-domain discovery for object category recognition.

Abstract: Random Forest (RF) is a successful paradigm for learning classifiers due to its ability to learn from large feature spaces and seamlessly integrate multi-class classification, as well as the achieved accuracy and processing efficiency. However, as many other classifiers, RF requires domain adaptation (DA) provided that there is a mismatch between the training (source) and testing (target) domains which provokes classification degradation. Consequently, different RF-DA methods have been proposed, which not only require target-domain samples but revisiting the source-domain ones, too. As novelty, we propose three inherently different methods (Node-Adapt, Path-Adapt and Tree-Adapt) that only require the learned source-domain RF and a relatively few target-domain samples for DA, i.e. source-domain samples do not need to be available. To assess the performance of our proposals we focus on image-based object detection, using the pedestrian detection problem as challenging proof-of-concept. Moreover, we use the RF with expert nodes because it is a competitive patch-based pedestrian model. We test our Node-, Path- and Tree-Adapt methods in standard benchmarks, showing that DA is largely achieved.

Abstract: Best student paper award.
Pain detection using facial images is of critical importance in many Health applications. Since pain is a spatiotemporal process, recent works on this topic employ facial spatiotemporal features to detect pain. These systems extract such features from the entire area of the face. In this paper, we show that by employing super-pixels we can divide the face into three regions, in a way that only one of these regions (about one third of the face) contributes to the pain estimation and the other two regions can be discarded. The experimental results on the UNBCMcMaster database show that the proposed system using this single region outperforms state-of-the-art systems in detecting no-pain scenarios, while it reaches comparable results in detecting weak and severe pain scenarios.