Abstract: The comprehension of animal behavior, especially human behavior, is one of the most ancient and studied problems since the beginning of civilization. The big list of factors that interact to determine a person action require the collaboration of different disciplines, such as psichology, biology, or sociology. In the last years the analysis of human behavior has received great attention also from the computer vision community, given the latest advances in the acquisition of human motion data from image sequences.
Despite the increasing availability of that data, there still exists a gap towards obtaining a conceptual representation of the obtained observations. Human behavior analysis is based on a qualitative interpretation of the results, and therefore the assignment of concepts to quantitative data is linked to a certain ambiguity.
This Thesis tackles the problem of obtaining a proper representation of human behavior in the contexts of computer vision and animation. On the one hand, a good behavior model should permit the recognition and explanation the observed activity in image sequences. On the other hand, such a model must allow the generation of new synthetic instances, which model the behavior of virtual agents.
First, we propose methods to automatically learn the models from observations. Given a set of quantitative results output by a vision system, a normal behavior model is learnt. This results provides a tool to determine the normality or abnormality of future observations. However, machine learning methods are unable to provide a richer description of the observations. We confront this problem by means of a new method that incorporates prior knowledge about the enviornment and about the expected behaviors. This framework, formed by the reasoning engine FMTL and the modeling tool SGT allows the generation of conceptual descriptions of activity in new image sequences. Finally, we demonstrate the suitability of the proposed framework to simulate behavior of virtual agents, which are introduced into real image sequences and interact with observed real agents, thereby easing the generation of augmented reality sequences.
The set of approaches presented in this Thesis has a growing set of potential applications. The analysis and description of behavior in image sequences has its principal application in the domain of smart video--surveillance, in order to detect suspicious or dangerous behaviors. Other applications include automatic sport commentaries, elderly monitoring, road traffic analysis, and the development of semantic video search engines. Alternatively, behavioral virtual agents allow to simulate accurate real situations, such as fires or crowds. Moreover, the inclusion of virtual agents into real image sequences has been widely deployed in the games and cinema industries.

Abstract: Symbol spotting problem requires feature extraction strategies able to generalize from training samples and to localize the target object while discarding most part of the image. In the case of document analysis, symbol spotting techniques have to deal with a high variability of symbols' appearance. In this paper, we propose the Circular Blurred Shape Model descriptor. Feature extraction is performed capturing the spatial arrangement of significant object characteristics in a correlogram structure. Shape information from objects is shared among correlogram regions, being tolerant to the irregular deformations. Descriptors are learnt using a cascade of classifiers and Abadoost as the base classifier. Finally, symbol spotting is performed by means of a windowing strategy using the learnt cascade over plan and old musical score documents. Spotting and multi-class categorization results show better performance comparing with the state-of-the-art descriptors.

Abstract: This paper presents a novel approach for estimating the shortest Euclidean distance from a given point to the corresponding implicit quadric fitting surface. It first estimates the orthogonal orientation to the surface from the given point; then the shortest distance is directly estimated by intersecting the implicit surface with a line passing through the given point according to the estimated orthogonal orientation. The proposed orthogonal distance estimation is easily obtained without increasing computational complexity; hence it can be used in error minimization surface fitting frameworks. Comparisons of the proposed metric with previous approaches are provided to show both improvements in CPU time as well as in the accuracy of the obtained results. Surfaces fitted by using the proposed geometric distance estimation and state of the art metrics are presented to show the viability of the proposed approach.

Abstract: Abstract. Text detection in urban scenes is a hard task due to the high variability of text appearance: different text fonts, changes in the point of view, or partial occlusion are just a few problems. Text detection can be specially suited for georeferencing business, navigation, tourist assistance, or to help visual impaired people. In this paper, we propose a general methodology to deal with the problem of text detection in outdoor scenes. The method is based on learning spatial information of gradient based features and Census Transform images using a cascade of classifiers. The method is applied in the context of Mobile Mapping systems, where a mobile vehicle captures urban image sequences. Moreover, a cover data set is presented and tested with the new methodology. The results show high accuracy when detecting multi-linear text regions with high variability of appearance, at same time that it preserves a low false alarm rate compared to classical approaches

Abstract: In this paper we propose the construction of a visual content layer which describes the visual appearance of geographic locations in a city. We captured, by means of a Mobile Mapping system, a huge set of georeferenced images (>500K) which cover the whole city of Barcelona. For each image, hundreds of region descriptions are computed off-line and described as a hash code. All this information is extracted without an object of reference, which allows to search for any type of objects using their visual appearance. A new Visual Content layer is built over Google Maps, allowing the object recognition information to be organized and fused with other content, like satellite images, street maps, and business locations.

Abstract: In this work, we present a first approach to activity patterns discovery by mean of topic models. Using motion data collected with a wearable device we prototype, TheBadge, we analyse raw accelerometer data using Latent Dirichlet Allocation (LDA), a particular instantiation of topic models. Results show that for particular values of the parameters necessary for applying LDA to a countinous dataset, good accuracies in activity classification can be achieved.

Abstract: We describe an architecture for a multi-camera, multi-resolution surveillance system. The aim is to support a set of distributed static and pan-tilt-zoom (PTZ) cameras and visual tracking algorithms, together with a central supervisor unit. Each camera (and possibly pan-tilt device) has a dedicated process and processor. Asynchronous interprocess communications and archiving of data are achieved in a simple and effective way via a central repository, implemented using an SQL database. Visual tracking data from static views are stored dynamically into tables in the database via client calls to the SQL server. A supervisor process running on the SQL server determines if active zoom cameras should be dispatched to observe a particular target, and this message is effected via writing demands into another database table. We show results from a real implementation of the system comprising one static camera overviewing the environment under consideration and a PTZ camera operating under closed-loop velocity control, which uses a fast and robust level-set-based region tracker. Experiments demonstrate the effectiveness of our approach and its feasibility to multi-camera systems for intelligent surveillance.