Prototipo de herramienta para el inventariado automático de piezas de Lego basado en visión artificial

Permitida la difusión del código bajo los términos de la licencia BSD de tres cláusulas.

Computer vision based eyewear selector

[EN]This paper describes a low-cost system that allows the user to visualize different glasses models in live video. The user can also move the glasses to adjust its position on the face. The system, which runs at 9.5 frames/s on general-purpose hardware, has a homeostatic module that keeps image parameters controlled. This is achieved by using a camera with motorized zoom, iris, white balance, etc. This feature can be specially useful in environments with changing illumination and shadows, like in an optical shop. The system also includes a face and eye detection module and a glasses management module.

Useful computer vision techniques for human-robot interaction

[ES]This paper describes some simple but useful computer vision techniques for human-robot interaction. First, an omnidirectional camera setting is described that can detect people in the surroundings of the robot, giving their angular positions and a rough estimate of the distance. The device can be easily built with inexpensive components. Second, we comment on a color-based face detection technique that can alleviate skin-color false positives. Third, a simple head nod and shake detector is described, suitable for detecting affirmative/negative, approval/dissaproval, understanding/disbelief head gestures.

Deep learning for computer vision: a comparison between convolutional neural networks and hierarchical temporal memories on object recognition tasks

In recent years, Deep Learning techniques have shown to perform well on a large variety of problems both in Computer Vision and Natural Language Processing, reaching and often surpassing the state of the art on many tasks. The rise of deep learning is also revolutionizing the entire field of Machine Learning and Pattern Recognition pushing forward the concepts of automatic feature extraction and unsupervised learning in general. However, despite the strong success both in science and business, deep learning has its own limitations. It is often questioned if such techniques are only some kind of brute-force statistical approaches and if they can only work in the context of High Performance Computing with tons of data. Another important question is whether they are really biologically inspired, as claimed in certain cases, and if they can scale well in terms of "intelligence". The dissertation is focused on trying to answer these key questions in the context of Computer Vision and, in particular, Object Recognition, a task that has been heavily revolutionized by recent advances in the field. Practically speaking, these answers are based on an exhaustive comparison between two, very different, deep learning techniques on the aforementioned task: Convolutional Neural Network (CNN) and Hierarchical Temporal memory (HTM). They stand for two different approaches and points of view within the big hat of deep learning and are the best choices to understand and point out strengths and weaknesses of each of them. CNN is considered one of the most classic and powerful supervised methods used today in machine learning and pattern recognition, especially in object recognition. CNNs are well received and accepted by the scientific community and are already deployed in large corporation like Google and Facebook for solving face recognition and image auto-tagging problems. HTM, on the other hand, is known as a new emerging paradigm and a new meanly-unsupervised method, that is more biologically inspired. It tries to gain more insights from the computational neuroscience community in order to incorporate concepts like time, context and attention during the learning process which are typical of the human brain. In the end, the thesis is supposed to prove that in certain cases, with a lower quantity of data, HTM can outperform CNN.

Evaluating the Impact of the Number of Access Points in Mobile Robots Localization Using Artificial Neural Networks

Localization is information of fundamental importance to carry out various tasks in the mobile robotic area. The exact degree of precision required in the localization depends on the nature of the task. The GPS provides global position estimation but is restricted to outdoor environments and has an inherent imprecision of a few meters. In indoor spaces, other sensors like lasers and cameras are commonly used for position estimation, but these require landmarks (or maps) in the environment and a fair amount of computation to process complex algorithms. These sensors also have a limited field of vision. Currently, Wireless Networks (WN) are widely available in indoor environments and can allow efficient global localization that requires relatively low computing resources. However, the inherent instability in the wireless signal prevents it from being used for very accurate position estimation. The growth in the number of Access Points (AP) increases the overlap signals areas and this could be a useful means of improving the precision of the localization. In this paper we evaluate the impact of the number of Access Points in mobile nodes localization using Artificial Neural Networks (ANN). We use three to eight APs as a source signal and show how the ANNs learn and generalize the data. Added to this, we evaluate the robustness of the ANNs and evaluate a heuristic to try to decrease the error in the localization. In order to validate our approach several ANNs topologies have been evaluated in experimental tests that were conducted with a mobile node in an indoor space.

Cognitive Comparative Advantage and the Organization of Work: Lessons from Herbert Simon's Vision of the Future

In a marvelous but somewhat neglected paper, 'The Corporation: Will It Be Managed by Machines?' Herbert Simon articulated from the perspective of 1960 his vision of what we now call the New Economy the machine-aided system of production and management of the late twentieth century. Simon's analysis sprang from what I term the principle of cognitive comparative advantage: one has to understand the quite different cognitive structures of humans and machines (including computers) in order to explain and predict the tasks to which each will be most suited. Perhaps unlike Simon's better-known predictions about progress in artificial intelligence research, the predictions of this 1960 article hold up remarkably well and continue to offer important insights. In what follows I attempt to tell a coherent story about the evolution of machines and the division of labor between humans and machines. Although inspired by Simon's 1960 paper, I weave many other strands into the tapestry, from classical discussions of the division of labor to present-day evolutionary psychology. The basic conclusion is that, with growth in the extent of the market, we should see humans 'crowded into' tasks that call for the kinds of cognition for which humans have been equipped by biological evolution. These human cognitive abilities range from the exercise of judgment in situations of ambiguity and surprise to more mundane abilities in spatio-temporal perception and locomotion. Conversely, we should see machines 'crowded into' tasks with a well-defined structure. This conclusion is not based (merely) on a claim that machines, including computers, are specialized idiots-savants today because of the limits (whether temporary or permanent) of artificial intelligence; rather, it rests on a claim that, for what are broadly 'economic' reasons, it will continue to make economic sense to create machines that are idiots-savants.

Vision-based vehicle detection and tracking with a mobile camera using a statistical framework

En esta tesis se aborda la detección y el seguimiento automático de vehículos mediante técnicas de visión artificial con una cámara monocular embarcada. Este problema ha suscitado un gran interés por parte de la industria automovilística y de la comunidad científica ya que supone el primer paso en aras de la ayuda a la conducción, la prevención de accidentes y, en última instancia, la conducción automática. A pesar de que se le ha dedicado mucho esfuerzo en los últimos años, de momento no se ha encontrado ninguna solución completamente satisfactoria y por lo tanto continúa siendo un tema de investigación abierto. Los principales problemas que plantean la detección y seguimiento mediante visión artificial son la gran variabilidad entre vehículos, un fondo que cambia dinámicamente debido al movimiento de la cámara, y la necesidad de operar en tiempo real. En este contexto, esta tesis propone un marco unificado para la detección y seguimiento de vehículos que afronta los problemas descritos mediante un enfoque estadístico. El marco se compone de tres grandes bloques, i.e., generación de hipótesis, verificación de hipótesis, y seguimiento de vehículos, que se llevan a cabo de manera secuencial. No obstante, se potencia el intercambio de información entre los diferentes bloques con objeto de obtener el máximo grado posible de adaptación a cambios en el entorno y de reducir el coste computacional. Para abordar la primera tarea de generación de hipótesis, se proponen dos métodos complementarios basados respectivamente en el análisis de la apariencia y la geometría de la escena. Para ello resulta especialmente interesante el uso de un dominio transformado en el que se elimina la perspectiva de la imagen original, puesto que este dominio permite una búsqueda rápida dentro de la imagen y por tanto una generación eficiente de hipótesis de localización de los vehículos. Los candidatos finales se obtienen por medio de un marco colaborativo entre el dominio original y el dominio transformado. Para la verificación de hipótesis se adopta un método de aprendizaje supervisado. Así, se evalúan algunos de los métodos de extracción de características más populares y se proponen nuevos descriptores con arreglo al conocimiento de la apariencia de los vehículos. Para evaluar la efectividad en la tarea de clasificación de estos descriptores, y dado que no existen bases de datos públicas que se adapten al problema descrito, se ha generado una nueva base de datos sobre la que se han realizado pruebas masivas. Finalmente, se presenta una metodología para la fusión de los diferentes clasificadores y se plantea una discusión sobre las combinaciones que ofrecen los mejores resultados. El núcleo del marco propuesto está constituido por un método Bayesiano de seguimiento basado en filtros de partículas. Se plantean contribuciones en los tres elementos fundamentales de estos filtros: el algoritmo de inferencia, el modelo dinámico y el modelo de observación. En concreto, se propone el uso de un método de muestreo basado en MCMC que evita el elevado coste computacional de los filtros de partículas tradicionales y por consiguiente permite que el modelado conjunto de múltiples vehículos sea computacionalmente viable. Por otra parte, el dominio transformado mencionado anteriormente permite la definición de un modelo dinámico de velocidad constante ya que se preserva el movimiento suave de los vehículos en autopistas. Por último, se propone un modelo de observación que integra diferentes características. En particular, además de la apariencia de los vehículos, el modelo tiene en cuenta también toda la información recibida de los bloques de procesamiento previos. El método propuesto se ejecuta en tiempo real en un ordenador de propósito general y da unos resultados sobresalientes en comparación con los métodos tradicionales. ABSTRACT This thesis addresses on-road vehicle detection and tracking with a monocular vision system. This problem has attracted the attention of the automotive industry and the research community as it is the first step for driver assistance and collision avoidance systems and for eventual autonomous driving. Although many effort has been devoted to address it in recent years, no satisfactory solution has yet been devised and thus it is an active research issue. The main challenges for vision-based vehicle detection and tracking are the high variability among vehicles, the dynamically changing background due to camera motion and the real-time processing requirement. In this thesis, a unified approach using statistical methods is presented for vehicle detection and tracking that tackles these issues. The approach is divided into three primary tasks, i.e., vehicle hypothesis generation, hypothesis verification, and vehicle tracking, which are performed sequentially. Nevertheless, the exchange of information between processing blocks is fostered so that the maximum degree of adaptation to changes in the environment can be achieved and the computational cost is alleviated. Two complementary strategies are proposed to address the first task, i.e., hypothesis generation, based respectively on appearance and geometry analysis. To this end, the use of a rectified domain in which the perspective is removed from the original image is especially interesting, as it allows for fast image scanning and coarse hypothesis generation. The final vehicle candidates are produced using a collaborative framework between the original and the rectified domains. A supervised classification strategy is adopted for the verification of the hypothesized vehicle locations. In particular, state-of-the-art methods for feature extraction are evaluated and new descriptors are proposed by exploiting the knowledge on vehicle appearance. Due to the lack of appropriate public databases, a new database is generated and the classification performance of the descriptors is extensively tested on it. Finally, a methodology for the fusion of the different classifiers is presented and the best combinations are discussed. The core of the proposed approach is a Bayesian tracking framework using particle filters. Contributions are made on its three key elements: the inference algorithm, the dynamic model and the observation model. In particular, the use of a Markov chain Monte Carlo method is proposed for sampling, which circumvents the exponential complexity increase of traditional particle filters thus making joint multiple vehicle tracking affordable. On the other hand, the aforementioned rectified domain allows for the definition of a constant-velocity dynamic model since it preserves the smooth motion of vehicles in highways. Finally, a multiple-cue observation model is proposed that not only accounts for vehicle appearance but also integrates the available information from the analysis in the previous blocks. The proposed approach is proven to run near real-time in a general purpose PC and to deliver outstanding results compared to traditional methods.

EPIC (Entorno de Páncreas Artificial con Interfaz Clínica)

La Diabetes Mellitus se define como el trastorno del metabolismo de los carbohidratos, resultante de una producción insuficiente o nula de insulina en las células beta del páncreas, o la manifestación de una sensibilidad reducida a la insulina por parte del sistema metabólico. La diabetes tipo 1 se caracteriza por la nula producción de insulina por la destrucción de las células beta del páncreas. Si no hay insulina en el torrente sanguíneo, la glucosa no puede ser absorbida por las células, produciéndose un estado de hiperglucemia en el paciente, que a medio y largo plazo si no es tratado puede ocasionar severas enfermedades, conocidos como síndromes de la diabetes. La diabetes tipo 1 es una enfermedad incurable pero controlable. La terapia para esta enfermedad consiste en la aplicación exógena de insulina con el objetivo de mantener el nivel de glucosa en sangre dentro de los límites normales. Dentro de las múltiples formas de aplicación de la insulina, en este proyecto se usará una bomba de infusión, que unida a un sensor subcutáneo de glucosa permitirá crear un lazo de control autónomo que regule la cantidad optima de insulina aplicada en cada momento. Cuando el algoritmo de control se utiliza en un sistema digital, junto con el sensor subcutáneo y bomba de infusión subcutánea, se conoce como páncreas artificial endocrino (PAE) de uso ambulatorio, hoy día todavía en fase de investigación. Estos algoritmos de control metabólico deben de ser evaluados en simulación para asegurar la integridad física de los pacientes, por lo que es necesario diseñar un sistema de simulación mediante el cual asegure la fiabilidad del PAE. Este sistema de simulación conecta los algoritmos con modelos metabólicos matemáticos para obtener una visión previa de su funcionamiento. En este escenario se diseñó DIABSIM, una herramienta desarrollada en LabViewTM, que posteriormente se trasladó a MATLABTM, y basada en el modelo matemático compartimental propuesto por Hovorka, con la que poder simular y evaluar distintos tipos de terapias y reguladores en lazo cerrado. Para comprobar que estas terapias y reguladores funcionan, una vez simulados y evaluados, se tiene que pasar a la experimentación real a través de un protocolo de ensayo clínico real, como paso previo al PEA ambulatorio. Para poder gestionar este protocolo de ensayo clínico real para la verificación de los algoritmos de control, se creó una interfaz de usuario a través de una serie de funciones de simulación y evaluación de terapias con insulina realizadas con MATLABTM (GUI: Graphics User Interface), conocido como Entorno de Páncreas artificial con Interfaz Clínica (EPIC). EPIC ha sido ya utilizada en 10 ensayos clínicos de los que se han ido proponiendo posibles mejoras, ampliaciones y/o cambios. Este proyecto propone una versión mejorada de la interfaz de usuario EPIC propuesta en un proyecto anterior para gestionar un protocolo de ensayo clínico real para la verificación de algoritmos de control en un ambiente hospitalario muy controlado, además de estudiar la viabilidad de conectar el GUI con SimulinkTM (entorno gráfico de Matlab de simulación de sistemas) para su conexión con un nuevo simulador de pacientes aprobado por la JDRF (Juvenil Diabetes Research Foundation). SUMMARY The diabetes mellitus is a metabolic disorder of carbohydrates, as result of an insufficient or null production of insulin in the beta cellules of pancreas, or the manifestation of a reduced sensibility to the insulin from the metabolic system. The type 1 diabetes is characterized for a null production of insulin due to destruction of the beta cellules. Without insulin in the bloodstream, glucose can’t be absorbed by the cellules, producing a hyperglycemia state in the patient and if pass a medium or long time and is not treated can cause severe disease like diabetes syndrome. The type 1 diabetes is an incurable disease but controllable one. The therapy for this disease consists on the exogenous insulin administration with the objective to maintain the glucose level in blood within the normal limits. For the insulin administration, in this project is used an infusion pump, that permit with a subcutaneous glucose sensor, create an autonomous control loop that regulate the optimal insulin amount apply in each moment. When the control algorithm is used in a digital system, with the subcutaneous senor and infusion subcutaneous pump, is named as “Artificial Endocrine Pancreas” for ambulatory use, currently under investigate. These metabolic control algorithms should be evaluates in simulation for assure patients’ physical integrity, for this reason is necessary to design a simulation system that assure the reliability of PAE. This simulation system connects algorithms with metabolic mathematics models for get a previous vision of its performance. In this scenario was created DIABSIMTM, a tool developed in LabView, that later was converted to MATLABTM, and based in the compartmental mathematic model proposed by Hovorka that could simulate and evaluate several different types of therapy and regulators in closed loop. To check the performance of these therapies and regulators, when have been simulated and evaluated, will be necessary to pass to real experimentation through a protocol of real clinical test like previous step to ambulatory PEA. To manage this protocol was created an user interface through the simulation and evaluation functions od therapies with insulin realized with MATLABTM (GUI: Graphics User Interface), known as “Entorno de Páncreas artificial con Interfaz Clínica” (EPIC).EPIC have been used in 10 clinical tests which have been proposed improvements, adds and changes. This project proposes a best version of user interface EPIC proposed in another project for manage a real test clinical protocol for checking control algorithms in a controlled hospital environment and besides studying viability to connect the GUI with SimulinkTM (Matlab graphical environment in systems simulation) for its connection with a new patients simulator approved for the JDRF (Juvenil Diabetes Research Foundation).

Detection and Tracking of Dynamic Objects. A MultiRobot approach to Critical Infrastructures Surveillance.

Despite that Critical Infrastructures (CIs) security and surveillance are a growing concern for many countries and companies, Multi Robot Systems (MRSs) have not been yet broadly used in this type of facilities. This dissertation presents a novel study of the challenges arisen by the implementation of this type of systems and proposes solutions to specific problems. First, a comprehensive analysis of different types of CIs has been carried out, emphasizing the influence of the different characteristics of the facilities in the design of a security and surveillance MRS. One of the most important needs for the surveillance of a CI is the detection of intruders. From a technical point of view this problem can be abstracted as equivalent to the Detection and Tracking of Mobile Objects (DATMO). This dissertation proposes algorithms to solve this specific problem in a CI environment. Using 3D range images of the environment as input data, two detection algorithms for ground robots have been developed. These detection algorithms provide a list of moving objects in the robot detection area. Direct image differentiation and computer vision techniques are used when the robot is static. Alternatively, multi-layer ground reconstructions are compared to detect the dynamic objects when the robot is moving. Since CIs usually spread over large areas, it is very useful to incorporate aerial vehicles in the surveillance MRS. Therefore, a moving object detection algorithm for aerial vehicles has been also developed. This algorithm compares the real optical flow obtained from a down-face oriented camera with an artificial optical flow computed using a RANSAC based homography matrix. Two tracking algorithms have been developed to follow the moving objects trajectories. These algorithms can efficiently handle occlusions and crossings, as well as exchange information among robots. The multirobot tracking can be applied to any type of communication structure: centralized, decentralized or a combination of both. Even more, the developed tracking algorithms are independent of the detection algorithms and could be potentially used with other detection procedures or even with static sensors, such as cameras. In addition, using the 3D point clouds available to the robots, a relative localization algorithm has been developed to improve the position estimation of a given robot with observations from other robots. All the developed algorithms have been extensively tested in different simulated CIs using the Webots robotics simulator. Furthermore, the algorithms have also been validated with real robots operating in real scenarios. In conclusion, this dissertation presents a multirobot approach to Critical Infrastructure Surveillance, mainly focusing on Detecting and Tracking Dynamic Objects.

A vision-based strategy for autonomous aerial refueling tasks

Autonomous aerial refueling is a key enabling technology for both manned and unmanned aircraft where extended flight duration or range are required. The results presented within this paper offer one potential vision-based sensing solution, together with a unique test environment. A hierarchical visual tracking algorithm based on direct methods is proposed and developed for the purposes of tracking a drogue during the capture stage of autonomous aerial refueling, and of estimating its 3D position. Intended to be applied in real time to a video stream from a single monocular camera mounted on the receiver aircraft, the algorithm is shown to be highly robust, and capable of tracking large, rapid drogue motions within the frame of reference. The proposed strategy has been tested using a complex robotic testbed and with actual flight hardware consisting of a full size probe and drogue. Results show that the vision tracking algorithm can detect and track the drogue at real-time frame rates of more than thirty frames per second, obtaining a robust position estimation even with strong motions and multiple occlusions of the drogue.

OMNIWORKS: Omnidirectional vision for human-UAV co-working

The OMNIWORKS project objective is to develop an autonomous and modular aerial inspection system for an off-shore meteorological mast up to 90m in length. The UAV was equipped with an omni-directional camera and vertical take-off/landing capabilities that should be simple enough to operate as to not need the interventions of a professional pilot under challenging situations. Therefore the tests included different aspects used to evaluate both the technical performance of the UAV behavior as well as the operators? point of view.

A vision based aerial rbot solution for the IARC 2014 by the Technical University of Madrid

The IARC competitions aim at making the state of the art in UAV progress. The 2014 challenge deals mainly with GPS/Laser denied navigation, Robot-Robot interaction and Obstacle avoidance in the setting of a ground robot herding problem. We present in this paper a drone which will take part in this competition. The platform and hardware it is composed of and the software we designed are introduced. This software has three main components: the visual information acquisition, the mapping algorithm and the Aritificial Intelligence mission planner. A statement of the safety measures integrated in the drone and of our efforts to ensure field testing in conditions as close as possible to the challenge?s is also included.

Robust Real-time Vision-based Aircraft Tracking From Unmanned Aerial Vehicles.

Aircraft tracking plays a key and important role in the Sense-and-Avoid system of Unmanned Aerial Vehicles (UAVs). This paper presents a novel robust visual tracking algorithm for UAVs in the midair to track an arbitrary aircraft at real-time frame rates, together with a unique evaluation system. This visual algorithm mainly consists of adaptive discriminative visual tracking method, Multiple-Instance (MI) learning approach, Multiple-Classifier (MC) voting mechanism and Multiple-Resolution (MR) representation strategy, that is called Adaptive M3 tracker, i.e. AM3. In this tracker, the importance of test sample has been integrated to improve the tracking stability, accuracy and real-time performances. The experimental results show that this algorithm is more robust, efficient and accurate against the existing state-of-art trackers, overcoming the problems generated by the challenging situations such as obvious appearance change, variant surrounding illumination, partial aircraft occlusion, blur motion, rapid pose variation and onboard mechanical vibration, low computation capacity and delayed information communication between UAVs and Ground Station (GS). To our best knowledge, this is the first work to present this tracker for solving online learning and tracking freewill aircraft/intruder in the UAVs.

Parallel Computer Vision Algorithms for Graphics Processing Units

La evolución de los teléfonos móviles inteligentes, dotados de cámaras digitales, está provocando una creciente demanda de aplicaciones cada vez más complejas que necesitan algoritmos de visión artificial en tiempo real; puesto que el tamaño de las señales de vídeo no hace sino aumentar y en cambio el rendimiento de los procesadores de un solo núcleo se ha estancado, los nuevos algoritmos que se diseñen para visión artificial han de ser paralelos para poder ejecutarse en múltiples procesadores y ser computacionalmente escalables. Una de las clases de procesadores más interesantes en la actualidad se encuentra en las tarjetas gráficas (GPU), que son dispositivos que ofrecen un alto grado de paralelismo, un excelente rendimiento numérico y una creciente versatilidad, lo que los hace interesantes para llevar a cabo computación científica. En esta tesis se exploran dos aplicaciones de visión artificial que revisten una gran complejidad computacional y no pueden ser ejecutadas en tiempo real empleando procesadores tradicionales. En cambio, como se demuestra en esta tesis, la paralelización de las distintas subtareas y su implementación sobre una GPU arrojan los resultados deseados de ejecución con tasas de refresco interactivas. Asimismo, se propone una técnica para la evaluación rápida de funciones de complejidad arbitraria especialmente indicada para su uso en una GPU. En primer lugar se estudia la aplicación de técnicas de síntesis de imágenes virtuales a partir de únicamente dos cámaras lejanas y no paralelas—en contraste con la configuración habitual en TV 3D de cámaras cercanas y paralelas—con información de color y profundidad. Empleando filtros de mediana modificados para la elaboración de un mapa de profundidad virtual y proyecciones inversas, se comprueba que estas técnicas son adecuadas para una libre elección del punto de vista. Además, se demuestra que la codificación de la información de profundidad con respecto a un sistema de referencia global es sumamente perjudicial y debería ser evitada. Por otro lado se propone un sistema de detección de objetos móviles basado en técnicas de estimación de densidad con funciones locales. Este tipo de técnicas es muy adecuada para el modelado de escenas complejas con fondos multimodales, pero ha recibido poco uso debido a su gran complejidad computacional. El sistema propuesto, implementado en tiempo real sobre una GPU, incluye propuestas para la estimación dinámica de los anchos de banda de las funciones locales, actualización selectiva del modelo de fondo, actualización de la posición de las muestras de referencia del modelo de primer plano empleando un filtro de partículas multirregión y selección automática de regiones de interés para reducir el coste computacional. Los resultados, evaluados sobre diversas bases de datos y comparados con otros algoritmos del estado del arte, demuestran la gran versatilidad y calidad de la propuesta. Finalmente se propone un método para la aproximación de funciones arbitrarias empleando funciones continuas lineales a tramos, especialmente indicada para su implementación en una GPU mediante el uso de las unidades de filtraje de texturas, normalmente no utilizadas para cómputo numérico. La propuesta incluye un riguroso análisis matemático del error cometido en la aproximación en función del número de muestras empleadas, así como un método para la obtención de una partición cuasióptima del dominio de la función para minimizar el error. ABSTRACT The evolution of smartphones, all equipped with digital cameras, is driving a growing demand for ever more complex applications that need to rely on real-time computer vision algorithms. However, video signals are only increasing in size, whereas the performance of single-core processors has somewhat stagnated in the past few years. Consequently, new computer vision algorithms will need to be parallel to run on multiple processors and be computationally scalable. One of the most promising classes of processors nowadays can be found in graphics processing units (GPU). These are devices offering a high parallelism degree, excellent numerical performance and increasing versatility, which makes them interesting to run scientific computations. In this thesis, we explore two computer vision applications with a high computational complexity that precludes them from running in real time on traditional uniprocessors. However, we show that by parallelizing subtasks and implementing them on a GPU, both applications attain their goals of running at interactive frame rates. In addition, we propose a technique for fast evaluation of arbitrarily complex functions, specially designed for GPU implementation. First, we explore the application of depth-image–based rendering techniques to the unusual configuration of two convergent, wide baseline cameras, in contrast to the usual configuration used in 3D TV, which are narrow baseline, parallel cameras. By using a backward mapping approach with a depth inpainting scheme based on median filters, we show that these techniques are adequate for free viewpoint video applications. In addition, we show that referring depth information to a global reference system is ill-advised and should be avoided. Then, we propose a background subtraction system based on kernel density estimation techniques. These techniques are very adequate for modelling complex scenes featuring multimodal backgrounds, but have not been so popular due to their huge computational and memory complexity. The proposed system, implemented in real time on a GPU, features novel proposals for dynamic kernel bandwidth estimation for the background model, selective update of the background model, update of the position of reference samples of the foreground model using a multi-region particle filter, and automatic selection of regions of interest to reduce computational cost. The results, evaluated on several databases and compared to other state-of-the-art algorithms, demonstrate the high quality and versatility of our proposal. Finally, we propose a general method for the approximation of arbitrarily complex functions using continuous piecewise linear functions, specially formulated for GPU implementation by leveraging their texture filtering units, normally unused for numerical computation. Our proposal features a rigorous mathematical analysis of the approximation error in function of the number of samples, as well as a method to obtain a suboptimal partition of the domain of the function to minimize approximation error.

3D model based map building

Comunicación presentada en el IX Workshop de Agentes Físicos (WAF'2008), Vigo, 11-12 septiembre 2008.