Analysis, monitoring, and management of quality of experience in video delivery services over IP

Esta tesis estudia la monitorizaci��n y gesti��n de la Calidad de Experiencia (QoE) en los servicios de distribuci��n de v��deo sobre IP. Aborda el problema de c��mo prevenir, detectar, medir y reaccionar a las degradaciones de la QoE desde la perspectiva de un proveedor de servicios: la soluci��n debe ser escalable para una red IP extensa que entregue flujos individuales a miles de usuarios simult��neamente. La soluci��n de monitorizaci��n propuesta se ha denominado QuEM(Qualitative Experience Monitoring, o Monitorizaci��n Cualitativa de la Experiencia). Se basa en la detecci��n de las degradaciones de la calidad de servicio de red (p��rdidas de paquetes, disminuciones abruptas del ancho de banda...) e inferir de cada una una descripci��n cualitativa de su efecto en la Calidad de Experiencia percibida (silencios, defectos en el v��deo...). Este an��lisis se apoya en la informaci��n de transporte y de la capa de abstracci��n de red de los flujos codificados, y permite caracterizar los defectos m��s relevantes que se observan en este tipo de servicios: congelaciones, efecto de ���cuadros���, silencios, p��rdida de calidad del v��deo, retardos e interrupciones en el servicio. Los resultados se han validado mediante pruebas de calidad subjetiva. La metodolog��a usada en esas pruebas se ha desarrollado a su vez para imitar lo m��s posible las condiciones de visualizaci��n de un usuario de este tipo de servicios: los defectos que se eval��an se introducen de forma aleatoria en medio de una secuencia de v��deo continua. Se han propuesto tambi��n algunas aplicaciones basadas en la soluci��n de monitorizaci��n: un sistema de protecci��n desigual frente a errores que ofrece m��s protecci��n a las partes del v��deo m��s sensibles a p��rdidas, una soluci��n para minimizar el impacto de la interrupci��n de la descarga de segmentos de Streaming Adaptativo sobre HTTP, y un sistema de cifrado selectivo que encripta ��nicamente las partes del v��deo m��s sensibles. Tambi��n se ha presentado una soluci��n de cambio r��pido de canal, as�� como el an��lisis de la aplicabilidad de los resultados anteriores a un escenario de v��deo en 3D. ABSTRACT This thesis proposes a comprehensive approach to the monitoring and management of Quality of Experience (QoE) in multimedia delivery services over IP. It addresses the problem of preventing, detecting, measuring, and reacting to QoE degradations, under the constraints of a service provider: the solution must scale for a wide IP network delivering individual media streams to thousands of users. The solution proposed for the monitoring is called QuEM (Qualitative Experience Monitoring). It is based on the detection of degradations in the network Quality of Service (packet losses, bandwidth drops...) and the mapping of each degradation event to a qualitative description of its effect in the perceived Quality of Experience (audio mutes, video artifacts...). This mapping is based on the analysis of the transport and Network Abstraction Layer information of the coded stream, and allows a good characterization of the most relevant defects that exist in this kind of services: screen freezing, macroblocking, audio mutes, video quality drops, delay issues, and service outages. The results have been validated by subjective quality assessment tests. The methodology used for those test has also been designed to mimic as much as possible the conditions of a real user of those services: the impairments to evaluate are introduced randomly in the middle of a continuous video stream. Based on the monitoring solution, several applications have been proposed as well: an unequal error protection system which provides higher protection to the parts of the stream which are more critical for the QoE, a solution which applies the same principles to minimize the impact of incomplete segment downloads in HTTP Adaptive Streaming, and a selective scrambling algorithm which ciphers only the most sensitive parts of the media stream. A fast channel change application is also presented, as well as a discussion about how to apply the previous results and concepts in a 3D video scenario.

Generaci��n de Casos de Prueba en Programaci��n Orientada a Objetos = Test Case Generation in Object-Oriented Programming

Las pruebas de software (Testing) son en la actualidad la t��cnica m��s utilizada para la validaci��n y la evaluaci��n de la calidad de un programa. El testing est�� integrado en todas las metodolog��as pr��cticas de desarrollo de software y juega un papel crucial en el ��xito de cualquier proyecto de software. Desde las unidades de c��digo m��s peque��as a los componentes m��s complejos, su integraci��n en un sistema de software y su despliegue a producci��n, todas las piezas de un producto de software deben ser probadas a fondo antes de que el producto de software pueda ser liberado a un entorno de producci��n. La mayor limitaci��n del testing de software es que contin��a siendo un conjunto de tareas manuales, representando una buena parte del coste total de desarrollo. En este escenario, la automatizaci��n resulta fundamental para aliviar estos altos costes. La generaci��n autom��tica de casos de pruebas (TCG, del ingl��s test case generation) es el proceso de generar autom��ticamente casos de prueba que logren un alto recubrimiento del programa. Entre la gran variedad de enfoques hacia la TCG, esta tesis se centra en un enfoque estructural de caja blanca, y m��s concretamente en una de las t��cnicas m��s utilizadas actualmente, la ejecuci��n simb��lica. En ejecuci��n simb��lica, el programa bajo pruebas es ejecutado con expresiones simb��licas como argumentos de entrada en lugar de valores concretos. Esta tesis se basa en un marco general para la generaci��n autom��tica de casos de prueba dirigido a programas imperativos orientados a objetos (Java, por ejemplo) y basado en programaci��n l��gica con restricciones (CLP, del ingl��s constraint logic programming). En este marco general, el programa imperativo bajo pruebas es primeramente traducido a un programa CLP equivalente, y luego dicho programa CLP es ejecutado simb��licamente utilizando los mecanismos de evaluaci��n est��ndar de CLP, extendidos con operaciones especiales para el tratamiento de estructuras de datos din��micas. Mejorar la escalabilidad y la eficiencia de la ejecuci��n simb��lica constituye un reto muy importante. Es bien sabido que la ejecuci��n simb��lica resulta impracticable debido al gran n��mero de caminos de ejecuci��n que deben ser explorados y a tama��o de las restricciones que se deben manipular. Adem��s, la generaci��n de casos de prueba mediante ejecuci��n simb��lica tiende a producir un n��mero innecesariamente grande de casos de prueba cuando es aplicada a programas de tama��o medio o grande. Las contribuciones de esta tesis pueden ser resumidas como sigue. (1) Se desarrolla un enfoque composicional basado en CLP para la generaci��n de casos de prueba, el cual busca aliviar el problema de la explosi��n de caminos interprocedimiento analizando de forma separada cada componente (p.ej. m��todo) del programa bajo pruebas, almacenando los resultados y reutiliz��ndolos incrementalmente hasta obtener resultados para el programa completo. Tambi��n se ha desarrollado un enfoque composicional basado en especializaci��n de programas (evaluaci��n parcial) para la herramienta de ejecuci��n simb��lica Symbolic PathFinder (SPF). (2) Se propone una metodolog��a para usar informaci��n del consumo de recursos del programa bajo pruebas para guiar la ejecuci��n simb��lica hacia aquellas partes del programa que satisfacen una determinada pol��tica de recursos, evitando la exploraci��n de aquellas partes del programa que violan dicha pol��tica. (3) Se propone una metodolog��a gen��rica para guiar la ejecuci��n simb��lica hacia las partes m��s interesantes del programa, la cual utiliza abstracciones como generadores de trazas para guiar la ejecuci��n de acuerdo a criterios de selecci��n estructurales. (4) Se propone un nuevo resolutor de restricciones, el cual maneja eficientemente restricciones sobre el uso de la memoria din��mica global (heap) durante ejecuci��n simb��lica, el cual mejora considerablemente el rendimiento de la t��cnica est��ndar utilizada para este prop��sito, la \lazy initialization". (5) Todas las t��cnicas propuestas han sido implementadas en el sistema PET (el enfoque composicional ha sido tambi��n implementado en la herramienta SPF). Mediante evaluaci��n experimental se ha confirmado que todas ellas mejoran considerablemente la escalabilidad y eficiencia de la ejecuci��n simb��lica y la generaci��n de casos de prueba. ABSTRACT Testing is nowadays the most used technique to validate software and assess its quality. It is integrated into all practical software development methodologies and plays a crucial role towards the success of any software project. From the smallest units of code to the most complex components and their integration into a software system and later deployment; all pieces of a software product must be tested thoroughly before a software product can be released. The main limitation of software testing is that it remains a mostly manual task, representing a large fraction of the total development cost. In this scenario, test automation is paramount to alleviate such high costs. Test case generation (TCG) is the process of automatically generating test inputs that achieve high coverage of the system under test. Among a wide variety of approaches to TCG, this thesis focuses on structural (white-box) TCG, where one of the most successful enabling techniques is symbolic execution. In symbolic execution, the program under test is executed with its input arguments being symbolic expressions rather than concrete values. This thesis relies on a previously developed constraint-based TCG framework for imperative object-oriented programs (e.g., Java), in which the imperative program under test is first translated into an equivalent constraint logic program, and then such translated program is symbolically executed by relying on standard evaluation mechanisms of Constraint Logic Programming (CLP), extended with special treatment for dynamically allocated data structures. Improving the scalability and efficiency of symbolic execution constitutes a major challenge. It is well known that symbolic execution quickly becomes impractical due to the large number of paths that must be explored and the size of the constraints that must be handled. Moreover, symbolic execution-based TCG tends to produce an unnecessarily large number of test cases when applied to medium or large programs. The contributions of this dissertation can be summarized as follows. (1) A compositional approach to CLP-based TCG is developed which overcomes the inter-procedural path explosion by separately analyzing each component (method) in a program under test, stowing the results as method summaries and incrementally reusing them to obtain whole-program results. A similar compositional strategy that relies on program specialization is also developed for the state-of-the-art symbolic execution tool Symbolic PathFinder (SPF). (2) Resource-driven TCG is proposed as a methodology to use resource consumption information to drive symbolic execution towards those parts of the program under test that comply with a user-provided resource policy, avoiding the exploration of those parts of the program that violate such policy. (3) A generic methodology to guide symbolic execution towards the most interesting parts of a program is proposed, which uses abstractions as oracles to steer symbolic execution through those parts of the program under test that interest the programmer/tester most. (4) A new heap-constraint solver is proposed, which efficiently handles heap-related constraints and aliasing of references during symbolic execution and greatly outperforms the state-of-the-art standard technique known as lazy initialization. (5) All techniques above have been implemented in the PET system (and some of them in the SPF tool). Experimental evaluation has confirmed that they considerably help towards a more scalable and efficient symbolic execution and TCG.

Automatic 3DTV Quality Assessment Based On Depth Perception Analysis

Quality assessment is a key factor for stereoscopic 3D video content as some observers are affected by visual discomfort in the eye when viewing 3D video, especially when combining positive and negative parallax with fast motion. In this paper, we propose techniques to assess objective quality related to motion and depth maps, which facilitate depth perception analysis. Subjective tests were carried out in order to understand the source of the problem. Motion is an important feature affecting 3D experience but also often the cause of visual discomfort. The automatic algorithm developed tries to quantify the impact on viewer experience when common cases of discomfort occur, such as high-motion sequences, scene changes with abrupt parallax changes, or complete absence of stereoscopy, with a goal of preventing the viewer from having a bad stereoscopic experience.

Versatile Bayesian classifier for moving object detection by non-parametric background-foreground modeling

Along the recent years, several moving object detection strategies by non-parametric background-foreground modeling have been proposed. To combine both models and to obtain the probability of a pixel to belong to the foreground, these strategies make use of Bayesian classifiers. However, these classifiers do not allow to take advantage of additional prior information at different pixels. So, we propose a novel and efficient alternative Bayesian classifier that is suitable for this kind of strategies and that allows the use of whatever prior information. Additionally, we present an effective method to dynamically estimate prior probability from the result of a particle filter-based tracking strategy.

Enhanced recommendations for e-learning authoring tools based on a proactive context-aware recommender

Authoring tools are powerful systems in the area of e-Learning that make easier for teachers to create new learning objects by reusing or editing existing educational resources coming from learning repositories or content providers. However, due to the overwhelming number of resources these tools can access, sometimes it is difficult for teachers to find the most suitable resources taking into account their needs in terms of content (e.g. topic) or pedagogical aspects (e.g. target level associated to their students). Recommender systems can take an important role trying to mitigate this problem. In this paper we propose a new model to generate proactive context-aware recommendations on resources during the creation process of a new learning object that a teacher carries out by using an authoring tool. The common use cases covered by the model for having recommendations in online authoring tools and details about the recommender model itself are presented.

Adaptive cross-device videoconferencing solution for wireless networks based on QoS monitoring

The increase in CPU power and screen quality of todays smartphones as well as the availability of high bandwidth wireless networks has enabled high quality mobile videoconfer- encing never seen before. However, adapting to the variety of devices and network conditions that come as a result is still not a trivial issue. In this paper, we present a multiple participant videoconferencing service that adapts to different kind of devices and access networks while providing an stable communication. By combining network quality detection and the use of a multipoint control unit for video mixing and transcoding, desktop, tablet and mobile clients can participate seamlessly. We also describe the cost in terms of bandwidth and CPU usage of this approach in a variety of scenarios.

Aphasic seizures in patients with temporopolar and anterior temporobasal lesions: a video-EEG study

Studies of patients with temporal lobe epilepsy provide few descriptions of seizures that arise in the temporopolar and the anterior temporobasal brain region. Based on connectivity, it might be assumed that the semiology of these seizures is similar to that of medial temporal lobe epilepsy. However, accumulating evidence suggests that the anterior temporobasal cortex may play an important role in the language system, which could account for particular features of seizures arising here. We studied the electroclinical features of seizures in patients with circumscribed temporopolar and temporobasal lesions in order to identify specific features that might differentiate them from seizures that originate in other temporal areas. Among 172 patients with temporal lobe seizures registered in our epilepsy unit in the last 15 years, 15 (8.7%) patients had seizures caused by temporopolar or anterior temporobasal lesions (11 left-sided lesions). The main finding in our study is that patients with left-sided lesions had aphasia during their seizures as the most prominent feature. In addition, while all patients showed normal to high intellectual functioning in standard neuropsychological testing, semantic impairment was found in a subset of 9 patients with left-sided lesions. This case series demonstrates that aphasic seizures without impairment of consciousness can result from small, circumscribed left anterior temporobasal and temporopolar lesions. Thus, the presence of speech manifestation during seizures should prompt detailed assessment of the structural integrity of the basal surface of the temporal lobe in addition to the evaluation of primary language areas.

Multiple Hashing Integration for Real-Time Large Scale Part-to-Part Video Matching

A real-time large scale part-to-part video matching algorithm, based on the cross correlation of the intensity of motion curves, is proposed with a view to originality recognition, video database cleansing, copyright enforcement, video tagging or video result re-ranking. Moreover, it is suggested how the most representative hashes and distance functions - strada, discrete cosine transformation, Marr-Hildreth and radial - should be integrated in order for the matching algorithm to be invariant against blur, compression and rotation distortions: (R; _) 2 [1; 20]_[1; 8], from 512_512 to 32_32pixels2 and from 10 to 180_. The DCT hash is invariant against blur and compression up to 64x64 pixels2. Nevertheless, although its performance against rotation is the best, with a success up to 70%, it should be combined with the Marr-Hildreth distance function. With the latter, the image selected by the DCT hash should be at a distance lower than 1.15 times the Marr-Hildreth minimum distance.

EVA: Endoscopic Video Analysis of the surgical scene for the assessment of MIS psychomotor skills

The present work covers the first validation efforts of the EVA Tracking System for the assessment of minimally invasive surgery (MIS) psychomotor skills. Instrument movements were recorded for 42 surgeons (4 expert, 22 residents, 16 novice medical students) and analyzed for a box trainer peg transfer task. Construct validation was established for 7/9 motion analysis parameters (MAPs). Concurrent validation was determined for 8/9 MAPs against the TrEndo Tracking System. Finally, automatic determination of surgical proficiency based on the MAPs was sought by 3 different approaches to supervised classification (LDA, SVM, ANFIS), with accuracy results of 61.9%, 83.3% and 80.9% respectively. Results not only reflect on the validation of EVA for skills? assessment, but also on the relevance of motion analysis of instruments in the determination of surgical competence.

New generation of human machine interfaces for controlling UAV through depth based gesture recognition

New forms of natural interactions between human operators and UAVs (Unmanned Aerial Vehicle) are demanded by the military industry to achieve a better balance of the UAV control and the burden of the human operator. In this work, a human machine interface (HMI) based on a novel gesture recognition system using depth imagery is proposed for the control of UAVs. Hand gesture recognition based on depth imagery is a promising approach for HMIs because it is more intuitive, natural, and non-intrusive than other alternatives using complex controllers. The proposed system is based on a Support Vector Machine (SVM) classifier that uses spatio-temporal depth descriptors as input features. The designed descriptor is based on a variation of the Local Binary Pattern (LBP) technique to efficiently work with depth video sequences. Other major consideration is the especial hand sign language used for the UAV control. A tradeoff between the use of natural hand signs and the minimization of the inter-sign interference has been established. Promising results have been achieved in a depth based database of hand gestures especially developed for the validation of the proposed system.

Higher-order regularization and morphological techniques for image segmentation

La segmentaci��n de im��genes es un campo importante de la visi��n computacional y una de las ��reas de investigaci��n m��s activas, con aplicaciones en comprensi��n de im��genes, detecci��n de objetos, reconocimiento facial, vigilancia de v��deo o procesamiento de imagen m��dica. La segmentaci��n de im��genes es un problema dif��cil en general, pero especialmente en entornos cient��ficos y biom��dicos, donde las t��cnicas de adquisici��n imagen proporcionan im��genes ruidosas. Adem��s, en muchos de estos casos se necesita una precisi��n casi perfecta. En esta tesis, revisamos y comparamos primero algunas de las t��cnicas ampliamente usadas para la segmentaci��n de im��genes m��dicas. Estas t��cnicas usan clasificadores a nivel de pixel e introducen regularizaci��n sobre pares de p��xeles que es normalmente insuficiente. Estudiamos las dificultades que presentan para capturar la informaci��n de alto nivel sobre los objetos a segmentar. Esta deficiencia da lugar a detecciones err��neas, bordes irregulares, configuraciones con topolog��a err��nea y formas inv��lidas. Para solucionar estos problemas, proponemos un nuevo m��todo de regularizaci��n de alto nivel que aprende informaci��n topol��gica y de forma a partir de los datos de entrenamiento de una forma no param��trica usando potenciales de orden superior. Los potenciales de orden superior se est��n popularizando en visi��n por computador, pero la representaci��n exacta de un potencial de orden superior definido sobre muchas variables es computacionalmente inviable. Usamos una representaci��n compacta de los potenciales basada en un conjunto finito de patrones aprendidos de los datos de entrenamiento que, a su vez, depende de las observaciones. Gracias a esta representaci��n, los potenciales de orden superior pueden ser convertidos a potenciales de orden 2 con algunas variables auxiliares a��adidas. Experimentos con im��genes reales y sint��ticas confirman que nuestro modelo soluciona los errores de aproximaciones m��s d��biles. Incluso con una regularizaci��n de alto nivel, una precisi��n exacta es inalcanzable, y se requeire de edici��n manual de los resultados de la segmentaci��n autom��tica. La edici��n manual es tediosa y pesada, y cualquier herramienta de ayuda es muy apreciada. Estas herramientas necesitan ser precisas, pero tambi��n lo suficientemente r��pidas para ser usadas de forma interactiva. Los contornos activos son una buena soluci��n: son buenos para detecciones precisas de fronteras y, en lugar de buscar una soluci��n global, proporcionan un ajuste fino a resultados que ya exist��an previamente. Sin embargo, requieren una representaci��n impl��cita que les permita trabajar con cambios topol��gicos del contorno, y esto da lugar a ecuaciones en derivadas parciales (EDP) que son costosas de resolver computacionalmente y pueden presentar problemas de estabilidad num��rica. Presentamos una aproximaci��n morfol��gica a la evoluci��n de contornos basada en un nuevo operador morfol��gico de curvatura que es v��lido para superficies de cualquier dimensi��n. Aproximamos la soluci��n num��rica de la EDP de la evoluci��n de contorno mediante la aplicaci��n sucesiva de un conjunto de operadores morfol��gicos aplicados sobre una funci��n de conjuntos de nivel. Estos operadores son muy r��pidos, no sufren de problemas de estabilidad num��rica y no degradan la funci��n de los conjuntos de nivel, de modo que no hay necesidad de reinicializarlo. Adem��s, su implementaci��n es mucho m��s sencilla que la de las EDP, ya que no requieren usar sofisticados algoritmos num��ricos. Desde un punto de vista te��rico, profundizamos en las conexiones entre operadores morfol��gicos y diferenciales, e introducimos nuevos resultados en este ��rea. Validamos nuestra aproximaci��n proporcionando una implementaci��n morfol��gica de los contornos geod��sicos activos, los contornos activos sin bordes, y los turbop��xeles. En los experimentos realizados, las implementaciones morfol��gicas convergen a soluciones equivalentes a aqu��llas logradas mediante soluciones num��ricas tradicionales, pero con ganancias significativas en simplicidad, velocidad y estabilidad. ABSTRACT Image segmentation is an important field in computer vision and one of its most active research areas, with applications in image understanding, object detection, face recognition, video surveillance or medical image processing. Image segmentation is a challenging problem in general, but especially in the biological and medical image fields, where the imaging techniques usually produce cluttered and noisy images and near-perfect accuracy is required in many cases. In this thesis we first review and compare some standard techniques widely used for medical image segmentation. These techniques use pixel-wise classifiers and introduce weak pairwise regularization which is insufficient in many cases. We study their difficulties to capture high-level structural information about the objects to segment. This deficiency leads to many erroneous detections, ragged boundaries, incorrect topological configurations and wrong shapes. To deal with these problems, we propose a new regularization method that learns shape and topological information from training data in a nonparametric way using high-order potentials. High-order potentials are becoming increasingly popular in computer vision. However, the exact representation of a general higher order potential defined over many variables is computationally infeasible. We use a compact representation of the potentials based on a finite set of patterns learned fromtraining data that, in turn, depends on the observations. Thanks to this representation, high-order potentials can be converted into pairwise potentials with some added auxiliary variables and minimized with tree-reweighted message passing (TRW) and belief propagation (BP) techniques. Both synthetic and real experiments confirm that our model fixes the errors of weaker approaches. Even with high-level regularization, perfect accuracy is still unattainable, and human editing of the segmentation results is necessary. The manual edition is tedious and cumbersome, and tools that assist the user are greatly appreciated. These tools need to be precise, but also fast enough to be used in real-time. Active contours are a good solution: they are good for precise boundary detection and, instead of finding a global solution, they provide a fine tuning to previously existing results. However, they require an implicit representation to deal with topological changes of the contour, and this leads to PDEs that are computationally costly to solve and may present numerical stability issues. We present a morphological approach to contour evolution based on a new curvature morphological operator valid for surfaces of any dimension. We approximate the numerical solution of the contour evolution PDE by the successive application of a set of morphological operators defined on a binary level-set. These operators are very fast, do not suffer numerical stability issues, and do not degrade the level set function, so there is no need to reinitialize it. Moreover, their implementation is much easier than their PDE counterpart, since they do not require the use of sophisticated numerical algorithms. From a theoretical point of view, we delve into the connections between differential andmorphological operators, and introduce novel results in this area. We validate the approach providing amorphological implementation of the geodesic active contours, the active contours without borders, and turbopixels. In the experiments conducted, the morphological implementations converge to solutions equivalent to those achieved by traditional numerical solutions, but with significant gains in simplicity, speed, and stability.

A framework for the analysis and optimization of delay in multiview video coding schemes

Esta tesis presenta un novedoso marco de referencia para el an��lisis y optimizaci��n del retardo de codificaci��n y descodificaci��n para v��deo multivista. El objetivo de este marco de referencia es proporcionar una metodolog��a sistem��tica para el an��lisis del retardo en codificadores y descodificadores multivista y herramientas ��tiles en el dise��o de codificadores/descodificadores para aplicaciones con requisitos de bajo retardo. El marco de referencia propuesto caracteriza primero los elementos que tienen influencia en el comportamiento del retardo: i) la estructura de predicci��n multivista, ii) el modelo hardware del codificador/descodificador y iii) los tiempos de proceso de cuadro. En segundo lugar, proporciona algoritmos para el c��lculo del retardo de codificaci��n/ descodificaci��n de cualquier estructura arbitraria de predicci��n multivista. El n��cleo de este marco de referencia consiste en una metodolog��a para el an��lisis del retardo de codificaci��n/descodificaci��n multivista que es independiente de la arquitectura hardware del codificador/descodificador, completada con un conjunto de modelos que particularizan este an��lisis del retardo con las caracter��sticas de la arquitectura hardware del codificador/descodificador. Entre estos modelos, aquellos basados en teor��a de grafos adquieren especial relevancia debido a su capacidad de desacoplar la influencia de los diferentes elementos en el comportamiento del retardo en el codificador/ descodificador, mediante una abstracci��n de su capacidad de proceso. Para revelar las posibles aplicaciones de este marco de referencia, esta tesis presenta algunos ejemplos de su utilizaci��n en problemas de dise��o que afectan a codificadores y descodificadores multivista. Este escenario de aplicaci��n cubre los siguientes casos: estrategias para el dise��o de estructuras de predicci��n que tengan en consideraci��n requisitos de retardo adem��s del comportamiento tasa-distorsi��n; dise��o del n��mero de procesadores y an��lisis de los requisitos de velocidad de proceso en codificadores/ descodificadores multivista dado un retardo objetivo; y el an��lisis comparativo del comportamiento del retardo en codificadores multivista con diferentes capacidades de proceso e implementaciones hardware. ABSTRACT This thesis presents a novel framework for the analysis and optimization of the encoding and decoding delay for multiview video. The objective of this framework is to provide a systematic methodology for the analysis of the delay in multiview encoders and decoders and useful tools in the design of multiview encoders/decoders for applications with low delay requirements. The proposed framework characterizes firstly the elements that have an influence in the delay performance: i) the multiview prediction structure ii) the hardware model of the encoder/decoder and iii) frame processing times. Secondly, it provides algorithms for the computation of the encoding/decoding delay of any arbitrary multiview prediction structure. The core of this framework consists in a methodology for the analysis of the multiview encoding/decoding delay that is independent of the hardware architecture of the encoder/decoder, which is completed with a set of models that particularize this delay analysis with the characteristics of the hardware architecture of the encoder/decoder. Among these models, the ones based in graph theory acquire special relevance due to their capacity to detach the influence of the different elements in the delay performance of the encoder/decoder, by means of an abstraction of its processing capacity. To reveal possible applications of this framework, this thesis presents some examples of its utilization in design problems that affect multiview encoders and decoders. This application scenario covers the following cases: strategies for the design of prediction structures that take into consideration delay requirements in addition to the rate-distortion performance; design of number of processors and analysis of processor speed requirements in multiview encoders/decoders given a target delay; and comparative analysis of the encoding delay performance of multiview encoders with different processing capabilities and hardware implementations.

A knowledge based application for industrialization design

After being designed, a product has to be manufactured, which means converting concepts and information into a real, physical object. This requires a big amount of resources and a careful planning. The product manufacturing must be designed too, and that is called Industrialization Design. An accepted methodology for this activity is starting defining simple structures and then progressively increasing the detail degree of the manufacturing solution. The impact of decisions taken at first stages of Industrialization Design is remarkable, and software tools to assist designers are required. In this paper a Knowledge Based Application prototype for the Industrialization Design is presented. The application is implemented within the environment CATIA V5/DELMIA. A case study with a simple Product from aerospace sector illustrates the prototype development.

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.

300 GHz CMOS video detection using broadband and active planar antennas

Using CMOS transistors for terahertz detection is currently a disruptive technology that offers the direct integration of a terahertz detector with video preamplifiers. The detectors are based on the resistive mixer concept and performance mainly depends on the following parameters: type of antenna, electrical parameters (gate to drain capacitor and channel length of the CMOS device) and foundry. Two different 300 GHz detectors are discussed: a single transistor detector with a broadband antenna and a differential pair driven by a resonant patch antenna.