Teaching and Students

xxx - xxx (from xxx)

xxx.

xxx - xxx (from xxx)

xxx.

xx

xx.

Alba Garín - Improving the Process of Difussion Tensor Image Analysis for the Study of Neurological Disorders (March 23, 2015)

In this work, the DTI analysis process has been improved with the development of a tool called DTIStatistics for the automatic and easy calculation of DTI group statistics. This tool reduces analysis times and possible errors and therefore, contributes to the study of neurological and psychiatric diseases. We have performed a validation of the final version of DTIStatistics with target users, comparing the user time of this tool and the standard pipeline normally used. Target users were not able to finish all the tasks with the standard pipeline as they could not display the analysis results. Nevertheless, using DTIStatistics they were able to complete all the tasks and barely made errors. Target users needed significantly less time to complete the tasks with DTIStatistics, reducing the analysis time in a 96%. Besides, a study in patients over 60 years of age with late-onset unipolar major depression is presented. The results obtained in the group of late-onset depressive patients were compared with a healthy control group and a control group of patients with early-onset late-life depression. These comparisons could lead us to propose targets for better knowledge of unipolar major depression in the elderly. The whole brain analysis confirms that the diferences in fractional anisotropy, mean diffusivity and radial diffusivity between the late-onset depressive patients and early-onset depressive patients controlled for gender are significant. The differences were found mainly in the frontal lobe and it is thought that this lobe is important in pathophysiology of depression. To our knowledge, this is the first study to analyse fractional anisotropy, mean diffusivity, radial diffusivity and axial diffusivity in order to distinguish between late-onset depressive patients, early-onset depressive patients and healthy subjects. (pdf).

Álvaro Bertelsen - Planning and Registration Techniques for Image-Guided Robotic Spinal Surgery (December 14, 2012)

This work presents the implementation of the image processing applications required by a robotic assistant designed for transpedicular fixation, a particular type of spine surgery consisting in the immobilisation of multiple vertebrae by means of screws connected by metal bars. Usually, transpedicular fixation, also known as inter-body fusion, is done to prevent further damage on a vertebra afected by fracture, disc problems or tumours. This surgery demands high levels of precision, as a badly inserted screw can damage the spinal cord, the surrounding blood vessels or the pedicle itself. Misplacement of screws is not uncommon: figures vary between studies but can reach values up to 30%, which can be effectively lowered by the use of Computer Assisted Surgery (CAS) systems. However, further improvement could be obtained by means of a robotic assistant like the one described here, which would cooperate with the surgeon steadily holding the instrumental on precise locations. In this way, the surgeon would always remain in charge of the intervention, while the robot performs tasks which are either repetitive or demand too much precision. (pdf).

Hugo Álvarez - Study of Augmented Reality methods for Real Time Recognition and Tracking of Untextured 3D Models in Monocular Images (December 22, 2011)

This thesis presents a study of the existing optical tracking methods and provides some improvements for some of them, particularly for those that are real time. More precisely, monocular optical marker tracking and model based monocular optical markerless tracking are discussed in detail. The proposed improvements are focused on industrial environments, which is a difficult challenge due to the lack of texture in these scenes. The response of model based monocular optical markerless tracking methods is jeopardized for untextured scenes, so this thesis proposes a 3D object recognition method that uses geometric properties instead of texture to initialize the tracking, as well as a markerless tracking method that uses multiple visual cues to update the tracking. Additionally, the details of the augmented reality system that has been developed to help in disassembly operations are given along this document. This serves as a tool to validate the proposed methods and shows a real world applicability of them. (pdf).

Carlos Buchart - Study of Parallel Techniques Applied to Surface Reconstruction from Unorganized and Unoriented Point Clouds (December 13, 2010)

This thesis studies the application of parallel techniques in the surface reconstruction problem, in order to improve the processing time required to obtain the final mesh. It is shown how local interpolating triangulations are suitable for global reconstruction, at the time that it is possible to take advantage of the independent nature of these triangulations to design highly eficient parallel methods. A parallel surface reconstruction method is presented, based on local Delaunay triangulations. The input points do not present any additional information, such as normals, nor any known structure. This method has been designed to be GPU friendly, and two implementations are presented. To deal the inherent problems related to interpolating techniques (such as noise, outliers and non-uniform distribution of points), a consolidation process is studied and a parallel points projection operator is presented, as well as its implementation in the GPU. This operator is combined with the local triangulation method to obtain a better reconstruction.This work also studies the possibility of using dynamic reconstruction techniques in a parallel fashion. The method proposed looks for a better interpretation and recovery of the shape and topology of the target model. (pdf).

Jairo Sánchez - A Stochastic Parallel Method for Real Time Monocular SLAM applied to Augmented Reality (December 10, 2010)

Simultaneous Localization and Mapping (SLAM) techniques can deal with completely unknown scenes, simultaneously estimating the camera pose and the 3D structure. Traditionally, this problem is solved using nonlinear minimization techniques that are very accurate but hardly used in real time. In this way, this thesis presents a highly parallelizable random sampling approach based on Monte Carlo simulations that fits very well on the graphics hardware. As demonstrated in the text, the proposed algorithm achieves the same precision as nonlinear optimization, getting real time performance running on commodity graphics hardware. (pdf).

 Josune Hernantes - Interaction in Multimodal Haptic Systems: Collision Response and Usability Improvements (December 19, 2008)

Haptic devices allow users to interact with virtual objects through the sense of touch receiving force feedback as a response to collisions with the environment. However, haptic technology is still recent and as a result it is difficult to get a stable and realistic response in complex environments. This thesis investigates and provides solutions in two haptic key areas: the calculation of contact forces and torques as a result of user actions in virtual environments and multisensory perception. Real-time collision detection and response is a complicated and essential task for the effectiveness of these systems. This work develops a haptic rendering method able to compute an improved and stable haptic response even in complex situations such as disassembly operations in aeronautic environments. Furthermore, it analyses the effectiveness of combining different sensory modalities available on the system to improve the overall perception of the system as well as user immersion. (pdf).

Iker Aguinaga - Automatic Analysis of the Precedence Relationships and Disassembly Routes for Selective Disassembly Planning in Mechanical Systems Using Virtual Mock-ups (May 04, 2007)

This thesis presents a method for the automatic generation of selective disassembly sequences, the sequence of part and subassembly removals required to extract a single target part. This problem requires finding the precedence relationships among the removals of the different parts. This thesis presents a method to obtain automatically these precedence relationships from the geometric definition of the model defined as triangle meshes. Due to the nature of this type of input the method needs to deal with a non-exact geometric definition of the model. The planning of disassembly sequences also requires the identification of the extraction trajectories of the different parts or assemblies. The failure to find these trajectories can make a planner fail to generate correct sequences or not evaluate potential solutions. This thesis analyses the disassembly path planning problem, its relation with the general path planning problem and the main differences between both of them. We propose several strategies of path planning to solve this problem and we analyse its behaviour and performance.

Pablo Delgado - xxx (September xx, 2017)

xxx.

Francisco Recio - Desarrollo en Android para aplicaciones de Realidad Aumentada (July 17, 2012)

This project aims to make applications of augmented reality on Android. To develop these applications have been reused existing libraries at CEIT implemented in C++. A study of alternatives for their inclusion in Android has been performed as well.

Javier Lebrero - Restaurant Menu Application for Android based on a Client Server Model (July 9, 2012)

This project aims to develop an application which is an easy manner for the restaurants to serve their customers with a better and more efficient service. The application consists specifically on an electronic menu card for restaurants. This program allows the customers to order the meals through a tablet. So they can see all the info in the interface, choose their food, and then check the state of their meals. The waiter will just give the device to the customer, who will choose and check, or even customize the menu in case he doesn’t want to eat one or more types of food.

Marta Santamaría - Advanced GUI's development for Oncology (April 25, 2012)

The project is developing a software that will allow on one hand to accurately calculate the dose of radiation energy absorbed and on the other hand, must provide tools for data analysis, i.e., develop user interface tools such as: display data in graphs, compare different graphs to evaluate their correlation, calculate statistics of those graphs, etc.

Jesús Iglesias - A Haptic System for an Allergy Application (June 7, 2011)

The present project approaches the design and development of an application to capture and represent the allergy tests using a haptic device and analyzing these tests. The new application called “Alergi”, obtains the data collection with a haptic, an electronic device similar to a pen, capable of capturing 3D positions. Thus, test results can be measured in situ and sent in digital format to a doctor with great time and work savings. As the information is represented on screen in the same form as the brands in the skin, this allows measuring captured samples and store them in a document.

Miguel Ángel Ortiz - InTrainer: A Virtual Reality System for Cardiac Rehabilitation (September 8, 2010)

InTrainer is a whole solution, integrated by hardware and software elements, designed to be used in the heart recovery section of a hospital, by patients attending programs after some cardiovascular event. It allows patients to immerse in a virtual world, where they can perform their training, while a biomedical signal reading system takes note about their different data. The virtual experience is rich and interesting, because of the different state-of-the-art tools used to build it. This experience is aimed to incentivize the team work, and also, to promote support between different members of the patient group. The information read from each sensor is constantly updated in the GUI and modify the behaviour of the avatar that represent a patient. In the same way, the log system keeps track of every important data trace from each patient, and stores it in a file for further usage.

Isabel Aguiriano - Minimum Requirements for Virtual Reality Helicopter Cockpit Applications (July 28, 2010)

This thesis shows possible field of applications for helicopter cockpit development using Virtual Reality technology and specifying minimum requirements. The human senses perceiving cockpit and outside environment have been also taken into account. The purpose is to precise criteria for evaluating the visualization in virtual reality from the point of view of an engineer, a pilot, and presenting virtual products to a customer. Thus, minimum requirements for each application have been identified. In the case that the requirements can not be full filled, some negative side effects are explained regarding recent studies that took place. Further today’s available VR hardware and software have been identified and verified regarding the minimum requirements. This study gives the opportunity to be witness of the numerous advantages of using Virtual Reality. However, also has been checked the numerous factors take place in the design of the virtual reality simulation. These could be and issue for further work.

Xabier Sistiaga - Illumination estimation from a picture (July 6, 2010)

In augmented reality, there is the need to fusionate computer rendered objects in real world scenes, in photographs or in videos. It is getting more extended, for example, in urban construction where there is an urge to know the kind of impact will a new building have. Consecuently, if the result needs to be fine, the 3D object needs to be rendered within some especifics conditions. Whilst, the most important condition is none other than the impact that the local illumination has to the object. Thus, knowing the scene illumination, number of lights, ubication from the object and their relative intensities, is crucial. This project studies the different posibilities for light source detecting models and implements the most convinient one for each cause. 3D object needs to be rendered within some especifics conditions. Whilst, the most important condition is none other than the impact that the local illumination has to the object. Thus, knowing the scene illumination, number of lights, ubication from the object and their relative intensities, is crucial. This project studies the different posibilities for light source detecting models and implements the most convinient one for each cause.

Pedro José Areizaga - Development of software modules for comercial aviation simulators (February 18, 2010)

The goal is to develop software plugins for X-Plane simulator (or modigy the source code) in order to improve or add more functionalities like for example a spherical projection of the three views to a Elumens Vision Station.

Denis Sánchez - Automatic cluster set-up and API for distributing computing (December 14, 2009)

This project is divided in two parts. In the first part, the aim was to create an installer for a Linux system to make the configuration of a cluster in a few clicks. Besides, the type of cluster created, is disk-less having the advantage of being a dynamical and not having to install a complete operating system in all the nodes. Regarding the second part of the project, is to develop an API which encapsulates MPI and makes it easier to develop parallel programs for the cluster. The future idea will be to use the cluster for complex computations in fields like image processing, simulation, etc. (more info).

Francisco Ostiategui - Analysis of OpenCL library for parallel applications (November 13, 2009)

The goal is the study of OpenCL capacities for parallel programming. Several applications (image field, mass-spring field) will be analysed comparing the perfomance using different strategies: CPU, GPU, OpenCL and clustering.

Imanol Herrera - Modular API for a graphics kernel of volumetric visualization (September 16, 2009)

Volumetric visualization allows the visualization of large data sets as medical images or experiments results. Hence it is commonly used for medical and scientific visualization. This project aims to create a general volumetric visualization library. For that purpose, an existing volumetric engine has been extracted from a former program. It has also been generalized and improved by adding tools for creating and editing the transfer styles used for the visualization.

Ibon Eskudero - Particle Filter in GPU for 3D Tracking (April 29, 2009)

Te goal of this project is to develop 3D Tracker based on CAD model matching and using GPGPU. The matching will use a particle filter algorithm.

Alfonso Callejo - Deformable Object Simulation using Mass-Spring Models: Numerical Methods and Stiffness Determination (September 26, 2008)

This project makes an overview of models used in real-time simulation of deformable objects, specifically soft tissue simulation for surgery simulators, and analyses in depth the most popular model among them: the mass-spring model (MSM), describing the different numerical methods used to solve it and making a proposal for the determination of its constitutive parameters. The discussed theoretical parameters and methods have been implemented in a simple program, in order to compare the methods and check them in practice. Finally, the experiments carried out at the end of the project, which compare MSM and nonlinear finite element method (FEM) solutions, corroborate that the MSM, using the parameters adjusted with the developed method, can be physically-realistic, allowing to calculate the exact shape of a deformed hyperelastic solid under certain loads.

Imanol Egido - Study and Implementation of Markerless Camera Auto-Calibration Methods (September 10, 2008)

In Augmented Reality, it is needed a camera that captures reality and a visualizer that shows those real elements next to the computer generated images. The aim of this project is to generate an stratified calibration method to obtain the intrinsic parameters of the camera.

Hugo Álvarez - Improvement of Collision Detection and Response for Haptic Devices (October 10, 2007)

This work shows the steps to encapsulate and interface a collision detection module. Furthermore, the new interface and structures offer to the users a simply API. The main idea of this design is to lower the effort required for the integration of the module in new projects. On the other hand, the process to integrate a new collision response oriented to haptic systems is also showed. This part describes the implemented optimizations that reduce the time of response and improve the calculations of the resulting force. The objective is to improve realism and to achieve optimal haptic frequency (1 kHz).

Javier Heras - Cooperative Augmented Reality System for Mobile Devices (January 26, 2007)

The aim of the Augmented Reality is to add computer generated data to real images. This technology is very interesting in very different areas like industrial processes, environmental studies, surgery or entertainment. In this work we pretend to apply this technology to a mobile device adding the functionality of sending those images to another device which we’ll call “client device”.

Jairo Sánchez - Study and Development of Techniques for the Markeless Camera Pose (June 2006)

Computation of the camera pose for synthetic and real scenes. The camera is pre-calibrated and the method doesn't need any kind of marker nor sensor. For the 3D tracker, several known algorithms have been used and compared among them.

Borja Morales - Simulation of Augmented Reality using the 3D Model of the Mechanical Department at Ceit (April 2006)

The first part of this work was the 3D modelling of our mechanical department at Ceit. After the modelling and the texturing, we had a very good platform for checking augmented reality algorithms. Before using the ARToolkit library in a real scene, we can check the library in a virtual environment. Thanks to this, it's possible to change materials, lights, etc. in order to study their influence in the algorithm.

Jokin Santurde - Gesture Recognition for controlling applications (September 07, 2005)

Gesture language is a powerful communication tool that complements the verbal communication and it can be used to interact with computers as people do in real life for talking each other. To integrate these gesture interfaces in an ambient intelligence environment it is necessary to recognise the gesture in a non invasive way. This research is focused on the study about markerless hand movement capture systems oriented to gesture interfaces. Thanks to this study we have developed a skin colour based hand detection application using new strategies like assimilative background that combined with HSV, they result in a more robust skin detector.

Álvaro Espinosa - Study and Comparison of different Collision Detection Methods (March 2005)

Comparison between our collision detection method (based on voxels) and the public library V-Collide.

Juan María Terrés - Study and comparison of alternatives for dimensional control in industry (May 15, 2017)

In the present project, we have experimented with three different camera models and four different optics (three of them monocentric as well as a telecentric one), in order to find the best set focusing on dimensional metrology applications. The best set will be the one which works with the minimum error, subject to proper performance.

Pablo Amadoz - Mechanical design of  a modular belt convayor (September 12, 2016)

This project focuses on the study of light and small pieces by means of a dimensional analysis of them (including the belt). We focus on the design of a structure where the belt can be supplemented with other elements such as sensors and cameras.

Pablo Delgado - Traffic (velocity) signals detection (June 4, 2015)

This work presents a fog detection and analysis system based on the different characteristics of the HSV image histograms and image classification program according to the concentration of fog in them, consisting of an analysis of the HSV characteristics and the image grayscale.

Carlos Álvarez - ADAS fog detection system based on computer vision (July 27, 2015)

This work presents a fog detection and analysis system based on the different characteristics of the HSV image histograms and image classification program according to the concentration of fog in them, consisting of an analysis of the HSV characteristics and the image grayscale.

Carlos Navarro - Research and development of low complexity multiuser environments for mobile devices (September 19, 2014)

This project seeks to explore the technical challenges of low complexity multiuser environments, that is, real time interaction between players, usually through an avatar.

Miguel Contreras - Desarrollo de una aplicación para dispositivos móviles IOS para el acceso a Consultas Académicas de la Universidad de Navarra (September 19, 2014)

El presente trabajo detalla el desarrollo de una aplicación destinada a dispositivos móviles con el sistema operativo iOS mediante la cual los alumnos de la Universidad de Navarra puedan realizar de una manera más amena, rápida y sencilla distintas consultas académicas.

Mikel Joaristi - Sistema vSLAM monocular optimizado para aplicaciones de Realidad Aumentada (July 16, 2014)

En este trabajo se presenta el estudio y desarrollo de un sistema de Visual Simultaneous Localization and Mapping (vSLAM) monocular y de bajo coste para su uso en aplicaciones de Realidad Aumentada en tiempo real. La característica principal del sistema desarrollado es que está basado en el método de optimización conocido como Bundle Ajustment.

Clara Molinuevo - Design of a procedure for the analysis of brain areas during cognitive tasks in schizophrenia patients using FSL (June 30, 2014)

This  project aims  to study  the  physiological abnormalities  that occur in  schizophrenic patients’  brain functioning, compared to healthy subjects’.  Clinica Universidad de Navarra (CUN) started a study on  task-related Functional Magnetic Resonance Imaging (fMRI)  and created an experiment for the subjects of the study to undergo. This work aims to develop an analysis procedure for the fMRI images obtained from the subjects. Due to lack of time, only three healthy subjects underwent the experiment. For testing the procedure developed, the data obtained from those 3 subjects were analyzed, and the results validated the design. The results of the study showed a greater activation in frontal areas of the  subjects’ brains, sometimes related to movement planning or thinking. The procedure was prepared to be used in a larger number of samples and in comparison of schizophrenic patients and healthy subjects.

David Seca - Reconocimiento y tracking de objetos 3D en dispositivos móviles (September 27, 2013)

Este trabajo presenta una aplicación de realidad aumentada orientada a dispositivos móviles que permite el reconocimiento y seguimiento (tracking) de objetos 3D a partir de nubes de puntos obtenidas en tiempo real por el sensor de profundidad Microsoft Kinect. La aplicación se ha desarrollado utilizando la librería Point Cloud; y el objetivo ha consistido en estudiar la viabilidad de utilizar en dispositivos móviles los algoritmos de reconocimiento y tracking de la librería Point Cloud. En este trabajo se han adaptado los algoritmos de la librería Point Cloud de reconocimiento y estimación de posición (con seis grados de libertad) mediante un nuevo enfoque híbrido reduciendo su coste computacional y acercándose a tiempo real. Además, se han modificado los parámetros de tracking mejorando su funcionamiento.

Unai Gurrutxaga - Estudio de la plataforma IOS para aplicaciones gráficas (July 12, 2013)

El objetivo de este proyecto es hacer un estudio de la plataforma iOS a la hora de desarrollar aplicaciones de realidad virtual para conocer las posibilidades de la misma. Para ello, se han identificado los siguientes objetivos específicos: aprender los fundamentos sobre la plataforma, estudiar el API gráfico, y desarrollar ejemplos que muestren las principales funcionalidades de una aplicación de realidad virtual en un dispositivo IOS.

Jon Roncal - Reconocimiento de expresiones faciales mediante técnicas de procesamiento de imágenes (May 30, 2013)

El presente proyecto describe el desarrollo de una aplicación software que dada una imagen que contiene una cara, capturada con una cámara web  convencional, es capaz de detectar la posición de la cara y reconocerla. Adicionalmente, en el proceso de reconocimiento la aplicación estima la posición de sus características faciales más relevantes (ojos, nariz, boca) e identifica la identidad del individuo, así como su expresión (feliz, sonriente, triste, etc.), género o edad aproximada.