Work is focused on visualization of large data sets, especially on high resolution terrain models. Thus the goal of this work is to design a library which can divide large models to small regular parts for fast render of whole scene using level of detail techniques. Further goal is theoretical analysis how the library for fast and smooth visualization of large scenes in real time can be used.

No description available.

A height field is a set of height distance values sampled at a finite set of sample points in a two-dimensional parameter domain. A height field usually contains a lot of redundant information, much of which can be removed without a substantial degradation of its quality. A common approach to reducing the size of a height field representation is to use a piecewise polygonal surface approximation. This consists of a mesh of polygons that approximates the surfaces of the original data at a desired level of accuracy. Polygonal surface approximation of height fields has numerous applications in the fields of computer graphics and computer vision. Triangular mesh approximations are a popular means of representing three-dimensional surfaces, and multiresolution analysis (MRA) is often used to obtain compact representations of dense input data, as well as to allow surface approximations at varying spatial resolution. Multiresolution approaches, particularly those moving from coarse to fine resolutions, can often improve the computational efficiency of mesh generation as well as can provide easy control of level of details for approximations. This dissertation concerns the use of wavelet-based MRA methods to produce a triangular-mesh surface approximation from a single height field dataset. The goal of this study is to obtain a fast surface approximation for a set of height data, using a small number of approximating elements to satisfy a given error criterion. Typically, surface approximation techniques attempt to balance error of fit, number of approximating elements, and speed of computation. A novel aspect of this approach is the direct evaluation of wavelet coefficients to assess surface shape characteristics within each triangular element at a given scale. Our approach hierarchically subdivides and refines triangles as the resolution level increases. / Ph. D.

Simulation en temps réel de scènes audio-visuelles complexes reste difficile en raison du processus de rendu techniquement indépendant mais perceptivement lié à chaque modalité. En raison de la dépendance cross-modale potentiel de la perception auditive et visuelle, l'optimisation de graphiques et de rendu sonore, tels que le niveau de détail (LOD), doit être considéré de manière combinée, mais pas comme des questions distinctes. Par exemple, dans l'audition et de la vision, les gens ont des limites perceptives sur la qualité de l'observation. Techniques de LOD conduit par la perception pour les graphismes ont été grandement progressé depuis des décennies. Cependant, le concept de LOD est rarement pris en compte dans l'évaluation et le rendu crossmodal. Cette thèse porte sur l'évaluation de la perception crossmodale sur le rendu LOD audiovisuel par des méthodes psychophysiques, sur lequel on peut appliquer une méthode fonctionnelle et générale, à terme, d'optimiser le rendu. La première partie de la thèse est une étude des problématiques. Dans cette partie, nous passons en revue les différentes approches LOD et discutons les issues, en particulier du point de vue au niveau de la perception crossmodale. Nous discutons également les résultats principaux sur le design, le rendu et les applications interactives des scènes audio et graphiques dans le cadre du projet ANR Topophonie dont la thèse a eu lieu. Une étude des méthodes psychophysiques pour l'évaluation de la perception audio-visuelle est également présentée afin de fournir une solide connaissance du design expérimentale. Dans la deuxième partie, nous nous concentrons sur l'étude de la perception des artefacts d'image dans le rendu LOD audio-visuel. Une série d'expériences a été conçue pour étudier comment la modalité audio supplémentaire peut influer sur la détection visuelle des artefacts produits par la méthode LOD d’imposteur. La troisième partie de la thèse est axée sur le X3D étendu que nous avons conçu pour la modélisation de LOD audio-visuel. Dans la dernière partie, nous présentons le design et l'évaluation du système original par le rendu LOD crossmodal. L'évaluation de la perception audio-visuelle sur le système LOD crossmodal a été atteinte grâce à une série d'expériences psychophysiques. Notre contribution principale est que nous offrons une compréhension originale de LOD crossmodal avec de nouvelles observations, et l'explorer par des expériences et des analyses perceptives. Les résultats de notre travail peuvent être, éventuellement, les preuves empiriques et des lignes directrices pour un système de rendu LOD crossmodale conduit par la perception. / Real-time simulation of complex audio-visual scenes remains challenging due to the technically independent but perceptually related rendering process in each modality. Because of the potential crossmodal dependency of auditory and visual perception, the optimization of graphics and sound rendering, such as Level of Details (LOD), should be considered in a combined manner but not as separate issues. For instance, in audition and vision, people have perceptual limits on observation quality. Techniques of perceptually driven LOD for graphics have been greatly advanced for decades. However, the concept of LOD is rarely considered in crossmodal evaluation and rendering. This thesis is concentrated on the crossmodal evaluation of perception on audiovisual LOD rendering by psychophysical methods, based on that one may apply a functional and general method to eventually optimize the rendering. The first part of the thesis is an overview of our research. In this part, we review various LOD approaches and discuss concerned issues, especially from a crossmodal perceptual perspective. We also discuss the main results on the design, rendering and applications of highly detailed interactive audio and graphical scenes of the ANR Topophonie project, in which the thesis took place. A study of psychophysical methods for the evaluation on audio-visual perception is also presented to provide a solid knowledge of experimental design. In the second part, we focus on studying the perception of image artifacts in audio-visual LOD rendering. A series of experiments was designed to investigate how the additional audio modality can impact the visual detection of artifacts produced by impostor-based LOD. The third part of the thesis is focused on the novel extended-X3D that we designed for audio-visual LOD modeling. In the fourth part, we present a design and evaluation of the refined crossmodal LOD system. The evaluation of the audio-visual perception on crossmodal LOD system was achieved through a series of psychophysical experiments. Our main contribution is that we provide a further understanding of crossmodal LOD with some new observations, and explore it through perceptual experiments and analysis. The results of our work can eventually be used as the empirical evidences and guideline for a perceptually driven crossmodal LOD system.

The use of Virtual Reality (VR) is growing in commercial use, one type of VR headset, called mobile stand-alone system has limited resources for computing and memory. Because of this, when developing real-time applications for this type of VR headset, performance needs to be heavily considered. One popular optimization technique is Level-of-Detail (LOD) which is a technique that represents a model at various resolutions. One type of LOD is called continuous LOD which can represent a model at a continuous spectrum of detail, this is not frequently used, however, because of being less intuitive and more difficult to implement than other versions. This project researched a type of continuous LOD used with a new metric called Pixel Quality Per Meter (PQPM). PQPM relies on having a minimum edge length calculated using a model's screen coverage in relation to its size. To answer whether PQPM can be used together with continuous LOD for intuitive, simple, and efficient updating and rendering in VR. The continuous LOD only uses one low-poly mesh which is tessellated with the help of a Vertex Displacement Map (VDM) to the desired quality. This approach is then evaluated using Nvidia FLIP, an image comparison application that emulates the human visual system. The result was an intuitive and easily implementable LOD which is used together with PQPM to decide the optimal quality given the models size and coverage on the screen. The usage of PQPM did not result in optimal quality at all distances, due to smaller segments being present, which could disappear completely at far distances. The continuous LOD combined with the PQPM did also not scale well but worked well at lower qualities. The study showed groundwork for how PQPM could work together with continuous LOD, it provides a more intuitive and easily implementable continuous LOD than previous approaches, however, because of the scalability issues, further work needs to be done to optimize this approach. / Användingen av Virtual Reality (VR) i kommersiellt syfte ökar, en typ av dessa VR headsets kallas mobila fristående system. Mobila fristående system har begränsade resurser när det kommer till att göra beräkningar och minne. På grund av detta måste man noga överväga prestanda. En populär optimeringsmetod kallas Level-of-Detail (LOD) vilket är en metod som kan representera en modell på ett antal olika upplösningar. En typ av LOD kallas kontinuerlig LOD som kan representera en modell med en kontinuerling detaljnivå. Denna används inte frekvent däremot på grund av att den är mindre intuitiv och svårare att implementera och uppdatera detaljnivå. Detta projektet undersökte en typ av kontinuerlig LOD där den används tillsammans med ett nytt metriskt mått som kallas Pixel Quality Per Meter (PQPM). PQPM byggs på att beräkna en minimal kantlängd via hur stor del av skärmen den täcker i relation till dess storlek. Detta undersöks för att besvara ifall PQPM kan användas tillsammans med kontinuerlig LOD för intuitiv, simpel och effektiv uppdatering och rendering i VR. Den kontinuerliga LOD använder sig bara av en low-poly mesh som är tessellerad med hjälp av en Vertex Displacement Map (VDM) till en önskad kvalitet. Detta evalueras sedan genom Nvidia FLIP, en bildjämförelseapplikation som emulerar människosyn. Resultatet var en intuitiv och enkelt implementerbar kontinuerlig LOD som användes tillsammans med PQPM för att bestämma den optimala detaljnivån baserat på modellens storlek och hur mycket av skärmen den täckte. Användingen av PQPM resulterade inte i optimal kvalitet för alla distanser, på grund av mindre segment som kunde försvinna helt på längre distanser. Den kontinuerliga LOD tillsammans med PQPM hade dålig skalning i prestanda men fungerade bra vid lägre kvalitetsnivåer. Studien visa grundarbetet för hur PQPM kan användas tillsammans med kontinuerlig LOD. Detta tillvägagångsättet tillät för ett mer intuitivt och simpelt sätt att implementera än tidigare tillvägagångssätt. Däremot, på grund av skalningsproblemen, fortsatt arbete måste göras för att optimera detta tillvägagångssättet.

<p>To be able to run a flight simulator a large area of terrain needs to be visualized. The simulator must update the screen in real-time to make the simulation work well. One way of managing large terrains is to tile the area into quadratic tiles to be able to work with different detailed representations of the terrain at different distances from the user. The tiles need to match with the adjacent tiles in the edge points to avoid cracks. When every tile can have a number of different levels of detail this means that the different levels of detail need to match too. This was previously done by having the same edge points in all levels of detail, giving an unnicessarily large polygon count in the less detailed levels. the method developed in this report uses different versions of borders with their own level of detail, adapting to different detail levels at the adjacent tile, reducing hte number of polygons.</p>

Parallel, adaptive Level-of-Detail in VR simulations. Parameters for efficient reuse of already calculated reductions are investigated. Choosen reduction algorithms are evaluated. Parameters for communication within the distributed application are determined. An implementation is introduced. / Paralleles, adaptives Level-of-Detail komplexer Objekten in VR-Simulationen. Die Parameter für eine effiziente Wiederverwendung bereits berechneter Reduktionen werden untersucht. Ausgewählte Reduktionsalgorithmen werden evaluiert. Die Parameter für die Kommunikation in der verteilten Applikation werden ermittelt. Eine Implementation wird vorgestellt.

Direct Volume Rendering (DVR) is a technique for creating images directly from a representation of a function defined over a three-dimensional domain. The technique has many application fields, such as scientific visualization and medical imaging. A striking property of the data sets produced within these fields is their ever increasing size and complexity. Despite the advancements of computing resources these data sets seem to grow at even faster rates causing severe bottlenecks in terms of data transfer bandwidths, memory capacity and processing requirements in the rendering pipeline. This thesis focuses on efficient methods for DVR of large data sets. At the core of the work lies a level-of-detail scheme that reduces the amount of data to process and handle, while optimizing the level-of-detail selection so that high visual quality is maintained. A set of techniques for domain knowledge encoding which significantly improves assessment and prediction of visual significance for blocks in a volume are introduced. A complete pipeline for DVR is presented that uses the data reduction achieved by the level-of-detail selection to minimize the data requirements in all stages. This leads to reduction of disk I/O as well as host and graphics memory. The data reduction is also exploited to improve the rendering performance in graphics hardware, employing adaptive sampling both within the volume and within the rendered image. The developed techniques have been applied in particular to medical visualization of large data sets on commodity desktop computers using consumer graphics processors. The specific application of virtual autopsies has received much interest, and several developed data classification schemes and rendering techniques have been motivated by this application. The results are, however, general and applicable in many fields and significant performance and quality improvements over previous techniques are shown. / On the defence date the status of article IX was Accepted.

Cette thèse porte sur la création interactive intuitive de formes 3D, justifié pour les artistes par un besoin d'efficacité et d'expressivité et pour le grand public par la démocratisation d'usages de la modélisation numérique pour les jeux vidéo ou pour les imprimantes 3D. Dans ce cadre, pour trouver des nouvelles formes d'interactions, nous avons développé des modeleurs et des techniques de modélisation : un modeleur de formes libres 3D par métaphore de peinture, un modeleur vectoriel temps réel de paysages ainsi qu'un modeleur de paysages par croquis vu de la première personne. Les contributions scientifiques vont d'un opérateur de mélange implicite qui garanti la localité du mélange, à l'utilisation qu'une formulation bi-harmonique pour la modélisation de terrains permettant une modélisation fine, intuitive et temps réel de terrains; en passant par la modélisation de paysages à partir de croquis composé de silhouettes vus de la première personne. / This thesis focuses on the interactive intuitive creation of 3D Models. Improvment of modeling tools for artists is still an active research mater especialy concerning efficiency and expressiveness. Moreother video games or 3D printers are make the use of modeling tools more common by the general public. In this context, to find new forms of interaction, we have developed modelers and modeling techniques: a 3D free-form modeler which uses a painting metaphor for creation, a vector-based real-time landscapes modeler as well as a sketch-based landscapes modeler that uses first person drawnings. Scientific contributions range from a blending operator for implicit modeling that guarantees the locality of the blend. The use for a bi-harmonic formulation for landscape modeling that allow fine, intuitive and real-time terrain modeling. Finally through landscape modelisation using sketches composed of silhouettes seen from the first person point of view.

To decrease the rendering time of a mesh, Level of Detail can be generated by reducing the number of polygons based on some geometrical error. While this works well for most meshes, it is not suitable for meshes with an associated texture atlas. By iteratively collapsing edges based on an extended version of Quadric Error Metric taking both spatial and texture coordinates into account, textured meshes can also be simplified. Results show that constraining edge collapses in the seams of a mesh give poor geometrical appearance when it is reduced to a few polygons. By allowing seam edge collapses and by using a pull-push algorithm to fill areas located outside the seam borders of the texture atlas, the appearance of the mesh is better preserved.