Malhas simpliciais são utilizadas em várias áreas da Computação Gráfica e Engenharia, como por exemplo, em vizualização, simulação, prototipação, além de outras aplicações. Este tipo de malha é, geralmente, utilizada como aproximações discretas de espaços contínuos, onde eles oferecem representações flexíveis e eficientes. Muito esforço é gasto visando gerar malhas de boa qualidade, porém, em alguns casos as malhas acabam sendo modificadas. Entretanto, este tipo de operação é geralmente custosa e inflexível, o que pode resultar na geraão de malhas bem diferentes das originais. A habilidade de manipular cenas dinâmicas revela-se um dos problemas mais desafiadores da computação gráfica. Este trabalho propõe um método alternativo para atualizar malhas simpliciais que vai além de mudanças geométricas e topológicas. Tal método explora uma das propriedade das Tringulações de Delaunay com Pesos, que permite a usá-las para definir implicitamente as relações de conectividade de uma malha. Ao contrário de manter as informações de conectividade explicitamente, a atual abordagem simplesmente armazena uma coleção de pesos associados a cada vértice. Além disso, criamos um algoritmo para calcular uma Tringulação de Delaunay com Pesos a partir de uma dada triangulação. O algoritmo consiste em uma busca em largura que atribui pesos aos vértices, e uma estratégia de de subdivisão para assegurar que a triangulação reconstruída será correspondente à original. Este método apresenta diversas aplicações e, em particular, permite a criação de um sistema simples de realizar combinação entre triangulações, que será ilustrada com exemplos em 2D e 3D. / Simplicial meshes are used in many fields of Computer Graphics and Engineering, for instance, in visualization, simulation, prototyping, among other applications. This kind of mesh is often used as discrete approximations of continuous spaces, where they offer flexible and efficient representations. Considerable effort is spent in generating good quality meshes, but in some applications the meshes can be modified over time. However, this kind of operation is often very expensive and inflexible, sometimes leading to results very different from the original meshes. The ability to handle dynamic scenes reveals itself as one of the most challenging problems in computer graphics. This work proposes an alternative technique for updating simplicial meshes that undergo geometric and topological changes. It explores the property that a Weighted Delaunay Triangulation (WDT) can be used to implicitly define the connectivity of a mesh. Instead of explicitly maintaining connectivity information, this approach simply keeps a collection of weights associated to each vertex. It consists of an algorithm to compute a WDT from any given triangulation, which relies on a breadth-first traversal to assign weights to vertices, and a subdivision strategy to ensure that the reconstructed triangulation conforms with the original one. This technique has many applications and, in particular, it allows for a very simple method of merging triangulations, which is illustrated with both 2D and 3d examples.

Acquisition of 3D model of real objects and scenes is indispensable and useful in many practical applications, such as digital archives, game and entertainment industries, engineering, advertisement. There are 2 main methods for 3D acquisition : laser-based reconstruction (active method) and image-based reconstruction from multiple images of the scene in different points of view (passive method). While laser-based reconstruction achieves high accuracy, it is complex, expensive and difficult to set up for large-scale outdoor reconstruction. Image-based, or multi-view stereo methods are more versatile, easier, faster and cheaper. By the time we begin this thesis, most multi-view methods could handle only low resolution images under controlled environment. This thesis targets multi-view stereo both both in large scale and high accuracy issues. We significantly improve some previous methods and combine them into a remarkably effective multi-view pipeline with GPU acceleration. From high-resolution images, we produce highly complete and accurate meshes that achieve best scores in many international recognized benchmarks. Aiming even larger scale, on one hand, we develop Divide and Conquer approaches in order to reconstruct many small parts of a big scene. On the other hand, to combine separate partial results, we create a new merging method, which can merge automatically and quickly hundreds of meshes. With all these components, we are successful to reconstruct highly accurate water-tight meshes for cities and historical monuments from large collections of high-resolution images (around 1600 images of 5 M Pixel images)

Acquisition of 3D model of real objects and scenes is indispensable and useful in many practical applications, such as digital archives, game and entertainment industries, engineering, advertisement. There are 2 main methods for 3D acquisition : laser-based reconstruction (active method) and image-based reconstruction from multiple images of the scene in different points of view (passive method). While laser-based reconstruction achieves high accuracy, it is complex, expensive and difficult to set up for large-scale outdoor reconstruction. Image-based, or multi-view stereo methods are more versatile, easier, faster and cheaper. By the time we begin this thesis, most multi-view methods could handle only low resolution images under controlled environment. This thesis targets multi-view stereo both both in large scale and high accuracy issues. We significantly improve some previous methods and combine them into a remarkably effective multi-view pipeline with GPU acceleration. From high-resolution images, we produce highly complete and accurate meshes that achieve best scores in many international recognized benchmarks. Aiming even larger scale, on one hand, we develop Divide and Conquer approaches in order to reconstruct many small parts of a big scene. On the other hand, to combine separate partial results, we create a new merging method, which can merge automatically and quickly hundreds of meshes. With all these components, we are successful to reconstruct highly accurate water-tight meshes for cities and historical monuments from large collections of high-resolution images (around 1600 images of 5 M Pixel images)

Malhas simpliciais são utilizadas em várias áreas da Computação Gráfica e Engenharia, como por exemplo, em vizualização, simulação, prototipação, além de outras aplicações. Este tipo de malha é, geralmente, utilizada como aproximações discretas de espaços contínuos, onde eles oferecem representações flexíveis e eficientes. Muito esforço é gasto visando gerar malhas de boa qualidade, porém, em alguns casos as malhas acabam sendo modificadas. Entretanto, este tipo de operação é geralmente custosa e inflexível, o que pode resultar na geraão de malhas bem diferentes das originais. A habilidade de manipular cenas dinâmicas revela-se um dos problemas mais desafiadores da computação gráfica. Este trabalho propõe um método alternativo para atualizar malhas simpliciais que vai além de mudanças geométricas e topológicas. Tal método explora uma das propriedade das Tringulações de Delaunay com Pesos, que permite a usá-las para definir implicitamente as relações de conectividade de uma malha. Ao contrário de manter as informações de conectividade explicitamente, a atual abordagem simplesmente armazena uma coleção de pesos associados a cada vértice. Além disso, criamos um algoritmo para calcular uma Tringulação de Delaunay com Pesos a partir de uma dada triangulação. O algoritmo consiste em uma busca em largura que atribui pesos aos vértices, e uma estratégia de de subdivisão para assegurar que a triangulação reconstruída será correspondente à original. Este método apresenta diversas aplicações e, em particular, permite a criação de um sistema simples de realizar combinação entre triangulações, que será ilustrada com exemplos em 2D e 3D. / Simplicial meshes are used in many fields of Computer Graphics and Engineering, for instance, in visualization, simulation, prototyping, among other applications. This kind of mesh is often used as discrete approximations of continuous spaces, where they offer flexible and efficient representations. Considerable effort is spent in generating good quality meshes, but in some applications the meshes can be modified over time. However, this kind of operation is often very expensive and inflexible, sometimes leading to results very different from the original meshes. The ability to handle dynamic scenes reveals itself as one of the most challenging problems in computer graphics. This work proposes an alternative technique for updating simplicial meshes that undergo geometric and topological changes. It explores the property that a Weighted Delaunay Triangulation (WDT) can be used to implicitly define the connectivity of a mesh. Instead of explicitly maintaining connectivity information, this approach simply keeps a collection of weights associated to each vertex. It consists of an algorithm to compute a WDT from any given triangulation, which relies on a breadth-first traversal to assign weights to vertices, and a subdivision strategy to ensure that the reconstructed triangulation conforms with the original one. This technique has many applications and, in particular, it allows for a very simple method of merging triangulations, which is illustrated with both 2D and 3d examples.

Malhas simpliciais são utilizadas em várias áreas da Computação Gráfica e Engenharia, como por exemplo, em vizualização, simulação, prototipação, além de outras aplicações. Este tipo de malha é, geralmente, utilizada como aproximações discretas de espaços contínuos, onde eles oferecem representações flexíveis e eficientes. Muito esforço é gasto visando gerar malhas de boa qualidade, porém, em alguns casos as malhas acabam sendo modificadas. Entretanto, este tipo de operação é geralmente custosa e inflexível, o que pode resultar na geraão de malhas bem diferentes das originais. A habilidade de manipular cenas dinâmicas revela-se um dos problemas mais desafiadores da computação gráfica. Este trabalho propõe um método alternativo para atualizar malhas simpliciais que vai além de mudanças geométricas e topológicas. Tal método explora uma das propriedade das Tringulações de Delaunay com Pesos, que permite a usá-las para definir implicitamente as relações de conectividade de uma malha. Ao contrário de manter as informações de conectividade explicitamente, a atual abordagem simplesmente armazena uma coleção de pesos associados a cada vértice. Além disso, criamos um algoritmo para calcular uma Tringulação de Delaunay com Pesos a partir de uma dada triangulação. O algoritmo consiste em uma busca em largura que atribui pesos aos vértices, e uma estratégia de de subdivisão para assegurar que a triangulação reconstruída será correspondente à original. Este método apresenta diversas aplicações e, em particular, permite a criação de um sistema simples de realizar combinação entre triangulações, que será ilustrada com exemplos em 2D e 3D. / Simplicial meshes are used in many fields of Computer Graphics and Engineering, for instance, in visualization, simulation, prototyping, among other applications. This kind of mesh is often used as discrete approximations of continuous spaces, where they offer flexible and efficient representations. Considerable effort is spent in generating good quality meshes, but in some applications the meshes can be modified over time. However, this kind of operation is often very expensive and inflexible, sometimes leading to results very different from the original meshes. The ability to handle dynamic scenes reveals itself as one of the most challenging problems in computer graphics. This work proposes an alternative technique for updating simplicial meshes that undergo geometric and topological changes. It explores the property that a Weighted Delaunay Triangulation (WDT) can be used to implicitly define the connectivity of a mesh. Instead of explicitly maintaining connectivity information, this approach simply keeps a collection of weights associated to each vertex. It consists of an algorithm to compute a WDT from any given triangulation, which relies on a breadth-first traversal to assign weights to vertices, and a subdivision strategy to ensure that the reconstructed triangulation conforms with the original one. This technique has many applications and, in particular, it allows for a very simple method of merging triangulations, which is illustrated with both 2D and 3d examples.

L'acquisition de modèles 3D des scènes réelles trouve son utilité dans de nombreuses applications pratiques, comme l'archivage numérique, les jeux vidéo, l'ingénierie, la publicité. Il existe principalement deux méthodes pour acquérir un modèle 3D: la reconstruction avec un scanner laser (méthode active) et la reconstruction à partir de plusieurs photographies d'une même scène prise dans des points de vues différentes (méthode passive). Si la méthode active permet d'acquérir des modèles avec une grande précision, il est cependant coûteux et difficile à mettre en place pour de grandes scènes extérieures. La méthode passive, ou la stéréo multi-vues est en revanche plus flexible, facile à mettre en oeuvre et surtout moins coûteuse que la méthode active. Cette thèse s'attaque au problème de la reconstruction de stéréo multi-vues à grande échelle et précise pour les scènes extérieures. Nous améliorons des méthodes précédentes et les assemblons pour créer une chaîne de stéréo multi-vues efficace tirant parti de l'accélération de cartes graphiques. La chaîne produit des maillages de qualité à partir d'images de haute résolution, ce qui permet d'atteindre les meilleurs scores dans de nombreuses évaluations. Aux plus grandes échelles, nous développons d'une part des techniques de type diviser-pour-régner pour reconstruire des morceaux partiaux de la scène. D'autre part, pour combiner ces résultats séparés, nous créons une nouvelle méthode qui fusionne rapidement des centaines de maillages. Nous réussissons à reconstruire de beaux maillages urbains et des monuments historiques précis à partir de grandes collections d'images (environ 1600 images de 5M Pixel) / Acquisition of 3D model of real objects and scenes is indispensable and useful in many practical applications, such as digital archives, game and entertainment industries, engineering, advertisement. There are 2 main methods for 3D acquisition : laser-based reconstruction (active method) and image-based reconstruction from multiple images of the scene in different points of view (passive method). While laser-based reconstruction achieves high accuracy, it is complex, expensive and difficult to set up for large-scale outdoor reconstruction. Image-based, or multi-view stereo methods are more versatile, easier, faster and cheaper. By the time we begin this thesis, most multi-view methods could handle only low resolution images under controlled environment. This thesis targets multi-view stereo both both in large scale and high accuracy issues. We significantly improve some previous methods and combine them into a remarkably effective multi-view pipeline with GPU acceleration. From high-resolution images, we produce highly complete and accurate meshes that achieve best scores in many international recognized benchmarks. Aiming even larger scale, on one hand, we develop Divide and Conquer approaches in order to reconstruct many small parts of a big scene. On the other hand, to combine separate partial results, we create a new merging method, which can merge automatically and quickly hundreds of meshes. With all these components, we are successful to reconstruct highly accurate water-tight meshes for cities and historical monuments from large collections of high-resolution images (around 1600 images of 5 M Pixel images)