A regular texture is formed from a regular congruent tiling of perceptually meaningful texture elements, also known as textons. If the tiling statistically deviates from regularity, either by texton structure, colour, or size, the texture is called near-regular. If we continue to perturb the tiling, the texture becomes stochastic. The set of possible textures that lie between regular and stochastic make up the texture spectrum: regular, near-regular, regular, near-stochastic, and stochastic. <p>In this thesis we provide a solution to the problem of creating, from a near-regular texture, a lattice which defines the placement of textons. We divide the problem into two distinct sub-areas: finding textons within an image, and lattice creation using both an ad-hoc method and a graph-theoretic method. <p>The problem of finding textons within an image is addressed using correlation. A texton selected by the user is correlated with the image and points of high correlation are extracted using non-maximal suppression. To extend this framework to irregular textures, we present early results on the use of feature space during correlation. We also present a method of correcting for a specific type of error in the texton finding result using frequency-space analysis. <p>Given texton locations, we provide two methods of creating a lattice. The ad-hoc method is able to create a lattice in spite of inconsistencies in the texton locating data. However, as texture becomes irregular the ad-hoc lattice construction method fails to correctly connect textons. To overcome this failure we adapt methods of creating proximity graphs, which join two textons whose neighbourhoods satisfy certain criteria, to our problem. The proximity graphs are parameterized for selection of the most appropriate graph choice for a given texture, solving the general lattice construction problem given correct texton locations. <p>In the output of the algorithm, centres of textons will be connected by edges in the lattice following the structure of texton placement within the input image. More precisely, for a texture T, we create a graph G = (V,E) dependent on T, where V is a set of texton centres, and E ={(v_i, v_j)} is a set of edges, where v_i, v_j are in V. Each edge e in E connects texton centre v in V to its most perceptually sensible neighbours.

A regular texture is formed from a regular congruent tiling of perceptually meaningful texture elements, also known as textons. If the tiling statistically deviates from regularity, either by texton structure, colour, or size, the texture is called near-regular. If we continue to perturb the tiling, the texture becomes stochastic. The set of possible textures that lie between regular and stochastic make up the texture spectrum: regular, near-regular, regular, near-stochastic, and stochastic. <p>In this thesis we provide a solution to the problem of creating, from a near-regular texture, a lattice which defines the placement of textons. We divide the problem into two distinct sub-areas: finding textons within an image, and lattice creation using both an ad-hoc method and a graph-theoretic method. <p>The problem of finding textons within an image is addressed using correlation. A texton selected by the user is correlated with the image and points of high correlation are extracted using non-maximal suppression. To extend this framework to irregular textures, we present early results on the use of feature space during correlation. We also present a method of correcting for a specific type of error in the texton finding result using frequency-space analysis. <p>Given texton locations, we provide two methods of creating a lattice. The ad-hoc method is able to create a lattice in spite of inconsistencies in the texton locating data. However, as texture becomes irregular the ad-hoc lattice construction method fails to correctly connect textons. To overcome this failure we adapt methods of creating proximity graphs, which join two textons whose neighbourhoods satisfy certain criteria, to our problem. The proximity graphs are parameterized for selection of the most appropriate graph choice for a given texture, solving the general lattice construction problem given correct texton locations. <p>In the output of the algorithm, centres of textons will be connected by edges in the lattice following the structure of texton placement within the input image. More precisely, for a texture T, we create a graph G = (V,E) dependent on T, where V is a set of texton centres, and E ={(v_i, v_j)} is a set of edges, where v_i, v_j are in V. Each edge e in E connects texton centre v in V to its most perceptually sensible neighbours.

Made available in DSpace on 2014-06-12T15:56:56Z (GMT). No. of bitstreams: 2 arquivo3093_1.pdf: 6869100 bytes, checksum: 84babf04dc9aa424f94c497c190576d6 (MD5) license.txt: 1748 bytes, checksum: 8a4605be74aa9ea9d79846c1fba20a33 (MD5) Previous issue date: 2010 / Conselho Nacional de Desenvolvimento Científico e Tecnológico / Nós apresentamos um método eficiente para síntese de padrões definidos como uma coleção de elementos vetoriais bidimensionais, a partir de uma amostra do padrão. Soluções recentes para este problema fazem uso de triangulação da entrada ou de medidas estatísticas da amostra para controlar o estágio de síntese. Nós propomos um método aplicável a texturas coloridas, desde regular até estocásticas, e que provê controle local sobre a densidade dos elementos. A amostra é segmentada em grupos de elementos similares e definimos uma nova métrica, que não ignora elementos isolados, para cálculo de distância entre vizinhanças de elementos, para comparar vizinhanças diferentes e incompletas. O laço principal de síntese consiste em um crescimento procedural, onde sementes são substituídas por referências a elementos da amostra, gerando novas sementes até que o espaço de síntese seja preenchido. Os resultados mostram a mesma qualidade visual de trabalhos anteriores, e resolvem padrões não abordados em trabalhos anteriores. Nós também mostramos que este método pode ser estendido para sintetizar padrões vetoriais sobre malhas poligonais

In this thesis a new image-based texturing method has been developed. This new method allows users to synthesize tiled textures that can be mapped to any quadrilateral mesh without discontinuity or singularity. An interface has been developed that allows user control over out put textures. Three methods have been included in the interface to create a periodic looking texture for 3D models and two methods have been developed to create wallpaper images (repeating textures on a 2D surface). Using these texturing methods, texturing problems are simplified, and more time can be spent solving artistic problems.

In this thesis a new image-based texturing method has been developed. This new method allows users to synthesize tiled textures that can be mapped to any quadrilateral mesh without discontinuity or singularity. An interface has been developed that allows user control over out put textures. Three methods have been included in the interface to create a periodic looking texture for 3D models and two methods have been developed to create wallpaper images (repeating textures on a 2D surface). Using these texturing methods, texturing problems are simplified, and more time can be spent solving artistic problems.

Traditionally, human texture perception has been studied using artificial textures made of random-dot patterns or abstract structured elements. At the same time, computer algorithms for the synthesis of natural textures have improved dramatically. The current study seeks to unify these two fields of research through a psychophysical assessment of a particular computational model, thus providing a sense of what image statistics are most vital for representing a range of natural textures. We employ Portilla and Simoncelli’s 2000 model of texture synthesis for this task (a parametric model of analysis and synthesis designed to mimic computations carried out by the human visual system). We find an intriguing interaction between texture type (periodic v. structured) and image statistics (autocorrelation function and filter magnitude correlations), suggesting different processing strategies may be employed for these two texture families under pre-attentive viewing.

Made available in DSpace on 2015-03-05T13:53:45Z (GMT). No. of bitstreams: 0 Previous issue date: 2 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Este trabalho apresenta um modelo para síntese de texturas a partir de amostras, mais especificamente para síntese de texturas com elementos que variam progressivamente ao longo da textura, conhecidas como Progressively-Variant Textures (PVT). Este tipo de textura é comum na Natureza, particularmente na pelagem de animais como leopardos. Os algoritmos de síntese existentes, na grande maioria, não permitem qualquer controle sobre o resultado final, ou então, os que permitem, são limitados ou tem problemas de performance inerente ao método de síntese utilizado (pixel-a-pixel). Para resolvermos estes problemas, propomos um algoritmo baseado numa nova unidade de síntese: o texel. O nosso algoritmo sintetiza uma nova textura agrupando texels que satisfazem um critério de similaridade numa vizinhança. As características de uma PVT são obtidas com transformações afim e operações morfológicas aplicadas sobre os texels e definidas pelo usuário. Os resultados obtidos apresentam uma ótima qualidade visual para uma / We present a model for texture synthesis from samples, particularly for synthesis of textures with smooth variation of the texture elements, known as Progressively-Variant Textures (PVT). This type of textures is common in Nature, specially in animal coat markings such as leopards. Current solutions, for the most part, do not allow much control by the user or the ones which do allow control are limited or have problems due to the pixel-at-a-time nature of synthesis. We address these problems with the introduction of an algorithm based on a new building block for synthesis: the texel. Our solution builds a new texture grouping texels which satisfy a neighborhood similarity criterion. The PVT features are introduced with affine transformations and morphological operators defined by the user and applied to the texels. The results show a good visual quality for a large number of sample textures, including cases where current solutions for natural textures fail. Besides, the many possibilities for variation of te

Existem milhares de espécies de serpentes no mundo, muitas com padrões distintos e intricados. Esta diversidade se torna um problema para usuários que precisam criar texturas de pele de serpente para aplicar em modelos 3D, pois a dificuldade em criar estes padrões complexos é considerável. Nós primeiramente propomos uma categorização de padrões de pele de serpentes levando em conta suas características visuais. Então apresentamos um modelo procedural capaz de sintetizar uma vasta gama de textura de padrões de pele de serpentes. O modelo usa processamento de imagem simples (tal como sintetizar bolinhas e listras) bem como autômatos celulares e geradores de ruído para criar texturas realistas para usar em renderizadores modernos. Nossos resultados mostram boa similaridade visual com pele de serpentes reais. As texturas resultantes podem ser usadas não apenas em computação gráfica, mas também em educação sobre serpentes e suas características visuais. Nós também realizamos testes com usuários para avaliar a usabilidade de nossa ferramenta. O escore da Escala de Usabilidade do Sistema foi de 85:8, sugerindo uma ferramenta de texturização altamente efetiva. / There are thousands of snake species in the world, many with intricate and distinct skin patterns. This diversity becomes a problem for users who need to create snake skin textures to apply on 3D models, as the difficulty for creating such complex patterns is considerable. We first propose a categorization of snake skin patterns considering their visual characteristics. We then present a procedural model capable of synthesizing a wide range of texture skin patterns from snakes. The model uses simple image processing (such as synthesizing spots and stripes) as well as cellular automata and noise generators to create realistic textures for use in a modern renderer. Our results show good visual similarity with real skin found in snakes. The resulting textures can be used not only for computer graphics texturing, but also in education about snakes and their visual characteristics. We have also performed a user study to assess the usability of our tool. The score from the System Usability Scale was 85:8, suggesting a highly effective texturing tool.

<div>In this thesis, we investigate methods for texture synthesis and texture re-rendering of indoor room scene images. The goal is to create a photorealistic redesign of interior spaces by replacing surface finishes with a new product based on a single room scene image. Specically, we focus on automating this process to reduce manual input while enabling high-quality and easy-to-use experience. The most common method of rendering textures into a scene is called texture mapping. Texture mapping involves mapping pixels in a texture sample to vertices in an object model. Typically, a large texture sample is required to perform texture mapping properly. Given a small texture sample, texture synthesis creates a large sized texture that appears to have</div><div>been made by the same underlying process. In the first part of this thesis, we present a method of texture synthesis that automatically determines a set of parameters to produce satisfactory results based on the texture types. The next challenge is to create a photorealistic re-rendering of the synthesized texture in the room scene image. 3D scene information such as geometry, lighting and reflectance is crucial to making the re-rendered image realistic. These properties contribute to the image formation process and must be estimated to create a scene-consistent modication. Knowing these parameters allows effects like highlights, shadows and inter-object reflections to be maintained during the re-rendering process. We detail methods for estimating</div><div>these parameters from a single indoor image. Finally, we will show a web-based implementation of these methods using the WebGL library ThreeJS.</div>

<p>We present a concept for using procedural techniques to represent media. Procedural methods allow us to represent digital media (2D images, 3D environments etc.) with very little information and to render it photo realistically. Since not all kind of content can be created procedurally, traditional media representations (bitmaps, polygons etc.) must be used as well. We have adopted an object-based media representation where an object can be represented either with a procedure or with its traditional representation. Since the objects are created on the client the procedures can be adapted to its properties such as screen resolution and rendering performance. To keep the application as small and flexible as possible, each procedure is implemented as a library which is only loaded when needed. The media representation iswritten in XML to make it human readable and easy editable. The application is document driven where the content of the XML document determines which libraries to be loaded. The media objects resulting from the procedures is composited into the media representation preferred by the renderer together with the non-procedural objects. The parameters in the XML document are relative to parameters determined by the system properties (resolution, performance etc.) and hence adapt the procedures to the client. By mapping objects to individual libraries, the architecture is easy to make multi threaded and/or distributed.</p>