Texture Synthesis of Contrasting Natural Patterns

The seamless integration of the shape and visual attributes of virtual objects is still one of the greatest challenges in Computer Graphics. For some natural objects, such as patterned animals, shape and appearance are mutually connected and therefore the individual treatment of these two aspects difficult the whole process and limits the visual results. One approach to solve this problem is to create shape and appearance together, thus generating so-called intelligent textures, since they can adapt to the surface of the object according to geometric information. The Clonal Mosaic model presented an approach for intelligent texturing of fur patterns seen in some mammals, particularly the big cats and giraffe. This paper extends this model to account for biologically plausible contrasting fur patterns, mostly seen in black and white, either regular - as seen in zebras, or irregular - as seen in cows and horses, among other animals. The main contributions of this work are the addition of a neural crest model, local control for parameters, and also vector field definition on the object's surface for simulation control. The results synthesized for various mammals with contrasting patterns such as cows, horses, and zebras, and other contrasting patterns found in frogs, for example, confirm the advantages of an integrated approach such as the one provided by the extended Clonal Mosaic procedural model.