Présentée par

Processing graphical data, either for its editing or the synthesis of new content, demands agood balance between the different sources of information one may exploit. Unlike "proce-dural" techniques, synthesis by example stands out thanks to its extreme ease-of-use : indeed,tasks such as identification, analysis and reproduction of the distinguishing features of the user-provided examples are left to the method itself. Such approaches, along with today’s intricateediting methods have greatly favored the production of compelling graphical content at a widescale, and henceforth facilitated the adoption of computer-assisted tools by artists. But in orderto meet with success, they also have to be highly controllable without being a mere extensionof the artist’s hand. We explore here such concerns in the context of expressive rendering andstudy the interactions, may they be collaborative or competitive, between the different sourcesof information at the core of such processes. In our opinion, there are three main sources ofinformation: the automatic analysis of the inputs before processing; the use of prior knowledgethrough predetermined models; and users’ explicit intervention. Through a clever combinationof these sources, we propose new expressive synthesis techniques which satisfy the aforemen-tioned usability. More than photographic realism, expressive rendering strives for the fulfillmentof less easily quantifiable goals such as the intelligibility or the aesthetic value of its results. Thesubjectivity behind the assessment of such criteria thus forces us to attach much importanceto the careful choice of the source of information to favor; the required amount of user inter-vention (without being detrimental to the method’s theoretical value); and the possible resort toprior models (without endangering its generality). Three main synthesis instances are studiedin this document: texture generation, image de-colorization, and artistic line rendering. Thegreat disparity of inputs (raster and vector textures, complex images, 3d meshes), terms of syn-thesis (imitation, conversion, depiction) and objectives (preservation of a texture’s visual signa-ture, plausible restitution of chromatic contrasts, creation of drawings in accordance with users’styles) gives rise to distinct balances between those sources of information and requires the con-sideration of various modes of user interaction.