Bayesian Rendering with Non-Parametric Multiscale Prior Model

This paper investigates the use of the Bayesian inference for devising an example-based rendering procedure. As prior model of this Bayesian inference, we exploit the multiscale non-parametric model recently proposed by Wei et al. for texture synthesis. This model appears to be interesting to also capture some characteristics of a rendering style from an artistic illustration example. Obtained results, with a prior model capturing the rendering style of drawing samples or trained with synthetic and real input textures, are presented. Our results indicate that the proposed method allows to simulate automatic synthesis of various illustration style. More generally, the proposed scheme is able to re-render an input image in the style of an other image allowing, in this way, to create a very broad range of artistic and visual effects.