Customizing painterly rendering styles using stroke processes

Abstract

In this paper, we study the stroke placement problem in painterly rendering, and present a solution named stroke processes, which enables intuitive and interactive customization of painting styles by mapping perceptual characteristics to rendering parameters. Using our method, a user can adjust styles (e.g., Fig.1) easily by controlling these intuitive parameters. Our model and algorithm are capable of reflecting various styles in a single framework, which includes point processes and stroke neighborhood graphs to model the spatial layout of brush strokes, and stochastic reaction-diffusion processes to compute the levels and contrasts of their attributes to match desired statistics. We demonstrate the rendering quality and flexibility of this method with extensive experiments.