“Intelligent brush strokes” by Wexler and Dezeustre

  • ©Daniel Wexler and Grills Dezeustre




    Intelligent brush strokes



    We present a fully automatic brush stroke placement algorithm for non-photorealistic rendering leveraging importance sampling techniques. Smarter brush placement enables creation of images using only a handful of well-placed strokes, avoiding the “textured” look common to prior techniques. We focus on the problem of converting source images to painterly renderings, but our technique could be extended to work on video source material or to generate abstract paintings without any source imagery. We leverage genetic search algorithms to evaluate the quality of the final images, guiding the solution through the large space of potential images.


    1. Hertzmann, Aaron, 1998. Painterly Rendering with Curved Brush Strokes of Multiple Sizes. In SIGGRAPH’98, Conference Proceedings, 453–460.
    2. Litwinowicz, Peter, 1997. Processing Images and Video for an Impressionist Effect. In SIGGRAPH ’97, Conference Proceedings, 407–414.
    3. Pharr, M., and Humphreys, G.. 2004. Infinite Area Light Source with Importance Sampling. In Pysically Based Rendering: From Theory to Implementation, PBRT Plugins from the Authors, http://pbrt.org/plugins/infinitesample.pdf.
    4. Zhao, M., and Zhu, S. C., 2011. Customizing Painterly Rendering Styles Using Stroke Processes. In NPAR’11, Proceedings of the ACM SIGGRAPH/Eurographics Symposium on Non-Photrealistic Animation and Rendering, 137–146.

ACM Digital Library Publication:

Overview Page: