Stylized multiresolution image representation

bstract. We integrate stylized rendering with an efficient multiesolution image representation, enabling a user to control how ompression affects the aesthetic appearance of an image. We dopt a point-based rendering approach to progressive image transission and compression. We use a novel, adaptive farthest point ampling algorithm to represent the image at progressive levels of etail, balancing global coverage with local precision. A progresively generated discrete Voronoi diagram forms the common founation for our sampling and rendering framework. This framework llows us to extend traditional photorealistic methods of image reonstruction by scattered data interpolation to encompass nonphoorealistic rendering. It supports a wide variety of artistic rendering tyles based on geometric subdivision or parametric procedural texures. Genetic programming enables the user to create original renering styles through interactive evolution by aesthetic selection. e compare our results with conventional compression, and we iscuss the implications of using nonphotorealistic representations or highly compressed imagery. © 2008 SPIE and IS&T. DOI: 10.1117/1.2898894