Artisic Vision: Automatic Digital Painting Using Computer Vision Algorithms
Author(s): Bruce Gooch.
Master Thesis: University of Utah, May,
2001.
[BibTeX] 
Abstract:
This thesis presents a method for creating simulated oil paintings. A raster image is next used as input to an algorithm that produces a painting-like image composed of brush strokes rather than pixels. Ultimately the sequence of brush strokes representing an interpreted image can be rendered in pixel form, however the brush stroke structure is far more compact for storage and transmission. Unlike previous automatic painting methods, this algorithm attempts to use very few brush-strokes. The algorithm achieves economy of brush strokes by first segmenting the image into features, finding the medial axes points of these features, converting the medial axes points into ordered lists of image tokens, and finally rendering these lists as brush strokes. The process creates images reminiscent of modern realist painters who often want an abstract or sketchy quality in their work.