No abstract

We have produced a software system that applies genetic programming to non-photorealistic rendering. Users of the system select a rendering algorithm to be used in combination with a genetic programming representation. There are multiple rendering algorithms and genetic programming representations to choose from. Users then select a target image. The system then attempts to render a likeness of the target image. Ove the course of many thousands of iterations, an evolutionary search mechanism is used to optimise the likeness. All incremental renderings are saved and converted into a video file that displays the entire evolutionary sequence. We report on experiments conducted to test the ability of the rendering algorithms to produce engaging art and on the efficiency of our genetic programming representations. We demonstrate that our methodology can be used to create interesting art and that our genetic programming representations are effective. Additionally we demonstrate that in our specific problem domain. faster convergence can be achieved using small populations of small programs.

We describe an approach to generating animations of drawings that start as a random collection of strokes and gradually resolve into a recognizable subject. The strokes are represented as tree based genetic programs. An animation is generated by rendering the best individual in a generation as a frame of a movie. The resulting animations have an engaging characteristic in which the target slowly emerges from a random set of strokes. We have generated two qualitatively different kinds of animations, ones that use grey level straight line strokes and ones that use binary Bezier curve stokes. Around 100,000 generations are needed to generate engaging animations. Population sizes of 2 and 4 give the best convergence behaviour. Convergence can be accelerated by using information from the target in drawing a stroke. Our approach provides a large range of creative opportunities for artists. Artists have control over choice of target and the various stroke parameters.

We describe an artist's journey of working with an evolutionary algorithm to create an artwork suitable for exhibition in a gallery. Software based on the evolutionary algorithm produces animations which engage the viewer with a target image slowly emerging from a random collection of greyscale lines. The artwork consists of a grid of movies of eucalyptus tree targets. Each movie resolves with different aesthetic qualities, tempo and energy. The artist exercises creative control by choice of target and values for evolutionary and drawing parameters.

No abstract