Comparison between Genetic Algorithm and Genetic Programming Performance for Photomosaic Generation Shahrul Badariah Mat Sah 1 , Vic Ciesielski 1 , Daryl D’Souza 1 , and Marsha Berry 2 1 School of Computer Science and Information Technology 2 School of Creative Media, RMIT University, GPO Box 2476V Melbourne Victoria 3001, Australia smatsah@cs.rmit.edu.au, {vic.ciesielski,daryl.dsouza,marsha.berry} @rmit.edu.au Abstract. Photomosaics are a new form of art in which smaller digital images (known as tiles) are used to construct larger images. Photomosaic generation not only creates interest in the digital arts area but has also attracted interest in the area of evolutionary computing. The photomo- saic generation process may be viewed as an arrangement optimisation problem, for a given set of tiles and suitable target to be solved using evolutionary computing. In this paper we assess two methods used to represent photomosaics, genetic algorithms (GAs) and genetic program- ming (GP), in terms of their flexibility and efficiency. Our results show that although both approaches sometimes use the same computational effort, GP is capable of generating finer photomosaics in fewer genera- tions. In conclusion, we found that the GP representation is richer than the GA representation and offers additional flexibility for future photo- mosaics generation. Keywords: Photomosaic, Genetic Programming (GP), Genetic Algo- rithm (GA). 1 Introduction Photomosaics are a new form of digital mosaic composed of a tessellation of thumbnail pictures known as tiles. When viewed from afar, the subject becomes evident as we perceive the association of the tiles rather than the individual tiles. When viewed close up, the subject is invisible as the details of each tile emerge. An example of a photomosaic appears in Figure 1. In a common approach photomosaics are generated by distributing one or more copies of tiles from among a set of small, carefully selected image tiles, across a two-dimensional gridded canvas. Besides being artistically interesting, a photomosaic may be viewed as a solution to a combinatorial optimisation problem based on the examples given by Mitchell et al [3]. Since photomosaic generation involves a tile collection and a set of fixed locations on a gridded X. Li et al. (Eds.): SEAL 2008, LNCS 5361, pp. 259–268, 2008. c  Springer-Verlag Berlin Heidelberg 2008