Evolving images for entertainment

  • Authors:
  • Qinying Xu;Daryl D'Souza;Vic Ciesielski

  • Affiliations:
  • RMIT University, Melbourne, Australia;RMIT University, Melbourne, Australia;RMIT University, Melbourne, Australia

  • Venue:
  • IE '07 Proceedings of the 4th Australasian conference on Interactive entertainment
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

Images are widely used in media contexts such as web design, games and video animation. The process of creating interesting images can be enjoyable if a useful tool is involved. In this paper we describe an interactive image generation tool called IMAGENE based on genetic programming, which can be used to create novel, surprising, and sometimes stunning images. A new method for presenting colour images is also proposed, which results in more colourful images being generated. The system permits the user to progressively evaluate and generate new images from previous sets of images. In doing so, the user is also able to control various settings and parameters. From a user perspective, the system has the following qualities which make generating images entertaining: can be created by choosing different function settings and families of images which have common characteristics between parents and children can be generated.