Skip to content

Interactively evolve various types of art (pictures, animations, shapes, and sounds) using Compositional Pattern Producing Networks

License

Notifications You must be signed in to change notification settings

schrum2/CPPNArtEvolution

Repository files navigation

CPPNArtEvolution

This project contains several forms of interactive art evolution, all based on CPPNs. All code in this project is part of the main MM-NEAT project, but this project is simplified to remove much extraneous code not required for interactive evolution. This project is also much easier to use and launch. Just follow the instructions below:

Windows

Just double-click any of the batch files with the Launch prefix:

  • Launch-Picbreeder.bat: Evolve 2D pictures, just like the original Picbreeder
  • Launch-AnimationBreeder.bat: Evolve 2D animations!
  • Launch-3DObjectBreeder.bat: Evolve 3D shapes, in a manner similar to Endless Forms
  • Launch-3DAnimationBreeder.bat: Evolve 3D animations!
  • Launch-Breedesizer.bat: Evolve tones that can be used to play MIDI files, similar to the original Breedesizer

These batch files all use the CPPNArtEvolution.jar file that is already in the repository. However, if you want to recompile this jar file, you can use the build.xml ANT build script, or create an executable jar file with the class edu.southwestern.MMNEAT.MMNEAT as the main class.

Mac/Linux/Unix

The batch files listed above do not use any Windows-specific commands, so they can easily be executed as bash scripts as well. Alternately, you can just copy-paste the command inside of any of these files to your terminal.

Help

If you need any help, then please contact me at [email protected]!

Also, you can evolve neural networks for lots of other interesting applications by using the original MM-NEAT project, which contains the interactive evolution code from this project, along with code to evolve agent behavior for various tasks (Ms. Pac-Man, Tetris, and Mario to name a few)

Citing

If you use any of this code for any project or publication, please cite the following paper:

@inproceedings{tweraser:gecco2018,
	title={Querying Across Time to Interactively Evolve Animations},
	author={Tweraser, Isabel and Gillespie, Lauren E and Schrum, Jacob},
	year={2018},
	booktitle={Proceedings of the Genetic and Evolutionary Computation Conference (GECCO 2018)},
	month={July},
	numpages = {8},
	url = {https://people.southwestern.edu/~schrum2/SCOPE/tweraser.gecco18.pdf},
	doi = {10.1145/3205455.3205460},
	publisher = {ACM},
	address = {New York, NY, USA},
	location={Kyoto, Japan}
}

About

Interactively evolve various types of art (pictures, animations, shapes, and sounds) using Compositional Pattern Producing Networks

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published