Picbreeder
  • About
  • Tips
  • Categories
    • Animals
    • Abstract

    • Faces
  • Details
    • Artist
    • Service 02
    • Service 03
Breed more images
Tip: select a parent(s) and press evolve
About
A science publication describing Picbreeder is available here and a presentation here.

Picbreeder is based on an idea by Jimmy Secretan, a graduate student in the School of Electrical Engineering and Computer Science at the University of Central Florida. He proposed his original idea in a meeting of the Evolutionary Complexity Research Group (EPlex). Its inspiration was sufficiently groundbreaking that four other EECS graduate students (Nick Beato, David D'Ambrosio, Adam Campbell, and Adelein Rodriguez) volunteered to work with Jimmy and under the supervision of Dr. Kenneth Stanley (who supervises the EPlex Group) for over a year to realize this vision for the world's largest online collaborative evolution experiment. This website is the result of that effort.

The idea was inspired by a recent proliferation of "NEAT-based Genetic Art" programs and before, which we had been discussing in the EPlex group. NEAT stands for NeuroEvolution of Augmenting Topologies, which is a method for artificially evolving neural networks on computers developed by Ken Stanley when he was a student at the University of Texas at Austin. Prior to Picbreeder, Ken also helped lead the development of an interactively evolutionary video game based on NEAT called NERO. While Genetic Art has been around for many years (for example see Karl Sims), the idea that NEAT-evolved neural networks can be used to produce pictures originated with Mattias Fagerlund in his DelphiNEAT-based Genetic Art program (DNGA). DNGA allowed the user to interactively evolve pictures that are drawn by NEAT neural networks.

While DNGA (in its initial release) produced only black and white images, it exhibited a remarkable property: It was actually possible to evolve images that look like known objects. While prior genetic art systems often produced beautiful images, they were often abstract and difficult to relate to the real world. DNGA, on the other hand, could produce images reminiscent of real things. One reason for this capability is the process of complexification implemented by NEAT, which means that the neural networks become larger and more complex as they evolve. This process means that the user can start with a simple image that captures a general idea and then refine and elaborate on it through evolution, allowing objects like spaceships and faces to evolve fluidly.

DNGA inspired a couple other NEAT-based genetic art programs, including Holger Ferstl's SharpNEAT-based Genetic Art and William Monti's JCPPN. These programs also featured the unique capability to evolve towards recognizable objects.

One consequence of these compelling results was Ken Stanley's theory of Compositional Pattern Producing Networks , which in part explains how networks evolved by NEAT are good at representing patterns with familiar regularities, an idea that is capitalized on by Picbreeder as well.

A perhaps vexing aspect of NEAT-based genetic art programs is that there was no way to share the genomes (i.e. the DNA) of evolved images with other people. Such sharing would be nice because if, for example, you evolved a nice looking spaceship, your friend may want to further evolve it, if only she had access to its genome.

Picbreeder addresses this problem in a most ambitious way: It puts an interactive evolutionary interface in the hands of the public in such a way that every lineage that has ever been produced by anyone can be explored and continued by anyone else. Thus, Picbreeder users are in effect participating in a massive online collaboration to explore the space of possible images.

Because Picbreeder images are evolved by NEAT, lineages never end. Any user can pick up where another left off. Furthermore, pictures evolved in Picbreeder have infinite resolution (because they are stored as mathematical objects), which means they are publication quality.

Perhaps most intriguing is that we do not know what is possible to evolve. As lineages are built upon lineages, complexification in NEAT means that images can theoretically become more and more complex. Thus, in theory, anything is possible. In this sense Picbreeder's users are pioneers, exploring collectively a space that has never been exhaustively explored, discovering together places that no one of us could have discovered alone.

Breeding Tips

There are a few facts about how evolution happens in Picbreeder that can help you evolve better images.

It's important to realize that the "DNA" in Picbreeder is designed to become more complex as more generations pass. That means that early on in a lineage, your best bet is to produce simple things. As you go along, images will naturally become more interesting, because the DNA actually expands as it evolves.

However, if you branch from an image that has already been evolved for a while, you get more advanced complexity right away. So if you want to start seeing a lot of detail right away, look for a good image from which to branch, either from yourself or from another user.

A funny thing about complex images is that while they may change a lot when you spawn a new generation, often they have a way of morphing into something that's still very interesting! Be on the lookout for these kinds of really interesting transformations. Everyone benefits when you publish something that represents a whole new kind of idea.

Because images can sometimes transform, it's possible to go from one type of image to another. For this reason, it is not always necessary to find exactly the right thing to start from. In particular, look for regularities in the image that you might want. For example, is it symmetric? If you want symmetry, even if the image is not something you like, it might be a good starting point.

To be a good breeder it helps to think about how things can be related. For example:

  • A pair of binoculars might be a good seed for a face because the two circles can becomes eyes.
  • A moon can become a sun if the shades of darkness invert.
  • A spaceship can become a butterfly by borrowing the wings.
  • A five-pointed star can become a humanoid by growing out its limbs.
  • A "C" can become a telephone by gaining some definition.

If you think in this way, you can see how many things are reachable in a whole new way.

Finally, because complexity is designed to increase over generations, the longer a lineage becomes, the more novel discoveries become possible. So it's a good idea to spend some time working on long lineages (either your own or someone else's) to bring them even farther along the path of discovery.

Because Picbreeder is a community of explorers, what you are doing as an individual user is not just finding new images, but creating new paths from which others may embark. By participating in this process, you are helping to make possible more than what you have done alone.