Steve DiPaola's Java-based abstract portraiture

Art imitates naturalist

February 20, 2008

Document Tools

Print This Article

E-mail This Page

Font Size
S      M      L      XL

Related Links

By Terry Lavender and Stuart Colcleugh

A computer program that “evolves” artistic renderings from a portrait of Charles Darwin has landed interactive arts and technology associate professor Steve DiPaola’s work in the journal Nature.

DiPaola’s computer-generated images were used to illustrate an article marking the 200th anniversary of the evolutionary-biology pioneer’s birth in Nature’s February 2008 edition.

DiPaola, who researches 3-D facial systems and intelligent user interfaces, says the Java-based program works by analyzing digitized versions of paintings and then uses evolutionary techniques such as crossover, mutation and survival to create new artworks.

“The evolving programs use recurring, emergent and merged-creative strategies to become good abstract portraitists,” he says.

Mimicking evolution to create art is not new, says DiPaola, and employing aesthetic criteria to judge fitness started well before computers.

“Selective breeding practices, where a human selects the parents for each generation from a given evolved set of choices, are the basis for centuries of ‘creatively’ modified roses, corn, dogs, cows and so on. Current evolutionary art systems borrow from this time-tested approach.”

But DiPaola selects for creativity as well as for best evolutionary practices. And it’s a slow process. His first prototype ran on a high-end PC for 50 days.

DiPaola says his system can be used for hands-on interactive experimentation to learn how evolution — and creativity — work. He’s currently working with the Vancouver Aquarium on an interactive system that evolves marine animal movement and survival strategies. The system will help children learn in a playful and open-ended way how ocean animals evolve differently.

“But our main research direction,” says DiPaola, “is to keep exploring computer creativity as a technique to better understand how human creativity works.”
Search SFU News Online