Creating Art Using Generative Adversarial Networks
Henry Abbott · Digital Kenyon (Kenyon College) · 2019
I set out to explore how artificial intelligence generated artwork compares to human created artwork. Using CycleGANs, my goal was to create art, and then conduct a Turing Test in order to determine whether or not one could distinguish between A.I. and human art. Using this test, as well as my own observations, I planned to analyze where CycleGANs succeeds in creating art, as well as where it struggles. Following this analysis, I thought more about what the results mean for art moving forward, and if artificial intelligence can truly create art.