By Robert. A Gonsalves
For this project, I gathered images of abstract paintings from WikiArt.org, processed them, and fed them into StyleGAN2 at the size of 1024x1024. I trained the GAN for three weeks on a GPU using Google Colab. I then processed the output images by adjusting the aspect ratio and running them through another ANN for a super-resolution resize. The resultant images are 4096 pixels wide or tall, depending on the aspect ratio.
You can read my article about this project on Medium here: MachineRay: Using AI to Create Abstract Art
This GitHub repositoray contains the source code for the project.
The source code is released under the CC BY-NC-SA license. See License.txt for details.
