ArtGAN: Artwork Synthesis with Conditional Categorical GANs
This addresses the challenge of synthesizing abstract artwork, which is more difficult than generating natural images, offering a domain-specific advancement for art generation.
The paper tackles the problem of generating complex artwork images by proposing ARTGAN, an extension of GANs that uses label feedback to improve learning, resulting in realistic artwork and natural-looking images on CIFAR-10 with clear shapes.
This paper proposes an extension to the Generative Adversarial Networks (GANs), namely as ARTGAN to synthetically generate more challenging and complex images such as artwork that have abstract characteristics. This is in contrast to most of the current solutions that focused on generating natural images such as room interiors, birds, flowers and faces. The key innovation of our work is to allow back-propagation of the loss function w.r.t. the labels (randomly assigned to each generated images) to the generator from the discriminator. With the feedback from the label information, the generator is able to learn faster and achieve better generated image quality. Empirically, we show that the proposed ARTGAN is capable to create realistic artwork, as well as generate compelling real world images that globally look natural with clear shape on CIFAR-10.