Conditional GANs for painting generation

Adeel Mufti,Biagio Antonelli,Julius Monello
DOI: https://doi.org/10.1117/12.2556551
2020-01-31
Abstract:We examined the use of modern Generative Adversarial Networks to generate novel images of oil paintings using the Painter By Numbers dataset. We implemented Spectral Normalization GAN (SN-GAN), and compared its outputs to a Deep Convolutional GAN. Visually, and quantitatively according to the Sliced Wasserstein Distance metric, we determined that the SN-GAN produced paintings that were most comparable to our training dataset. We then performed a series of experiments to add supervised conditioning to SN-GAN, the culmination of which is what we believe to be a novel architecture that can generate face paintings with user-specified characteristics.
What problem does this paper attempt to address?