AI Generated Nude Portrait (2018)

Using an implementation of Progressive Growing of GANs and a corpus of thousands of nude portraits scraped primarily from WikiArt, a neural network was trained to create nude portraits.

The resulting portraits are not realistic. The machine failed to learn the overarching structure of the human body, and instead has fallen into a local minima where it focuses too much on more local features like armpits, breasts, and folds of fat. In a way, these blobs of flesh are the network’s misinterpretation of the nude body. A kind of Francis Bacon / Sol Lewitt mashup that investigates the latent space of flesh as expressed through a global history of representation.