Results obtein after training a Stylegan2 neural network with the “Puri & me” serie. I grouped the results according to the similarity to the original paintings.

We should not forget that these neural network learn about forms and visual patterns, but it does not recognice what everything is, how things are called or what do they mean.

