Colab Notebooks: ART in ARTificial Intelligence

There are many approaches to train AI on Artworks.

One of them was provided via Reddit: StyleGAN trained on Artwork Dataset with 24k images from Kaggle.

You get interesting results, it could even be possible to trace the original artworks the model was trained on.

New art game, everybody? “Visual Etymology”

Another one is WikiART StyleGAN2 Conditional Model, provided by Peter Baylies (et al) and packed into a notebook by Doron Adler:

This model was trained on WikiART images. It allows even to choose between artists, genres and styles.

And the results are impressive:

Things to try out:

• Every new combination produces interesting artworks. Try to select artists from different epochs with an untypical style. Like Picasso & Renaissance or Shishkin & Pop Art. The results are sometimes unexpected and not always comprehensible, but that’s art after all.

Links:

• _C0D32_ Colab Notebook (trained on 24k Artworks)
• WikiART StyleGAN2 Colab Notebook

Read also:

Index of Series "Google Colab Notebook".

Full essay "12 Colab Notebooks that matter"