ARTIFICIAL IMAGES

I teach live, multiweek machine learning art courses in the Artificial Images class series with Derrick Schultz.

In live, online classes of 15-20 students, we teach artists and designers how to use machine learning and artificial intelligence in their creative practice— specifically, how to generate images and video. We record and open-source all course materials on the Artificial Images YouTube channel.

Since February 2020, we have developed and taught seven multi-week ML art courses together.

Here are a select few courses:

1. RunwayML Deep Dive: an introductory course focusing on RunwayML, a tool that allows artists to use machine learning tools without any coding or server knowledge. Contains high-level overviews and demos of many different models, including First Order Motion, Style Transfer, StyleGAN, Super-resolution, AttnGAN, GPT-2, and more. Class notes by me here.

2. StyleGAN2 Deep Dive: an advanced technical course focusing on the StyleGAN2 model— training, testing, image projection, vector interpolation, setting up your own GPU, and more. Class notes by me here.

3. Artificial Images: a mid-level course on machine learning/AI in the image-making context. Models covered include Style Transfer, Next Frame Prediction, StyleGAN2, as well as how to train your own models from command line and compile your own dataset for training.