Exploring latent space + GAN Image creation + Medieval bestiaries
(thisisnotabesitary)
I thought it would be cool to build a type of medieval bestiary using image Generative Adversarial Networks to create images, maybe even based on an alignment chart considering how these texts were morally coded.
Some thoughts:
- Adding a transitional/interpolation between creations (maybe even with the added layer of moral alignment).
- Using latent directions to modify images
- Building on our understanding of specific medieval cultures, could be a series with different religious backgrounds/ideas
This was hugely inspired by the Met Museum’s ‘Gen Studio’ and my own interest with visualizing datasets like the GeoCities archive.
This falls under the "preferably good" category in terms of using AI for history.
A silly way to interact with historical thinking and creativity, helping us immerse ourselves in past cultures.