“Generating images from an internet grab bag” with CLIP+VQGAN
AI Weirdness • Generating images from an internet grab bag. This is pretty neat –
Still amazed by this:
Here’s CLIP+VQGAN (trained on internet photos and their accompanying text), prompted two different ways:
“A car driving down a desert road in monument valley”
These are really cool to play around with, there is a how-to google doc here, on how to make these. I tried to contact to author of the guide to see if they are interested in getting it on our learn system, if I can’t find them I’ll see if we can make one that has some more details, photos, video, etc. and how this could be done in other ways too.
Post a Comment