I've generated a lot of terrible unicorn cakes by this point.
Last time I experimented with generating cakes
[https://www.aiweirdness.com/ai-cake-fails/], using CLIP's internet training to
guide a couple of image generating methods. Around the time I posted my
experiments, RiversHaveWings, who developed one of the most popular
Still amazed by this:
(trained on internet photos and their accompanying text), prompted two different
"A car driving down a desert road in monument valley"
A car driving down a desert road in monument valley"A car driving down a
There are upsides to working with a neural net that trained on a huge collection
of internet images and text. One is that, instead of ominous grey geometric
when it doesn't understand your prompt (there is a free interactive demo of
“Frodo Baggins delivering a pizza through the mines of Moria”
Remember my attempts to get CLIP+BigGAN to generate candy hearts
[https://ai-weirdness.ghost.io/curse-your-valentine-with-these-candy/]? Here’s what an
alternative method, CLIP+FFT [https://github.com/eps696/aphantasia], does with
the prompt “a candy heart with a message”.
“Leaning Tower of Pizza”, BigGAN steered by CLIP using Big SleepI wrote earlier about DALL-E [https://www.aiweirdness.com/the-drawings-of-dall-e/], an image
generating algorithm recently developed by OpenAI. One part of DALL-E
[https://openai.com/blog/dall-e/]’s success is another algorithm called CLIP