Re:Imagen, a text-to-image diffusion model
by dclowd9901 on Tuesday May 24, 2022 @02:26AM
(#31488665)
Looking at these… I can’t help but wonder if these are literal examples of AI imagination?
[ Parent
| on HN ]
Re:Imagen, a text-to-image diffusion model
by joeycodes on Tuesday May 24, 2022 @02:29AM
(#31488689)
I've started to ask myself if my own creativity is a result of random sampling from the diffusion tapestry of associated memories and experience on that topic.
Re:Imagen, a text-to-image diffusion model
by albertzeyer on Tuesday May 24, 2022 @02:51AM
(#31488821)
What else could creativity possible be?
I do wonder what Dall-E 2 would output for a request along the lines of "A still life of a vase of flowers in a completely new art style."
Re:Imagen, a text-to-image diffusion model
by zimpenfish on Tuesday May 24, 2022 @05:55AM
(#31490025)
Don't have access to Dall-E 2 or Imagen but I do have [1] and [2] locally and they produced [3] with that prompt.
[1] https://github.com/nerdyrodent/VQGAN-CLIP.git [2] https://github.com/CompVis/latent-diffusion.git [3] https://imgur.com/a/dCPt35K
Nice. Latent-diffusion has come out very traditional but the VQGAN/CLIP ones are fairly original.
Re:Imagen, a text-to-image diffusion model
by zimpenfish on Tuesday May 24, 2022 @05:04PM
(#31497525)
From my experiments, the LD one doesn't seem to have been trained on as big or as tagged data set - there's a whole bunch of "in the style of X" that the VQGAN knows* about but the LD doesn't. That might have something to do with it.