How a Stable Diffusion prompt changes its output for the style of 1500 artists

Funny that the Bob Ross version just makes them look like Bob Ross. Maybe there are more pictures of Bob Ross in the training set than his actual paintings.

Re: How a Stable Diffusion prompt changes its output for the style of 1500 artists

#5

Post by pessimizer » Sun, Oct 02, 2022, 3:55 PM UTC

From an artist to a brand identity.

Re: How a Stable Diffusion prompt changes its output for the style of 1500 artists

#6

Post by norwalkbear » Sun, Oct 02, 2022, 3:57 PM UTC

Truly impressive, though it gives me some great fear for all the people whose career is art. This will likely take the pareto to a new height instead of 20/80 maybe even 1/99

Re: How a Stable Diffusion prompt changes its output for the style of 1500 artists

#7

Post by ShamelessC » Sun, Oct 02, 2022, 3:58 PM UTC

pain2022 wrote: post #3

Quoted post unavailable.

> What is this list?! No Picasso, Renoir or Vrubel present

Comments like this are why people think HN is nothing but a place to receive a shallow dismissal of your efforts.

It's a list. Make your own.

Re: How a Stable Diffusion prompt changes its output for the style of 1500 artists

#8

Post by whywhywhywhy » Sun, Oct 02, 2022, 4:05 PM UTC

UI should allow you to choose from the images and get the artist name.

Re: How a Stable Diffusion prompt changes its output for the style of 1500 artists

#9

Post by ttoinou » Sun, Oct 02, 2022, 4:09 PM UTC

How come the faces features and body parts are almost always at the same place on the image ?

Re: How a Stable Diffusion prompt changes its output for the style of 1500 artists

#10

Post by mjamesaustin » Sun, Oct 02, 2022, 4:09 PM UTC

jerpint wrote: post #2

I wonder how constant everything else is kept, e.g. The seed. It's interesting that all the poses seem to align

I noticed that too! I'm guessing they used the same seeds, which does a great job of providing a comparison in style.