3 ms·

You seem to be under the impression that SD can only generate original art. However, it will literally recreate existing paintings for you if you just prompt it

You seem to be under the impression that SD can only generate original art. However, it will literally recreate existing paintings for you if you just prompt it with the title. Identical composition and everything.

Filligree 4y ago
I’m curious. Can you give an example of that happening for a painting that’s still in copyright?
Aerroon 4y ago
Photoshop can also recreate existing paintings if you just "prompt it with the correct input", no?
- codeflo 4y ago
  So you're saying "if I put the made-up straw man argument that I intend to knock down in quotation marks, it's less obvious that it's not actually what the other person wrote"?
  - Aerroon 4y ago
    I'm pointing out that the model has a piece missing - the input from the user. You can't just flippantly dismiss it, because input from the user is a crucial part of recreating the works the model is supposedly copying. It's an important factor to consider, because any digital painting program will let you recreate any art work with the correct user input. It's not sufficient to just consider whether it can reproduce an image, but also how much user input was required to do so.
gdubs 4y ago
Models like MidJourney and StableDiffusion can generate very close renditions of some very well known paintings – and if it has enough of a conceptual understanding of even a less known painting, it can do a pretty impressive rendition of that as well. But it's virtually impossible for these models to make an exact replica – a photocopy - of an existing painting, because that would make it break some laws of information theory probably. It's not a lossless compression engine. Paintings like, "Girl With the Pearl Earring" appear so frequently in the datasets that the models tend to overfit on them – which is actually not something you want when designing a model. It tends to create issues for you. But that's why a painting like that can be simulated somewhat accurately. But even then – it's never going to be 100%.

Filligree 4y ago

I’m curious. Can you give an example of that happening for a painting that’s still in copyright?

Aerroon 4y ago

Photoshop can also recreate existing paintings if you just "prompt it with the correct input", no?

codeflo 4y ago
So you're saying "if I put the made-up straw man argument that I intend to knock down in quotation marks, it's less obvious that it's not actually what the other person wrote"?
- Aerroon 4y ago
  I'm pointing out that the model has a piece missing - the input from the user. You can't just flippantly dismiss it, because input from the user is a crucial part of recreating the works the model is supposedly copying. It's an important factor to consider, because any digital painting program will let you recreate any art work with the correct user input. It's not sufficient to just consider whether it can reproduce an image, but also how much user input was required to do so.

gdubs 4y ago

Models like MidJourney and StableDiffusion can generate very close renditions of some very well known paintings – and if it has enough of a conceptual understanding of even a less known painting, it can do a pretty impressive rendition of that as well. But it's virtually impossible for these models to make an exact replica – a photocopy - of an existing painting, because that would make it break some laws of information theory probably. It's not a lossless compression engine. Paintings like, "Girl With the Pearl Earring" appear so frequently in the datasets that the models tend to overfit on them – which is actually not something you want when designing a model. It tends to create issues for you. But that's why a painting like that can be simulated somewhat accurately. But even then – it's never going to be 100%.