News for hackers. Stuff that matters.
Re:A Web UI for Stable Diffusion by MattRix on Saturday September 10, 2022 @01:27AM (#32788055)
This is exactly it. It’s pretty remarkable that it was trained on over 100 terabytes of images and yet the model has been distilled down to only 4gb.
Re:A Web UI for Stable Diffusion by rassibassi on Saturday September 10, 2022 @02:48AM (#32788441)
Yes, and another reason for the small model size and the novelty of the underlying paper [1], is that the diffusion model is not acting on the pixel space but rather on a latent space. This means that this 'latent diffusion model' does not only learn the task at hand (image synthesis) but in parallel also a powerful lossy compression model via an outer auto encoder structure. Now, the number of weights (model size) can be reduced drastically as the inner neural network layers act on a lower dimensional latent space rather than a high dimensional pixel space. It's fascinating because it shows that deep learning at its core comes down to compression/decompression (encoding/decoding), with close relation to Shannon's Information Theory (e.g. source coding/channel coding/data processing inequality).

[1] https://arxiv.org/abs/2112.10752

Re:A Web UI for Stable Diffusion by jrm4 on Sunday September 11, 2022 @04:23PM (#32802946)
Oh, wow. Now that you mention how it's similar to lossy (if not the same as) compression it all makes a LOT of sense. This is great. I teach IT and I already do a bit on how lossy compression works, (e.g. hey, if you see a blue pixel and then another slightly darker one next to it, what's the NEXT likely to be?) and this is something of an extension of that.
Re:A Web UI for Stable Diffusion by rassibassi on Saturday September 10, 2022 @03:09AM (#32788539)
Correction: the auto encoder is pre-trained :)
Re:A Web UI for Stable Diffusion by visarga on Saturday September 10, 2022 @01:40AM (#32788114)
Then maybe we should remind about this 25,000:1 ratio when an artist complains about his copyrights being abused. The model doesn't have space to actually copy his works inside, it can only memorise the equivalent of a thumbnail from each input. A very small thumbnail, scaled down 150:1 per width and height (square root of 25000). That's like a grain of rice on the screen.
Re:A Web UI for Stable Diffusion by DrSiemer on Saturday September 10, 2022 @05:36AM (#32789155)
That's not how it works though. Instead of applying arbitrary content detail reduction, the model is an attempt to distill the core of what makes a particular artist (or phrase, face, object etc) unique.

When programming, it will often take a long time and a lot of code to get to a few final lines that do what you want. You cannot say the final result is a "thumbnail" of all previous efforts. Rather, it is the apotheosis of it.

Some artists spend decades developing a style that looks like a kid could do it as well. Still, there is something unique in there, that a trained eye will recognize. Converting that particular style to a formula and making that freely available is at least somewhat morally ambiguous.

Re:A Web UI for Stable Diffusion by holoduke on Saturday September 10, 2022 @06:41AM (#32789415)
It's the same as someone trying to mimic a style. Nothing wrong with that. Certainly not something you could get copyrights from.
Re:A Web UI for Stable Diffusion by DrSiemer on Saturday September 10, 2022 @08:16AM (#32789964)
It is not the same as trying to mimic a style. It is cloning the essence of a style and making it readily available to anyone who asks for it.

Sure, it's not copyright infringement, but you could argue that this takes away from the hardship the original artist had to go through to perfect their style.

Re:A Web UI for Stable Diffusion by namarie on Saturday September 10, 2022 @01:46PM (#32792829)
> Sure, it's not copyright infringement

Which was precisely what the discussion was about.

> But you could argue that this takes away from the hardship the original artist had to go through to perfect their style

You could argue the same things about photoshop, a lot of other digital tools, drum machines, the photograph and the phonograph.

Re:A Web UI for Stable Diffusion by jrm4 on Saturday September 10, 2022 @10:04AM (#32790745)
Ah, I can step in here.

Fair use might work but maybe not? If I were to argue against it, I'd probably compare something like a recording of music vs. a MIDI file. Same raw data scaling.