Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Stable Attribution

stableattribution.com

Post reply on HN
Print view View original

291–300 of 365 posts

Re: Stable Attribution

#291

Post by Lockal »

goldemerald wrote: post #8

This is a great website, but not in the way the authors intended. Based on some of the examples they explicitly provided, it is clear to me Stable Diffusion creates novel art. Here's a random example https://www.stableattribution.com/?image=a2666aee-0a1a-411b-... I will admit this is a nice tool for verifying the creations of SD aren't pure copies, so I think it will be useful for a time. But as AI-generated images s…

> Based on some of the examples they explicitly provided, it is clear to me Stable Diffusion creates novel art

Fiasco of a tool to detect the source of plagiarism does not mean that Stable Diffusion creates novel art. When you need a character with dyed hair tips, SD will occasionally spit out copyrighted Harley Quinn portrait from DC Comics, Stable Attribution won't mention Paul Dini and Bruce Timm as character authors, but tool may show you some other pictures of people with colored head. Good luck with lawyers, what I can say.

Re: Stable Attribution

#292

Post by makomk »

space_fountain wrote: post #243
orangeoxidation wrote: post #228

Earlier quoted context omitted.

Interesting. So it only shows images that were transformed / mixed to get the output, but does not show images used to learn how to transform / select them? Sounds very much like a human would do it. If I 'know' how to recognize saurik and I know how anime is supposed to look like, I can check my digital photo library for a picture of saurik and than use that picture as a template to draw an anime version of saurik.…

I think saying exactly what an ai of sufficient complexity is doing is a mater for philosophy more than science. But idk if transforming or mixing is how I’d describe what this is doing. In particular it truly does not have a complete representation of any of the images it was trained on. They just wouldn’t fit. It does of course have an understanding of how embeddings relate to images that is informed by the images…

The author of this tool is even aware of that argument and just dismisses it with no real justification: https://twitter.com/atroyn/status/1622360994579357696

Re: Stable Attribution

#293

Post by irthomasthomas »

saurik wrote: post #35

I gave it a photo I had Stable Diffusion 1.4 generate from the prompt "avatar for saurik". If you dig through the CLIP database, you will find that the model was trained on a ridiculously large number of copies of my Twitter profile photo due to it being included when people screenshot popular tweets I've posted (which, notably, also means that it is rather low resolution). https://www.stableattribution.com/?image=e8…

Perhaps it's confused because the real you looks exactly like Leonidas Kavakos! http://thelistenersclub.timothyjuddviolin.com/wp-content/upl...

Re: Stable Attribution

#294

Post by jaimex2 »

Everything is a remix.

I don't know what it's trying to achieve other than to waste everyone's time.

Those human artists saw art from other artists who saw art from other artists who saw art from other artists etc.

Re: Stable Attribution

#295

Post by throwaway1851 »

saurik wrote: post #104
codetrotter wrote: post #82

Earlier quoted context omitted.

Am I seeing the same thing you are in those two images you linked to? The first one, generated by AI has you looking at the camera. The other one has you looking at an instrument. They don’t look much like each other to me, in terms of pose or anything. The first images suggested by Stable Attribution looks a lot more like the AI image to me, in terms of pose and everything.

The slight adjustment of where I'm looking is minor. The instrument is uncommon and honestly difficult to get stable diffusion to generate correctly at all; though, from my experience playing with this (I spent a lot of time trying to figure out why it knew who I was before discovering the CLIP database browser), I'm going to argue that the reason that hand is showing up in the position it is is because of my hand ho…

> Clearly seeing at least one photo of me (and AFAIK it was only trained on thousands of copies of that single photo) was absolutely crucial to the construction of this image, and yet this website isn't finding any

I think there are two separate ideas that are being conflated here. The first idea is that there is a mapping between text input and a joint text/image embedding space. For that mapping, yes, your profile photo (paired with some caption text that says saurik) would have to be the most important training input because, as you say, how else does SD know what “saurik” looks like?

But the second idea is the mapping from the shared latent space to an image output. It is NOT necessarily true that your profile photo is the most significant training example for the generated image. That’s because the text “saurik” can map to a neighborhood in the latent space that has all the photos this tool retrieved — photos which are indeed very similar to the generated image.

As an example, let’s say that I resemble Ben Affleck. And you type in “throwaway1851” and you get back an image that does look like me, but also looks an awful lot like Ben Affleck. In this case I really wouldn’t be surprised if my one profile photo didn’t contribute much to the generated image as compared to a dozen different photos of Ben Affleck. Perhaps the original mapping just pointed to Affleck anyway, because I wasn’t significant enough to end up taking space in the model’s parameters.

Re: Stable Attribution

#296

Post by l33tman »

romwell wrote: post #182
chii wrote: post #170

Earlier quoted context omitted.

> This isn't weavers or telegraph operators, this is coming for all human art forms. so somehow the arts is above the telegraph operators or weavers? > but people will happily pay 2 cents to hear GPTx's paraphrase of it. except if you can tell that the novel GPTx "paraphrased" is a derivative work from yours, rather than transformative, you can either sue for loyalties, or take action. If you cannot tell if GPTx is s…

>except if you can tell that the novel GPTx "paraphrased" is a derivative work from yours, rather than transformative, you can either sue for loyalties, or take action. Respectfully, no. The top comment in this thread is saurik asking Stable Diffisuion to generate "an avatar of saurik" and getting a veritable likeness of themselves back. It would be laughable to think that this is feasible without the model having la…

"If it does, then that data was indispensable for generating the output for that prompt."

It gets murkier than this, by far. The diffusion models are not just trained on images, but on text. It varies by implementation but Stable Diffusion for example used a pre-trained CLIP transformer network from OpenAI (and subsequently OpenCLIP). CLIP can have internal associations between words that in turn steer the diffusion image generation.

To given a simple example of how this could work (I'm not saying this particular example does work though): the model in total could understand that a "swedish flag" consists of a "yellow cross on a blue background", and it could have yellow crosses in its image training set, but no images of Swedish flags, and it would still understand how to draw a "Swedish flag" based on the language semantics.

Just as a human, actually.

So, this won't work. A style can be inferred and described and it doesn't necessarily need to be in the image training set at all.

Re: Stable Attribution

#297

Post by throwaway1851 »

XorNot wrote: post #253
ertian wrote: post #47

Earlier quoted context omitted.

Yeah, it seems like it'll just as happily 'attribute' human-made images as well, which calls the whole thing into question. If it's really showing the images that stable diffusion has 'stolen from', and it'll do the same for humans, does that not mean people are equally guilty?

Yep straight up just uploaded a picture from my camera phone and it confidently found a bunch of strikingly similar images.

That in itself isn’t an indictment of the website. The tool is built on the assumption of being provided an image generated from Stable Diffusion. If you violate that assumption, it isn’t surprising that the tool fails your “test.”

By comparison, a “dog vs. cat” classifier that has 100% accuracy on the dog/cat task will, nonetheless, tell you that a slice of pizza is a dog... or a cat.

(You could possibly interpret the results as “if SD had generated this image, it would have drawn most from these sources in doing so”.)

Re: Stable Attribution

#298

Post by HenriTEL »

> people who saw it knew who made it

Yeah for a small subset of insiders to a small subset of creators. Really that was already a fiction prior to stable diffusion.

And then why human-made artwork would not need attribution of prior work that influenced it but AI made work would require some. We could extend the right to cite there and if we can't make it obvious that more than 10% of a given image was from another one, attribution makes no sense.

Re: Stable Attribution

#299

Post by williamcotton »

makomk wrote: post #292
space_fountain wrote: post #243

Earlier quoted context omitted.

I think saying exactly what an ai of sufficient complexity is doing is a mater for philosophy more than science. But idk if transforming or mixing is how I’d describe what this is doing. In particular it truly does not have a complete representation of any of the images it was trained on. They just wouldn’t fit. It does of course have an understanding of how embeddings relate to images that is informed by the images…

The author of this tool is even aware of that argument and just dismisses it with no real justification: https://twitter.com/atroyn/status/1622360994579357696

> historically, creatives have been among the first to embrace new technologies. ever since the renaissance, artists have picked up every new tool as it's become available, and used it to make great things.

> these people aren't 'luddites'

This is just total bullshit. I know plenty of artists who are embracing this technology to make all sorts of things that tools like SD were not designed to do, like psychedelic music videos, etc.

What the author means is that a few loud blue check marks on Twitter who claim to be artists have been tweeting, get ready for it, inflammatory claims.

Re: Stable Attribution

#300

Post by pk-protect-ai »

Utter bullshit. Even image of Pele on a main page of the site does not have any actual style dependencies from the images proposed by the site.

I have uploaded my image made by SD: https://ibb.co/3NxPdNw The list of proposed sources seems to be very random.

EDIT: https://www.stableattribution.com/?image=541edf3c-6281-4177-...

Bullshit.

People learning every day to draw they are learning from other images. They using others styles and some time they are coming up with a unique style which was never used before. SD does the same. You can't recover original images. You can have similar style or composition, but you will never recover original image.

Post reply on HN