You literally cannot respond with "you are holding it wrong" specially when I'm claiming that even for the popular _example prompts_ SD authors used they had to hand-pick the best random result over a sea of extremely shitty images.
And even in the original paper they disclaim it by saying "oh, our model is just bad at limbs". No, it's not just bad at limbs. They just happened to try examples where it could particularly show how terrible it is at limbs (i.e. spider legged horses and the like). But in truth, it's just bad at everything.
I'll repeat myself. You have to play around with the models and learn how to use it (just like you have to do for everything) .
> But in truth, it's just bad at everything
Thousands of people (including myself) have had the complete opposite result and have gotten amazing pictures. You can play around with the finetuning with different models from civitai and get completely different art styles too.
Like, this is so dumb I don't even know how to respond lol.
You're like some guy who got a computer for the first time and couldn't figure out how to open the web browser, so he just dismissed it as useless.
Most egregious if you are even doing more tuning and cherry picking that the authors of the models are doing, which you definitely are.
Drawing isn't hard, you know
SD will never compare to the power of pencil and paper imo. Drawing is an essential skill for any visual artist not just for mechanics but for developing style, taste, and true understanding of the world around you visually.
I recommend Freehand Figure Drawing for Illustrators as a good starting point (along with some beginner art lessons). It won't take 1k hours before you see results. It's also fun!
It works surprisingly well.
Why do I want to avoid that reliance, other than to be smug to the nerds? And as far as general self-satisfaction, let's assume I would rather master a different skill with the time it would take.
Especially because that training cost only has to be done once, so on a per-person basis it beats learning to draw by a lot.
If you said something about flexibility and specificity of what you can create I could get behind that, but I think the arguments you're making are very unconvincing.
If drawing was that easy, no one would worry about disruption from AI image generators, because everyone who wanted images would be knocking them out by hand, not paying people for them, so there’d be nothing to disrupt.
Therefore, I would say most people hyped for SD and friends are the capitalists and consumers of the art world - not the creators.
Repeating an assertion more times doesn’t make it true.
> If you want to create, you can and will learn to draw.
“Drawing”, even when it comes to imagery, is far from the only form of creation, and, even if it were, will is not determinative of capacity.
But, yeah, I mean, people have said similar things about basically evey new means of creation ever.
No, the fact that something is made with it makes it creation. Imagination makes it creative, on top of being creation.
The problem I see here is you are trying to insert your moral/aesthetic judgement of the quality/value of the mechanism/process of creation into the objective description of whether creation is happening, probably because you have subscribed to a worldview which valorizes creation and denigrates consumption so accepting something you don’t want valorized as “creation” is incompatible with that worldview.
You can just say you don't like Generative AI and wish people wouldn’t use it. You don’t need to declare that creating things with it isn’t creation to try to mask your (valid as any other) aesthetic preference in a very silly circumlocution designed to resemble an objective description, albeit a patently self-contradictory one.
Besides, there are models that are much more capable than Stable Diffusion. The best one currently seems to be Midjourney V5.
No, it wouldn't have – not to people in the know. We just didn't have powerful enough computers back in the 90s. Sure, the techniques we've got now are better, but 90s algorithms (with modern supercomputers) can get you most of the way.
Transformers are awesome, but they're not that much of a stretch from 90s technology. GANs are… ridiculously obvious, in hindsight, and people have been doing similar things since the dawn of AI; I imagine the people who came up with the idea were pretty confident of its capabilities even before they tested them.
Both these kinds of system – and neural-net-based systems in general – are based around mimicry. Their inability to draw limbs, or to tell the truth, or count, are fundamental to how they function, and iterative improvement isn't going to fix them.
I'm not talking about the 90s. I'm talking about April 29, 2020.
If you're talking about popular awareness of GPT-style autocomplete, then I agree. If you're talking about academic awareness of what these things can and can't do, we've had that for a while.
Three years ago there was nothing remotely as impressive as modern GPT style or text to image models. Basically nobody predicted what was about to happen. The only exception I know is Scott Alexander [1]. I don't know about any similar predictions from the experts, but I'm happy to be proven wrong.
[1] https://slatestarcodex.com/2019/02/19/gpt-2-as-step-toward-g...
Well, yes (that file was), but actually no. StyleGAN1's public release was February 2019, and it's capable of far more than just faces.
> Three years ago there was nothing remotely as impressive as modern GPT style
I predicted that! Albeit, not publicly. (My predictions claimed it would have certain limitations; I can show that those still exist in GPT-4, but nobody on Hacker News seems to understand when I try to communicate it.¹)
> or text to image models.
Artbreeder (then called Ganbreeder) existed in early 2020, and it didn't take me by surprise when it came out. It parameterises the output of the model by mapping sliders to regions of the latent space; quite an obvious thing to do, if you want to try getting fine-grained control over the output. (A 2015 paper built on this technique: https://arxiv.org/abs/1508.06576)
I was using spaCy back in 2017–2018-time. It represents sentences as vectors, that you can do stuff like cosine similarity on.
If I'd been more interested in the field back then, I could have put two and two together, and realised you could train a net on labelled images (with supervised learning) to map a spaCy model's space to StyleGAN's, which would be a text-to-image model. It was very much imaginable back before April of 2020; a wealthy non-researcher hobbyist could've made one, using off-the-shelf tools.
If I were better at literature searches, I could probably find you an example of someone who'd done that, or something like it!
---
¹: See e.g. here: they tell me GPT-4 can translate more than just explicitly-specified meaning, and the "evidence" doesn't even manage that. https://news.ycombinator.com/item?id=35530316 (They also think translating the title of a game is the same as translating the game, for some reason; that confusion was probably my fault.)
I don't know. I've had chatbots for decades before "a few years ago", so I have never been particularly impressed. I would say that for someone who was already impressed with that you could practically describe a landscape in plain old 2000s Google Images and get a result, SD feels like just an incremental improvement over it -- the ability to create very surreal-looking 'melanges', at the cost of it almost always generating non-sensical ones. And also add that Google Images is much easier to use than SD...
I do a lot of my SD with a fixed seed and 1-image batches, once you know the specific model you are using getting decent pictures isn't hard, and zeroing in on a specific vision is easier with a fixed seed. Once I am happy with it, I might do multiple images without a fixed seed using the final prompt to see if I get something better
If you are using a web interface that only use the base SD models and don’t allow negative prompts, yes, its harder (negative prompts and in particular good, model specific, negative embeddings are an SD superpower.)
I do agree that it's bad at following the prompt exactly. It will produce most of the things you mentioned in the prompt but not necessarily in the same fashion you asked for. I don't agree that produced images are mostly shitty, just visit that subreddit.
This doesn't really say anything, because it's just survivor bias. The entire purpose of that site is to show the successes. Most people get disasters every single day, they just don't upload them to the site. Even if I try the same prompt I will get a disaster image, as long as I don't use e.g., exactly the same random seed they happened to use. This is not even "prompt engineering". It's just outright playing with the dice.
I madeots of wallpapers in bulk recently with a good prompt and model. None of the images, all via random seeds, was shitty by any measure. Only images I ended up deleting had some usual Ai artifacts I couldn't stand.
I have thousands upon thousands of beautiful images, each one more inspiring than the other and I did nothing or very close to nothing. You are belittling this amazing tech so much it sounds like you are scared of it. What’s it to you?
Have you ever lived in a time where it was even remotely possible to go from “astronaut on horse” and get even 1 decent result, in seconds?
I can’t even.
> You need to agree to share your contact informations to access this model
All of this does feel like another scam where the tech is exaggerated and side hustles are actually the endgame.
Obviously you have to play around with them and figure out how to prompt & tune them.
Once you do that, you can get pretty amazing results.
I make do a bunch of social media graphics and stable diffusion + midjourney has been insanely useful.