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

Imagen: An AI system that creates photorealistic images from input text

imagen.research.google

Post reply on HN
Print view View original

131–140 of 233 posts

Re: Imagen: An AI system that creates photorealistic images from input text

#131

Post by IanCal »

lm28469 wrote: post #119
HPsquared wrote: post #116

Earlier quoted context omitted.

It's division of labour. The toolmaker has a different set of skills to the user of the tools.

The user, me included, has less and less knowledge and skills though. When your tool becomes a megacorp owned subscription based service you barely understand is it really a tool ? It doesn't produce art in the way a brush and paint produce art, you barely have any control on what it does, you just become an image filter, you press a few keys, look at the image for 5 seconds and decide if it triggers the right part o…

> you barely have any control on what it does,

You have quite a lot of control from style and colour to composition by feeding in initial images. You can do this iteratively, selecting parts you want to keep and others you want to adjust.

Photography hasn't killed art, yet you can describe it as "point at something and press a button". Sure you'll get something out of it, and it might be alright. But the great outputs take more work and care, just like with the air art now.

Re: Imagen: An AI system that creates photorealistic images from input text

#132

Post by webmaven »

glenneroo wrote: post #71
hackernewds wrote: post #40

Earlier quoted context omitted.

Is Lexica finding results previously computed? Or generating them? I could only work with very simple queries like "photo of a cat".

It's ~1.5 million entries inputted by users during the beta period on Discord.

There are a lot of prompts and results that aren't being included. Not sure what the criteria were.

Re: Imagen: An AI system that creates photorealistic images from input text

#133

Post by jpeter »

Roark66 wrote: post #74

Every single time I see an article about a new AI model that has a section called "societal impact" I know immediately they are not releasing the model, the training set, nothing... It seems to be the kind of bullshit statement that those companies put in place of "we paid $500k training this model and we're not giving it for free to anyone".

No post body was provided.

Re: Imagen: An AI system that creates photorealistic images from input text

#134

Post by culi »

tomthe wrote: post #69
culi wrote: post #15

Earlier quoted context omitted.

Not a threat to stock photos? That's exactly what they are. Look at these photos from the Midjourney Discord today: crystal dragon thing: https://cdn.discordapp.com/attachments/951197655021797436/10... https://cdn.discordapp.com/attachments/951197655021797436/10... https://cdn.discordapp.com/attachments/951197655021797436/10... davinci-style notebook of flying machines: https://cdn.discordapp.com/attachments/10080491…

They all look great! But the usual customers of stock photos are not looking for dragons or cavemen taking a group selfie. And all the images are not the result of a beginner trying their first prompt, they all took many tries to be generated.

I've used MidJourney and I really don't think it takes that many tries to get what you want. When you enter in a prompt you get 4 results and then you can either upscale one of them or you can variate on one of them and produce 4 variations of that particular one.

Here's a couple "first results" that I personally tried and you can judge for yourself:

"the government is putting violence in our water"

https://mj-gallery.com/87f5a54d-7d59-44d3-aab4-1dd3ef34902e/...

"cherry monkey"

https://mj-gallery.com/5d2e14ba-8ea1-4797-ab6b-4a6807cfffa8/...

"permaculture garden city"

https://mj-gallery.com/088c18c1-8e61-44da-b109-edfbd32967ac/...

"Acmella oleracea"

https://mj-gallery.com/9d1bc9f3-3cdc-44ec-8577-0791c69aa942/...

"mondrian banana cloud forest"

https://mj-gallery.com/25a8ef06-1a07-4a40-97e5-bebbd2ee925e/...

"lonely neon rainforest at night"

https://mj-gallery.com/8e7b73c0-f519-4727-bf16-30e0a42ab412/...

Obviously these prompts are a bit more artsy than stock photos are meant to be but the point is just to give you an idea of how it does on the first try. All of these took less than a minute to produce

Re: Imagen: An AI system that creates photorealistic images from input text

#135

Post by tgv »

LudwigNagasena wrote: post #126
tgv wrote: post #124

For all those who are practically demanding this be turned over to them, here's a quote from a recent article (about Telegram): Filing a charge is pointless, says Ezra. Since two years, she's being harassed on Telegram. It started when she was sixteen: photoshopped nudes with her snapchat account were circulated. They had taken selfies from her social media, and those of her family, and combined them with porn fragme…

> If you read that, and think all these tools should be released, you're part of the problem. Such comments make me wonder whether a social credit system is actually a good idea. Yes, it can be abused to deny my rights, but how can it be worse than being assumed to be a sexual predator by default?

Nobody assumes that. But if you give it away to literally everybody that asks for it, you'd be worse than naive to assume that nobody is going to abuse it.

But your comment makes it sound as if you'd rather give up your rights than not have access to this system. I don't think it's that interesting, is it?

Re: Imagen: An AI system that creates photorealistic images from input text

#136

Post by prophesi »

SXX wrote: post #84
Roark66 wrote: post #74

Every single time I see an article about a new AI model that has a section called "societal impact" I know immediately they are not releasing the model, the training set, nothing... It seems to be the kind of bullshit statement that those companies put in place of "we paid $500k training this model and we're not giving it for free to anyone".

Fortunately we have Stability.AI and they release their image generation model already. Hopefully they'll follow with other projects too. https://stability.ai/blog/stable-diffusion-public-release

There's also a fork that requires a lot less VRAM. I was able to get this working with an Nvidia GTX 1070.

https://github.com/basujindal/stable-diffusion

You'd clone the fork, then download Stability AI's checkpoint, sd-v1-4.ckpt, from https://huggingface.co/CompVis/stable-diffusion-v-1-4-origin...

Follow the instructions in the forked repo, and you should be good to go in a manner of minutes.

Re: Imagen: An AI system that creates photorealistic images from input text

#137

Post by spyder »

bottlepalm wrote: post #7

Smart for Dalle, Midjourney, and Stable Diffusion to capitalize on this quickly. It looks like the technology is being commoditized at rapid speed. I wonder what’s next.

Stability.ai and probably others too already working on video and audio models too and also I think I heard a service to train / finetune the model with your own dataset.

Re: Imagen: An AI system that creates photorealistic images from input text

#138

Post by janekm »

echelon wrote: post #57
mshenfield wrote: post #28

These tools are amazing for prototyping. I had an idea for a promotional poster, and seeing my idea just by writing it felt like magic. The generated image had too many artifacts to use, but gave me a guideline to follow when creating the real thing in Pixlr. AI content generation (text, image, source code, video, music) will be a huge boon for prototyping where applied judiciously.

Google hasn't released squat . Google's product is vaporware and we shouldn't afford them any airtime until they release something usable. They're just trying to butt in and get press off of the backs of the teams actually working in the open, and that's super lame. Release your model, Google, or stop bragging and talking over the others here. You're greedily sucking oxygen out of the conversation, and as a trillion…

Well… midjourney used stable diffusion (with an additional guidance model I believe, not just prompt engineering) for their beta model which they already closed down again… it’s back to their old far inferior model.

Re: Imagen: An AI system that creates photorealistic images from input text

#139

Post by vagabund »

IanCal wrote: post #129
lm28469 wrote: post #112

Earlier quoted context omitted.

> I make ai art all day long! I feel like this is the epitome of modern "content creation" Typing a few sentences into a software you barely understand, said software shits 15 jpegs out, 2 are good, "hey I make art". What a sad state of affair, tech is consuming everything and people are cheering, one more step on the path to being complete useless key pressers.

Photography is well known to have killed art. Press a button on a box you barely understand and the camera shits out a pic. "Hey I make art". What a sad state of affair, tech is consuming everything and people are cheering, one more step on the path to being complete useless key pressers.

Also true

Re: Imagen: An AI system that creates photorealistic images from input text

#140

Post by motoboi »

Roark66 wrote: post #74

Every single time I see an article about a new AI model that has a section called "societal impact" I know immediately they are not releasing the model, the training set, nothing... It seems to be the kind of bullshit statement that those companies put in place of "we paid $500k training this model and we're not giving it for free to anyone".

in the case of imagen, I suppose the cost is at least two orders of magnitude over 500k.
Post reply on HN