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

161–170 of 233 posts

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

#161

Post by kyriakos »

goldenkey wrote: post #87
kyriakos wrote: post #72

Earlier quoted context omitted.

How do they get consistent looks? I mean if you are using it for a comic book your characters should look the same throughout the story. From my experience, the results between prompts are completely different even with minor variations.

If the seed remains the same, the difference between similar prompts is actually continuous. Here's a slew of images (1 through 5) I generated all from the same seed and same prompt sans a word or two: https://www.instagram.com/p/Chg60Fou6xB/

Thanks. Those are amazing by the way

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

#162

Post by KaoruAoiShiho »

lm28469 wrote: post #112
sheilaalien wrote: post #97

I'm surprised how very little this forum and the public in general knows or understands about the current capabilities and availability to these ai art generators. You have midjourney.com beta.dreamstudio.ai craiyon.com (real quick version no fuss, low quality) creator.nightcafe.studio and those are just some of the entry-level ones. I make ai art all day long! Check out my media feed on twitter @Sheilaaliens

> 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.

It's only sad if you think art or humans sacrificing their time has some kind of fundamental moral value rather than being just another type of product. As a materialist I don't really subscribe to that. I see work as a negative thing that society progresses to reduce.

The keypressing is clearly, objectively, not useless. It is useful, just less costly and more accessible to way more people. I'm generally happy with this.

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

#164

Post by sebringj »

This will be nice when a person wants a custom item tailor made to describe visually what it should look like and verify the picture before sending over. Taking that idea further into the future, we can clearly see how speech-to-text and dynamic adjustments coupled with 3d printing could lead.

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

#165

Post by dekhn »

gallabytes wrote: post #142
motoboi wrote: post #140

Earlier quoted context omitted.

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

... no it definitely wasn't. that's $50m. read the paper, they tell you how long it took on a v4-256, which you know the public rental price for.

Pretty much everybody who trains models trains tens to hundreds of models before their final run. But a better way to think about this is: Google spent money to develop the TPUv4, then paid money to have a bunch built, and they're sitting in data centers, either consuming electricity or not. Google has clearly decided the value of these results exceeds the amortized cost of running the TPU fleet (which is an enormous expense).

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

#166

Post by LelouBil »

vyrotek wrote: post #163

Is the next thing going to be image generation trained on your own photos? It seems like personalized generation would be a huge market. Ex. Show me walking on a beach. Show my dog wearing sunglasses, etc.

You might want to look at this :

https://textual-inversion.github.io/

It allows you to find a string that corresponds to some "thing" you have pictures of. Then you can use a text to image model containing a reference to your "thing".

Looks pretty amazing! However it needs 20gb of VRAM so I couldn't try it out.

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

#167

Post by mshenfield »

hackernewds wrote: post #39
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.

> Creating the real thing in Pixlr is it that good now? note that Pixlr was bought by Google

I like Pixlr, I use pixlr.com/e. It's free and web based and has always worked well for me.

Did that happen recently? It says in Pixlr's site they're owned by INMAGINE.

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

#168

Post by TigeriusKirk »

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…

I feel there's a strong argument to be made that these organizations should be required to release these models publicly. These are built on the works of the public at large, and the public should get the full benefit of them.

Whatever effort Google has put into building the model is infinitesimally small compared to the work of the creators they're harvesting.

I don't expect this to happen easily, if at all, but I'm strongly in favor of it, and would even support legislation to that effect.

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

#169

Post by gfodor »

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".

It feels like the wind has shifted because of Stability AI, and now companies doing this are going to look like those that have “call to get a demo” links on their homepage in a SaaS world.
Post reply on HN