Oh wow! That looks great!
Hilarious! is this a free tool?
It works on a credits system. You have to sign up for an account and at that point you get 50 free credits - each credit gives you one “generation” - like you get to type in one sentence to tell it what to generate, and it will pop out 4 different images for each “generation.” The images are 1024x1024 - and there doesn’t seem to be any way to get it to output a bigger image.
Anyhow, after your initial 50 credits they automatically refill your account with 15 free credits each month. You can buy additional credits but I don’t know how much they cost. I believe it’s still in the beta stage and there are obviously some “issues” that need to be worked out! I’d be very hesitant to pay for anything at this point since the results are so iffy, but it’s fun to play with, and I was able to get it to output a few non-people things (flowers and landscapes) that look vaguely usable.
Anyhow, here’s the URL if you want to play with it: https://labs.openai.com/
Hahahaha! Trigger Warning: You are about to encounter Demon Spawn!!! 
I haven’t found a place to leave them feedback yet. One would hope they actually look at some of the stuff their system puts out so they can tweak the software!
Ostensibly it should improve on its own as it works more and more… I’m not sure, however how they are “rewarding” their algorithm for getting it right.
1 Like
Lol. Yeah, I’ve been doing a bunch of AI art as well. Faces are pretty funny until you’ve done a bunch of tinkering. And don’t get me started on hands - AI has **serious** difficulty not giving hands all sorts of extra fingers folded and bent into inhuman shapes. Ha ha.
Once you’ve got to an image with vaguely human facial features, try GFP-GAN, a separate AI that does face-fixing. I was using it quite a lot, as it also scales up 512x512px image using a (third) AI without the enlarged image being all pixelated. But for some reason, very recently every face I tried started to look similar; it was modifying the eye colors, nose and lips so everybody looked the same! Don’t know if that was a glitch, and if so, whether it has been fixed yet. Beside that, though, the faces do look fairly real, depending on the original image - sometimes you’ll still need GIMP (or Photoshop, etc) to mess with the eyes, as they can still come out gazing off in opposite directions!
I’ve just posted a sci-fi portrait on my Twitter account, showing the evolution of what I got from a few hours worth of AI, followed by another day or so tinkering, re-doing clothing, background and lighting. Here’s the tweet:
https://twitter.com/artsiren/status/1577284581086351360
And for a simple-to-use web version of GFP-GAN, try Baseten, here:
That sci fi woman looks good, even features not all horrifically distorted. Yay!
Completely independent of AI, I just tried that Baseten link on an old vacation photo from when I had a worse camera. It just what it does and you can’t tweak but it did a far better job of sharpening than Photoshop’s inbuilt sharpening tool. So thanks for the link.
1 Like
I tried Dall-E some time ago after stumbling on it somewhere. It was free, no sign-up, and very slow to render then. I believe the output was also restricted from commercial use then too. So it was just “mmm, that was interesting” and moved on. Since your post, I’ve been playing with this new iteration that is Dall-E 2 and it is absolutely fascinating not to mention addicting.
On the main Dalle-E page there is this render in the gallery:
This got me wondering about copyright on the dog & cat images “pulled” to make this copyright-free “composite”. Which led me to do a bunch of reading on how Dalle-E works (much of which is way over my head) and it turns out, images aren’t being “pulled”, they are genuinely being created from nothing, based on your text input. It’s not regurgitating & combining images “pulled” from somewhere, it’s making a wholly new image from what it’s learned a dog and a cat and white etc is.
The model’s been trained on hundreds of millions of image/text pairs and draws on that memory to render things.
The best layman’s explanation I ran across explained it like this -
If I said “Cat, draw a bird perched in a tree holding a flower” you’d draw what according to your memory and life-experience you think of as a bird, flower, tree … You didn’t copy anyone else’s imagery, you just drew from nothing what you interpret those things to be. That’s what Dalle-E does. Amazing huh?
Your drawing is going to be completely different than my or anyone else’s drawing. But if I refine the request to “draw a woodpecker perched in a pine tree holding a yellow rose” everyone’s drawings are going to be more similar though still different.
I think as designers & promoters who are constantly thinking about tags & keywords and how I would describe this vrs how customers would describe this, I think we can use that experience in prompting Dalle-E to get closer to what we want. We should be naturals at this. 😉 Results are still hit or miss but to me that’s part of the fun. It’s like hiring an artist to paint a mural. You can tell them what you want but the final result will still be a surprise due to the artist’s own perception.
Anyways, thank you for bringing this newer Dalle-E 2 to light. It’s just amazing technology. I think calling it software is like calling the Neuschwanstein Castle a lil’ shack in the woods. I can see buying lots of credits to experiment with this. (So maybe not thank you LOL).
4 Likes
You can use “outpainting” to expand your image infinitely. You can Google it to find some amazing examples. Girl with Pearl Earrings is most-cited example. I haven’t tried it myself as each section you add costs a credit and you could blow through them in a heartbeat trying to learn.
You can buy 115 credits for $15. Which is 460 images! My initial assumption was that you’d get charged 1 credit per image downloaded but nope, you get charged 1 credit per “generation” (four image versions) which works out to only $0.03 - $0.04 per image.
Very interesting! What’s “outpainting?”
Yeah, AI is pretty interesting stuff. They say that the more specific you are in telling it what you want the better, but honestly, I had some trouble getting it to do what I wanted especially in terms of where I wanted xyz image inside the picture. I was trying to make some mock-ups for wedding signs and it kept giving me signs that weren’t centered or facing forward like this:
I tried everything I could think of to get it to give me a sign facing forward and this is what I got…
Hmmm… Perhaps I just need to learn how it thinks so I can better tailor my requests.
The other thing that drove me batty was that I was trying to get it to output some flowers, but it kept cutting them off like this… Gah!
Well anyhow, I think that it will probably get better as it “learns” and hopefully I’ll get better at telling it what I want. It’s certainly fun to play with!
1 Like
Come for the weding. Stay for the fonwand font. It is comical and frustrating. But I think we’re seeing its’ first steps (it was only just opened to public in the summer) and it will get better with time. Like a baby, it will eventually learn by being exposed to many inputs.
1 Like
Hahahaha! Hopefully it will get better, and even as is, I’ve been able to get some useful stuff out of it. I had to do a LOT of hand editing to make this one usable, but the final product isn’t half bad!
2 Likes
Mock-ups! That’s actually the first thing I tried. An input of “a square white throw pillow on a modern black couch against a yellow wall” yielded these four images:
I was trying for ‘green screen’ colors that would be easy to replace in PS. The lower right image has potential, if I wanted to use more credits I could probably even add prompts for frames for artwork on the wall and such. But since I do very little active promoting, this was just a test for potential that I didn’t pursue further yet. But yeah, this AI rendering is a game-changer for sure and I think as time goes by there will be more and more sites with tutes on how to word things to get closer to the results you want. We’re all in the learning phase right now.
1 Like
Never mind - I just saw the outpainting section of the software. Still trying to figure out exactly how it works but I was able to turn this:
into this:
Promising!
5 Likes
I’ve still been having good luck getting it to give me woodcuts. I edit them quite a lot, but the final outcomes are nice.
Winter Chickadee on Holly Woodcut Winter Holidays
by Works_For_Peanuts
Woodcut Owl with a Gift Snowy Night Winter Holiday
by Works_For_Peanuts
Christmas Lights Merry and Bright Woodcut
by Works_For_Peanuts
6 Likes
Those are great!
My Daughter made this one do you all like it?
Beautiful winged Angel lovers in snow believe
by DoodlesGifts
6 Likes
AI artwork is AWESOME. I’ve been playing with it for about 6 weeks now at Night Cafe, which allows you to 1)earn credits, 2) use any images you create for commercial purposes, 3) allows you to add the link to an art print for sale - so those in the Night Cafe community that are scrolling through and “liking/commenting” on creations can buy a print direct from you. You asked how else these designs could be used? Here’s a collection of ornaments and stocking I created from my wildlife series
https://www.zazzle.com/collections/wild_animals_christmas_stockings_ornaments-119550289641065863
and here’s my NightCafe profile: https://creator.nightcafe.studio/u/AlteredUrbane
4 Likes
Just an FYI per an email from Dalle-E:
You own the generations you create with DALL·E.
We’ve simplified our Terms of Use and you now have full ownership rights to the images you create with DALL·E — in addition to the usage rights you’ve already had to use and monetize your creations however you’d like. This update is possible due to improvements to our safety systems which minimize the ability to generate content that violates our content policy.
3 Likes