I've been a little bit obsessed with Machine learning text-to-image tech lately, and some of my recent results are finally good enough to post!
![cyberpunk cityscape with large red glowing sign 9_text -5 [Disco Diffusion v5.1 ViT-B32 ViT-B16 RN50] 505064725.png-upscaled.png](https://images.hive.blog/768x0/https://files.peakd.com/file/peakd-hive/ausbitbank/AJizD5WHydgTG1sedFexJkRDJk8yqvYN2MTrmJ34zV7DHEPYvZWX7HURzRQCE9e.jpg)
This is just a small collection, you can find all my renders being automagically posted to my discord (https://discord.gg/Z9QtqcFM) in the art-spam room. You'll also find pinned messages with everything you need to make your own.
Feel free to use anything I make, but credit would be nice :)
oh man... i am sure you've seen all the buzz around dall-e 2... miiiiind bloooown!!!
what kind of stack did you make yours with?
I know, that and Midjourney look amazing. I'm sticking with open source, usually a combination of latent diffusion + disco diffusion 5.1 with "Visions of Chaos".
I've checked the #artspam on discord... pretty funky stuff!
I wonder how you trained/tricked it into including the hive logo into all those renders because sometimes there's not even a hint to it in the keywords?! 🤔
I make dodgy init images like these first, with some opacity and rgb noise over the top, and find ways to justify it in the prompt like "massive red energy shield" etc
lol... ok... so it's actually a human-AI art collaboration! nice!!
You seem to really enjoy it, sir. Of course it is very fun and produces satisfaction for you. It's like playing around but yielding something in the end.
It will look like metaverse technology on one hand. I don't understand it at all.
Is this text-to-image machine learning technology only limited to generating images from text descriptions, sir?
Its pretty flexible, a lot of the images above I got by combining a basic sketch of what I want with a heap of text different prompts and fidding with settings.. Some people are using the same tech to make videos, there's heaps of possibilities with machine learning :)
Thank you
I will try to study it in my spare time. It will be fun. But, I'm afraid it will make me addicted, don't you think, sir?
As an artist, I have only one thing to say to this robot:
They terk err jerbs!
Those eyes in the second last one XD
AI makes some pretty cool art. And sounds like you're having way too much fun with it XD
Hey, this is amazing which software you're using?
a combination of latent diffusion + disco diffusion 5.1 with "Visions of Chaos"
Bro! This is legit!
!PIZZA
!PGM
BUY AND STAKE THE PGM TO SEND A LOT OF TOKENS!
The tokens that the command sends are: 0.1 PGM-0.1 LVL-2.5 BUDS-0.01 MOTA-0.05 DEC-15 SBT-1 STARBITS-0.00000001 BTC (SWWAP.BTC)
Discord
Support the curation account @ pgm-curator with a delegation 10 HP - 50 HP - 100 HP - 500 HP - 1000 HP
Get potential votes from @ pgm-curator by paying in PGM, here is a guide
I'm a bot, if you want a hand ask @ zottone444
PIZZA Holders sent $PIZZA tips in this post's comments:
@morenow(1/15) tipped @ausbitbank (x1)
thaddeusprime tipped ausbitbank (x1)
Learn more at https://hive.pizza.
Very cool! Which software did you use for that? Do you use the Hive logo or something like that as a base? Or just a text prompt? I use NightCafe, but if I use text prompt only then it returns bees and hives :)
I run disco diffusion 5.1 locally, and I have the same issues with text only prompts even if I use
bees:-9etc. I used modified versions of the logo with opacity and noise with a sketch of the scene layout as an image prompt + lots of experiments with text promptsYeah, I tried first with OpenAI and Google something but it was complicated (I don't know python) but find out that there are a lot of interfaces online for that, relatively easy to use, so I choose NightCafe. If you're interested, I have also some videos, also with AI created sound. Depressed Musician.
btw: not to advertise but I find the debate interesting, I haven't had any response so far, but I've been posting this and similar stuff for about three months now :)
Most interesting... 😀
AI art is actually getting pretty darn good these days, give it a couple years and it'll be spitting out what's in our minds better than we can...
This is amazing and will definitely take you up on the offer of using some of your artwork....and of course credit you !CTP
This is incredible.
!PIZZA
This is lovely... well done!
It's not my part, but I really enjoy it, it's a great work of art 👏
Hey @ausbitbank
Hoping to get in touch, trying to explore the teamaustralia community - and keep coming back to your account in my exploration. Seeing if this is still a group, and whether there is a discord community.
I'm at Port Macquarie in NSW, and wouldn't mind connecting with some others with green and gold blood.
Cheers, Tim
Gday, the new hive Australia discord is here https://discord.gg/dYJb6gKR
Thanks for sharing your beautiful images. I am going to check more on Discord.