Artificial Intelligence created Art

[JuanMena]

You know me, and you know I don't give a fuck when I get curious about something.

So I found this site that creates art by words with AI, and these are some of the results.

I'm mostly interested in the color harmony.

 

[JuanMena]

GOD DAMNIT! ACCIDENTALLY PRESSED POST THREAD.

 

[JuanMena]

 

[SG854]

This is the stupidest thread evar

 

[JuanMena]

This is the stupidest thread evar

It's on the EOF. What did you expect?

 

[Mr_Reaper]

 

[Mr_Reaper]

 

[WiiMiiSwitch]

Why is it so ugly?

 

[Milenko]

It's like they're all almost something recognisable. Pretty cool!

 

[Deleted member 397813]

King dedede

 

[JuanMena]

Why is it so ugly?

It's just AI. I don't know how it works, but maybe it's searching for pics on the internet and merges them together into a fucking mess.
As I said, I'm mostly interested in the color harmony the AI used. It's gonna be handy in the future for me

 

[WiiMiiSwitch]

It's just AI. I don't know how it works, but maybe it's searching for pics on the internet and merges them together into a fucking mess.
As I said, I'm mostly interested in the color harmony the AI used. It's gonna be handy in the future for me

I think in the future (3-4 years) we might see full on art
AI art has existed since 2014 and it does continue to evolve
This site tells you other cool things ai could do like music, military, and other stuff
https://digdeeper.neocities.org/ghost/capitalism.html

 

[WiiMiiSwitch]

Why is it so ugly?

also this was sarcasm

 

[JuanMena]

I think in the future (3-4 years) we might see full on art
AI art has existed since 2014 and it does continue to evolve
This site tells you other cool things ai could do like music, military, and other stuff
https://digdeeper.neocities.org/ghost/capitalism.html

I like to think it will be like that. It's certainly a pretty interesting topic.

 

[CoolMe]

Interesting stuff.. I like trippy looking aesthetic, these look like they were drawn in another dimension or something.

 

[JuanMena]

Interesting stuff.. I like trippy looking aesthetic, these look like they were drawn in another dimension or something.

It's fun to mess around with.
You never know what you're gonna see. It adds some surprise factors if you look for certain things.
I was going to post the outcome of searching for:

Gunshot Wounds
Beheading
Towers
Murder

That's why the pics jumps from 001 to 070 and so on.

 

[SG854]

⠄⠄⣿⣿⣿⣿⠘⡿⢛⣿⣿⣿⣿⣿⣧⢻⣿⣿⠃⠸⣿⣿⣿⠄⠄⠄⠄⠄
⠄⠄⣿⣿⣿⣿⢀⠼⣛⣛⣭⢭⣟⣛⣛⣛⠿⠿⢆⡠⢿⣿⣿⠄⠄⠄⠄⠄
⠄⠄⠸⣿⣿⢣⢶⣟⣿⣖⣿⣷⣻⣮⡿⣽⣿⣻⣖⣶⣤⣭⡉⠄⠄⠄⠄⠄
⠄⠄⠄⢹⠣⣛⣣⣭⣭⣭⣁⡛⠻⢽⣿⣿⣿⣿⢻⣿⣿⣿⣽⡧⡄⠄⠄⠄
⠄⠄⠄⠄⣼⣿⣿⣿⣿⣿⣿⣿⣿⣶⣌⡛⢿⣽⢘⣿⣷⣿⡻⠏⣛⣀⠄⠄
⠄⠄⠄⣼⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣦⠙⡅⣿⠚⣡⣴⣿⣿⣿⡆⠄
⠄⠄⣰⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣷⠄⣱⣾⣿⣿⣿⣿⣿⣿⠄
⠄⢀⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⢸⣿⣿⣿⣿⣿⣿⣿⣿⠄
⠄⣸⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡿⠣⣿⣿⣿⣿⣿⣿⣿⣿⣿⠄
⠄⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⠿⠛⠑⣿⣮⣝⣛⠿⠿⣿⣿⣿⣿⠄
⢠⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣶⠄⠄⠄⠄⣿⣿⣿⣿⣿⣿⣿⣿⣿⡟⠄

 

[Vila_]

Did somebody say ai? lol

Looks like you are using DALLE with CLIP guidance to generate the images, they are both fine but not as good as the new models released this year, I would recommend using VQGAN, RUdall-e, or StyleGAN2 (will be posting more links and examples soon since I'm on my e-reader right now)

 

[JuanMena]

Did somebody say ai? lol

Looks like you are using DALLE with CLIP guidance to generate the images, they are both fine but not as good as the new models released this year, I would recommend using VQGAN or RUstyleGAN2 (will be posting more links and examples soon since I'm on my e-reader right now)

Oh wow! I want to try them both!

 

[Vila_]

Oh wow! I want to try them both!

---

VQGAN paper: https://compvis.github.io/taming-transformers/
Colab: https://colab.research.google.com/github/eps696/aphantasia/blob/master/CLIP_VQGAN.ipynb
Example:

"a dark fantasy world rendered in unreal engine"

"colorful nebula forest"

"Frosted mountains, captured with sony alpha camera, unreal engine, rtx"

"GBAtemp flag" these ai don't just generate one image, they generate multiple and refine the image over time, here is the progression:

---

Rudalle: https://rudalle.ru/en/
Colab: https://colab.research.google.com/drive/1wGE-046et27oHvNlBNPH07qrEQNE04PQ?usp=sharing (you will have to translate any text to russian first since it was trained in russian lol)
Examples:
"Flat Minimalist House"

---

The model below doesn't work the same way the others do, instead you train it with a dataset of a lot of similar images and it tries to generate similar ones... it's the model used in https://thispersondoesnotexist.com/

Stylegan2: https://developer.nvidia.com/blog/synthesizing-high-resolution-images-with-stylegan2/
Demos: https://thisxdoesnotexist.com/ and https://github.com/justinpinkney/awesome-pretrained-stylegan2
My personal favourite model: https://github.com/Vilagamer999/SneakGAN (totally not by me I swear)
Examples:

Generated on a dataset of cat images:

Generated on a dataset of self portraits:

Generated on a dataset of cars:

And finally the best one (dedicated to @WiiMiiSwitch):

---

Enjoy