AI Image Generation

pilou

Lexica but not training more than 1 image!
Big zoom (66) : https://imgbox.com/g/Og1Ey38iML

pilou

Just a prompt = @@@ with 3 images for colors...
Leonardo
Big Zoom: https://imgbox.com/g/a75q5Z5PO3

pilou

Funny the prompt for all theses images except 2 last was "3.14" or "@@@"
Leonardo

pilou

Leonardo training test colors
Big Zoom : https://imgbox.com/g/KUsTUjZCb6

pilou

Leonardo
Test of the "Prompt image" with the 3.14 of yesterday with the Lady and Man as prompts (just 2 for him)
I would have learned something: the max you can put as images is 5!
The max for classic training is 30!
But the top is 8-15 according to the FAQ!
It's already calmer in the results!

Big Zoom: https://imgbox.com/g/zgSUHiJXWb

pilou

Leonardo
Moi Object
Big Zoom https://imgbox.com/g/tlSoDLaaQW

pilou

Previous prompt was "colored city"

Same above with a found image on the Net
prompt "post appocalyptic devasted city"

Capacity to gives human details is astonished!
Big Zoom : https://imgbox.com/g/fE5aBSAsI7

Amazing that a simple move of a slider gives different results!

pilou

Playground 64 variations mechanical piece
Big Zoom: https://imgbox.com/g/evvq3jAXAd
image Moi3D object

pilou

Leonardo portrait Moebius
big zoom : https://imgbox.com/g/L2507v4ekO

pilou

Funny replace portrait Moebius by "portrait by Moebius"
changes more the gender of the result!
Big zoom : https://imgbox.com/g/49lM7tdnt1

pilou

Wombo : 2 New Styles - Horror Cut v2.0 - Horror Cut v2.1
furturistic architecture at mountain

pilou

Mini tutorial for Leonardo AI
Put an image of a 3D object (or any image for which you want to see a "textured" projection of its prompt) in Image to Image - choose your favorite "Finetuned model" - a full-on Control Net (Depth to Image)!

Exemples big zoom...

pilou

Wombo : style Horror Cut v2.0, v2.1
futuristic architecture at mountain

Big Zoom : https://imgbox.com/g/zNKeH8KV5w

pilou

Not exactly Ai...
Funny this one! c:! and you can mouve your mouse during the process!
Reaction-Diffusion Explorer Tool
Interactive web application for creating reaction-diffusion textures and patterns!

Reaction-Diffusion Explorer Tool

Interactive web application for creating reaction-diffusion textures and patterns

(karlsims.com)

pilou

Inside Leonardo : textures to the infinite!

pilou

Bing: monobike
big Zoom : https://imgbox.com/g/20mFvitmWP

pilou

Leonardo Gaz engine
Big Zoom : https://imgbox.com/g/AtXk1XLras

pilou

Mini tut
(here I use Flowscape, which costs less than a pack of cigarettes at the parent company)

The image on the left is done in 30 seconds, and Leonard's ready to rock!
He knows how to put snow on the roof(s), electricity!
And the prompt is "winter landscape"!

the image on the left 1 second since it's automatic!
Of course, you can rotate it in any direction, add trees etc...Leonardo interprets it to his liking!