AI [AI Art] - Show Us Your AI Skill *NO TEENS*

4.80 star(s) 6 Votes
Discussion Reviews (6)
Mr-Fox

Mr-Fox

Well-Known Member
Jan 24, 2020
1,401
3,808
DreamingAway said:
Stable Diffusion formally unveiled August 22, 2022

Since then the area of AI technology has been swarmed by corporations & open source communities.
We aren't even a year into this ride and this is already what's happening.

What are we going to be posting in this thread in 2024? :eek:
Click to expand...
There will be a 3d printer that we use to print our ai companions or minions... But one day a rich asshole, because it's always a rich asshole.. He prints too many and they band together and starts the apocalypse, key the walking dead music... :LOL:

1678120869107.png
 
  • Like
  • Wow
Reactions: Sepheyer and DreamingAway
Bix Lewd

Bix Lewd

Newbie
Aug 10, 2016
69
146
DreamingAway said:
What is "Masked" is that just seeding the denoiser in Img2Img or a Control Net vector?
Click to expand...
I had to manually mask the background out in Photoshop. The Select > Subject command did most of the work though.

I was floored by the consistency, these results imply that we'll be able to make longer frame animations or even entire character spritesheets for games based on an arbitrary character prompt!

I'm going to see if I can make a Thicc Rogue mod for Oni's Rogue-Like evolution
 
DreamingAway

DreamingAway

Member
Aug 24, 2022
317
746
Bix Lewd said:
I had to manually mask the background out in Photoshop. The Select > Subject command did most of the work though.

I was floored by the consistency, these results imply that we'll be able to make longer frame animations or even entire character spritesheets for games based on an arbitrary character prompt!

I'm going to see if I can make a Thicc Rogue mod for Oni's Rogue-Like evolution
Click to expand...
But how are you passing that image into your generation?
Is there a "Mask" option in control net or your using that for image to image?
 
Bix Lewd

Bix Lewd

Newbie
Aug 10, 2016
69
146
DreamingAway said:
But how are you passing that image into your generation?
Is there a "Mask" option in control net or your using that for image to image?
Click to expand...
Sorry for being unclear, I throw the 'masked' image into img2img so I can extract the depth and pose maps from it. Then, I use controlNet in txt2img and feed it those depth and pose maps in order to generate the new art
 
  • Like
Reactions: VanMortis, Mr-Fox and DreamingAway
DreamingAway

DreamingAway

Member
Aug 24, 2022
317
746
Bix Lewd said:
Sorry for being unclear, I throw the 'masked' image into img2img so I can extract the depth and pose maps from it. Then, I use controlNet in txt2img and feed it those depth and pose maps in order to generate the new art
Click to expand...

Lovely ~

Thanks for sharing your workflow!
 
  • Like
Reactions: Bix Lewd
Jimwalrus

Jimwalrus

Well-Known Member
Sep 15, 2021
1,102
4,189
Jimwalrus said:
V1.0 of a
You must be registered to see the links
Textual Inversion embedding. View attachment 2447341
Click to expand...
Updated to
You must be registered to see the links
- trained for longer on a larger number of both full-body and face images. 00003-2967019595.png
One odd thing (and probably better asked / answered in the Prompt Sharing & Learning thread) is that, despite almost all of the full body training images being NSFW, SD really wants to draw clothes on her! Takes strong prompts to generate NSFW images. :unsure:
 
  • Like
Reactions: Slaver0, VanMortis, HawkyDawky and 3 others
You must log in or register to reply here.
Top Bottom
4.80 star(s) 6 Votes