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

4.80 star(s) 6 Votes
Discussion Reviews (6)
FoolsBeGone

FoolsBeGone

Well-Known Member
May 21, 2023
1,042
3,134
Bix Lewd said:
Yesterday I tried inpainting a full range of facial expressions in A1111 without training for a specific character:

View attachment 2438032

I couldn't get decent expressions out of RealVision or URPM and I think that's because they were not trained on Danbooru images. The model I ended up using is called AllysMix3 (it's on civitai)

You don't have permission to view the spoiler content. Log in or register now.
Click to expand...
I don't know why but this kinda reminds me of Doc Oc from Into the Spiderverse :cool:
 
FoolsBeGone

FoolsBeGone

Well-Known Member
May 21, 2023
1,042
3,134
atheran said:
I've been trying on my end too. Removed all the bloody references that could even insinuate a minor, then thought the model might be the issue, changed to Clarity. I either don't like the model at all, or I'm doing something very wrong and get a very 3d like appearance, too plastic-y, earlier model looked a lot more realistic.

As for the issue at hand.. You tell me, but I don't think it's doing much.
This is the render without any post processing, so all prompts and setting should be visible, if you (or anyone else) have any ideas. I want to be done with it and create the character I have in mind. And no, it's not a minor. :D

View attachment 2438840

Though I have to admit, Clarity manages the [Hawaiian|Scandinavian] much better, as far as face is concerned. Too bad it falls behind on realism.
Click to expand...
Since I have used up all my reactions that I am allowed for the day I will just reply and add :love::love::love::love:
 
Mr-Fox

Mr-Fox

Well-Known Member
Jan 24, 2020
1,401
3,808
A special mentioning, this is based off the awesome prompt by the eminent Sharlotte. You can find the original in this post.
I have done some work on it myself also of course. Instead of his Patricia Ti, I used my own Kendra Lora. I also made some additions to the prompt and changed the settings to my liking.

00003-234793821.png
 
  • Like
  • Heart
Reactions: sharlotte, stefbou, HawkyDawky and 2 others
M

Microtom

Well-Known Member
Sep 5, 2017
1,179
4,434
Do you mean the analog diffusion checkpoint or reliberate? The last image I just made used reliberate. I got grain and style from using keywords like extremely grainy wes anderson movie, analog style, 80s cinematic movie still, anamorphic lens, 1970s. The original picture had plenty of grain, but I fixed the girl in img2img using the tile resample controlnet model with a max denoise of 1 to give her more details. This lost a lot of grain so I added some in photoshop.

For reference, this was the original image:
00112-3263874275.png
 
Last edited:
  • Like
Reactions: Whitekink, Kass69 and azrazr
gardenhideaway

gardenhideaway

Newbie
Jul 1, 2019
25
47
I've been playing around with the local stable diffusion web ui for the past day or so hoping it could be a fun muse for "real art" and writing, etc. Quickly realized I know jack shit about prompts and especially danbooru tags, so I've spent most of my time taking selfies and letting it generate tags and it's surprisingly great. Plus I found it's super fun to play with thirst trap selfies since most models sure seem geared for that.

lgm_output_522331787.png

But I really need to learn how to use indrawing to fix poses, limb placement, and especially to compose group scenes. I was a little surprised to see that good tutorials were harder to find since AI is trending so much. :/ I'm not even sure how to approach concepts like foreshortening without working around an existing photo. Also how to like... diversify it a little lol.
 
  • Like
  • Haha
Reactions: Whitekink, sharlotte, azrazr and 1 other person
You must log in or register to reply here.
Top Bottom
4.80 star(s) 6 Votes