Search

(Supports wildcard *)
Statistics
  • Id: 182726
  • Posted: 2023-10-06 21:55:18
    by Fungip
  • Size: 1568x1391
  • Rating: Explicit
  • Score: 164 (vote up)
Options
History
Related Posts
afro ai_art altered_common_sense altered_perception before_and_after bimbofication blue_eyes breasts caption collar earrings femsub fungip_(generator) hand_on_hip hat huge_breasts manip navel navel_piercing original piercing police_uniform policewoman purple_lipstick stripper text thick_thighs thong unaware

Edit | Respond

8 comments (0 hidden)

Argonis
>> #522421
Posted on 2023-10-07 03:09:41
Score: 2 (vote Up)
I wonder if that gang recruiting.
lickitup
>> #522429
Posted on 2023-10-07 03:52:38
Score: 2 (vote Up)
As someone who isnt into Bimbofication, this is an exception with how good she looks after and before.
HypnoHammer
>> #522438
Posted on 2023-10-07 06:09:52
Score: 0 (vote Up)
Can we get an artist here?
Argonis
>> #522440
Posted on 2023-10-07 06:54:00
Score: 0 (vote Up)
I think Fungip IS the artist.
Mindwipe
>> #522441
Posted on 2023-10-07 06:59:26
Score: 1 (vote Up)
The artist is AI.
Fungip
>> #522464
Posted on 2023-10-07 12:26:37
Score: 0 (vote Up)
The art was made with AI, the editing and captionning is from me.
UnderBridge
>> #522502
Posted on 2023-10-07 22:21:11
Score: 0 (vote Up)
Super impressive for AI art. What tool did you use, and how'd you manage the girl to look so similar in the before and after images while still having her whole outfit/hairstyle change?
Fungip
>> #522528
Posted on 2023-10-08 06:23:13
Score: 1 (vote Up)
I use the Stable Diffusion UI, with a variation of models and loras. After I generate a satisfying first image, I reuse the seed from that first generation onto the second one, and change some of the prompts, while keeping enough of the original prompts to maintain a sense of consistency between the two images. I generate a bunch of images and pick the one that's closest to what I'm looking for. In this case the hairstyle change was a happy accident. It wasn't what I intended, but since it fits well with the context of the story, I kept it. When there are small things I wish to change, like a clothe color, the position of an arm, a specific accessory, etc...I make the rough modifications on Photoshop, then run the image with the AI once more (on img2img) to smoothen everything up.
This one here went pretty smoothly. Sometimes it takes a lot of retries, and jumping back and forth between Photoshop and the AI. At the end there are always some minor glitches I have to fix (within the limits of what I can do). Things like weird colored spots or bad hands. I also do the blank eyes of the second image on Phososhop.
1