Pregnant AI Artwork
|
AI Artwork Tutorials, Tips & Tricks (Mega-Thread)
- 35 Vote(s) - 3.69 Average
- 1
- 2
- 3
- 4
- 5
|
|
User 70749 () October 13, 2022, 8:22 am
|
October 13, 2022, 8:22 am by User 70749 ()
Re: Gwen, It is possible but for now cannot yet be done on demand. For specific characters to be generated using txt2img text prompts, they either need to be well represented in the training data used to produce the AI diffusion models, or someone will need to train a model themselves on the specific character which can be used in image generation. Training is not especially difficult to accomplish if you can run your own instance of Stable Diffusion you could learn to train a model. But it requires time and the appropriate hardware either your own or a leased cloud system. Maybe Gwen is more present in the Waifu diffusion or the NovelAI leak models but I do not have those.
What is easier for now is img2img, even more so when combined with other editing tools. Here is a lazy one, thanks for introducing me to this artist I like their style:
I am using Stable Diffusion on a service, I will discuss details more when I have my own setup.
|
Rextyd (91)
October 13, 2022, 9:37 am
|
Posts: 39
Joined: Jan 2022
Reputation: 91
October 13, 2022, 9:37 am by Rextyd (91)
Have you ever run out of credit in stabble diffusion
|
pregphil (3685)
Staff Member October 15, 2022, 2:55 pm
|
Staff Member
Posts: 764
Joined: Jan 2012
Reputation: 3685
October 15, 2022, 2:55 pm by pregphil (3685)
Hey everyone, @ Admin has split a bunch of posts into standalone threads and asked me to update this thread.
Moving forward, this will be the main area for AI tutorials, tips, etc. Please continue sharing whatever help you can offer, showcase experiments, anything that is helpful to others.
If for some reason you'd like your stuff returned to this thread, I can do that. We're recommending that you post collections to separate threads, whether they be theme mega-threads or creator standalone threads.
Hopefully this is a welcome change!
|
User 70749 () October 22, 2022, 1:03 pm
|
October 22, 2022, 1:03 pm by User 70749 ()
Interface - Fast Auto1111 - https://github.com/TheLastBen/fast-stable-diffusion
Model - SD 1.5
Prompt:Beautiful curvy cute young teen country girl age 19 standing in the shower, very pregnant belly, large cleavage, thick thighs, naked and seductive, bellybutton,
Negative prompt: deformed, lumpy, bizarre, weird, amputee, extra limbs, weird hands, plastic, huge hands, deformed hands, anime, cartoon, no bellybutton, multiple bellybuttons, grainy
Steps: 20, Sampler: Euler, CFG scale: 10, Seed: 222569307, Size: 512x512, Model hash: 81761151, Batch size: 4, Batch pos: 0
Selected from a batch of 512x512
Edited to remove the typical fingerstubs that occur, quick and dirty work with the clone stamp tool. then run through img2img with crop and resize:
Prompt: Beautiful curvy cute young teen country girl age 19 standing in the bathroom, very pregnant belly, large cleavage, thick thighs, bellybutton,
Negative prompt: deformed, lumpy, bizarre, weird, amputee, extra limbs, weird hands, plastic, huge hands, deformed hands, anime, cartoon, no bellybutton, multiple bellybuttons, grainy
Steps: 20, Sampler: Euler, CFG scale: 16, Seed: 1732918755, Size: 960x1024, Model hash: 81761151, Denoising strength: 0.2, Mask blur: 4
I don't know how that lingerie-jeans hybrid outfit is supposed to work but I'm not complaining.
By using the same seed, settings and prompts anyone can produce the same output at each step.
Liked by gingerlover (Sep 20, 2023), c1025018 (Mar 12, 2023), hardko (Oct 23, 2022), DeadmanSenji (Oct 22, 2022), JaxsonStark (Oct 22, 2022), OliviaCohen (Oct 22, 2022)
|
Porkchop (42752)
October 22, 2022, 1:30 pm
|
Posts: 4,864
Joined: Dec 2012
Reputation: 42752
October 22, 2022, 1:30 pm by Porkchop (42752)
Wow!! Impressive!!
I thought she was real for a moment!
Well done!
|
User 70749 () October 23, 2022, 12:55 am
|
October 23, 2022, 12:55 am by User 70749 ()
lexica.art was always a great site for exploring prompts and their outputs. Now they're doing generation as well without the need for credits. There's limits to what settings can be changed, results aren't polished and its txt2img only - but it's a great place to try out concepts before fine tuning them with better tools. It also has automatic upscale. It's notably better at artistic compositions rather than photographic content. If anyone wants to try out the technology but isn't sure where to start this is almost perfect for beginners.
Prompt: "Young female space marines, winning a heavy assault battle, large fit body, heavily pregnant, cleavage, big round bare belly, tight sexy costume armor, gorgeous cute makeup, formal hair, film still, award winning war documentary, cinematography 2020"
Guidance scale 10, size 640x640
more or less the same but with the near ubiquitous addition of "painting by greg rutkowski and artgerm"
Aside from that, many if not most AI art services offer a free trial which is usually generous enough to get a handle on things.
|
pregphil (3685)
Staff Member October 23, 2022, 3:06 am
|
Staff Member
Posts: 764
Joined: Jan 2012
Reputation: 3685
October 23, 2022, 3:06 am by pregphil (3685)
Excellent information @"AIEOU". To everyone else: this is a perfect example of what we're hoping for in this thread. Learning from one another, developing as artists, etc.
I might try something of my own!
|
User 27138 () October 23, 2022, 9:02 pm
|
Liked by Njameson (Mar 9, 2024), (Mar 24, 2023), (Jan 5, 2023)
|
User 70749 () October 24, 2022, 1:49 am
|
October 24, 2022, 1:49 am by User 70749 ()
Those don't look too bad, but some people are more sensitive to the uncanny valley than others.
The diffusion process relies in part on randomness so you're always going to be picking out the best of what gets generated. For everything you see shared there's a ton that get discarded for good reason.
What image size are you making? The diffusion models are trained at 512x512, the taller the output size the more likely it is that the subject will become a distorted monstrosity. The Automatic1111 implementation has a fix for that which basically automates the process I follow for higher resolution images - creating a smaller image then running it through img2img. It's great but it's not selective and will just run every iteration through the process, which can waste time.
There's also ways to redo just parts of an image and you can touch things minor flaws up in an editor before sending it back through img2img. For that approach it really depends how much work you want to put in.
Using an implementation that lets you use negative prompts helps. It doesn't fix everything and if overused becomes counterproductive, but it can reduce the probability of anomalies.
|
breydog (154)
October 24, 2022, 1:47 pm
|
Posts: 139
Joined: Aug 2016
Reputation: 154
October 24, 2022, 1:47 pm by breydog (154)
-__- Can't even get away from Ai "Art" in my porn.
if you want it that bad just learn to make morphs,
you'll get better results and its practiaclly the same thing since the ai is just taking from other pregnant pics and morphs and slotting them together.
why people think this is so great is beyond me.
|
|
Related Threads |
Author |
Replies |
Views |
Last Post |
Bing.com Generated Artwork |
BiBeBel |
774 |
210,914 |
May 2, 2024, 1:31 pm
Last Post: LynxWarlord65
|
JaxsonStark AI thread |
JaxsonStark |
23 |
14,966 |
April 29, 2024, 11:10 am
Last Post: KimmySmith
|
BBBenthusiast's Hyperpregnancy Ai thread |
BBBenthusiast |
47 |
34,147 |
March 27, 2024, 5:19 pm
Last Post: Njameson
|
Audio/Text AI Chatbot Thread |
anon4644 |
23 |
9,376 |
February 15, 2024, 12:14 am
Last Post: Timelord 2
|
Pinup artwork from Bing |
jeffjarrett5 |
20 |
6,456 |
January 21, 2024, 12:58 pm
Last Post: jeffjarrett5
|
Users browsing this thread: 1 Guest(s)
|
|