Create your AI Cum Slut for Valentine’s Day 60% OFF Now
x

Fan Art AI Netorase Phone (Le Stag) - Fan Art Thread

player two

Member
Dec 4, 2022
135
149
nickohoma and others, idea just crossed my mind. I know some folks are toying around with AI video generation and such.

But, is it possible to include in a prompt for that each scene, to create multiple copies of it but each one a half degree off of the previous? With just a few or so of them, it should be possible to create a "3D" illusion by placing them in a back & forth loop.

It's a smidge different than full on animation as it's not prompted for character motion but a still shot from barely different angles (aimed at a fixed point of still scene).

Imagine psuedo-3D Tara's fit tits hanging there in the air with her legs spread OR Laura crowning Nathan with a load oozing out of her mouth. Those could be next level PNGs or GIFs here.
 

nickohoma

Member
Jan 11, 2022
299
1,862
It's an interesting idea, since the better SaaS AI video makers out there right now seem to all limit NSFW stuff. It would be pretty easy to do something like an idle animation just by running minor variations on the same seed and prompt. I added a blink frame to this one. If I got pickier I could probably get frames where the boobs bounce around less, I was just focusing on the face for this loop.

wiggly.gif

Panning around a character would be hard since you'd have trouble getting those small angle adjustments. But you could probably do wiggly animations of dripping fluids or facial expression changes, stuff like that.
 

player two

Member
Dec 4, 2022
135
149
...
Panning around a character would be hard since you'd have trouble getting those small angle adjustments. But you could probably do wiggly animations of dripping fluids or facial expression changes, stuff like that.
Nicely done with the first crack at it. I like it. ++

Kind of reminds me of the animation styles that drops frames for cool style and pseudo-3D effects like "Into the Spiderverse" and "TNMT: Mutant Mayhem".

On a side note, I've been watching a few YT on folks using Krita's AI interface to local (or remote) Comfy. Some artists are using models trained on their art style, while others don't. But all of them are basically using Krita for region and layer work -- almost like old school Kai Power Tools but 30 years later amped on generative AI.

Check out . With a good prompt, one could use Krita to make a few still frames by hand "animating" to guide the AI. Then stack them for the PNG/GIF loop. And all NSFW. ;D

I still need to sort out what my "beginners" workflow will be by trying out different tooling before I level up. XD. Gotta have goals.

Update: Having some "motion" images in the game could definitely give that OF/TikTok DM feel.
 
Last edited:

HoM11

Member
Feb 28, 2021
121
913
Omg Tara's tan lines are niiiiiice
"Oh, you like 'em? I was actually wondering if maybe I should get rid of them".

BeachEpisode (2).jpg

(surely, Le Stag will give us a couple of episodes where the gang goes to the beach/on vacation?)


"Hey, dude. Could you apply sunscreen on my back?"

BeachEpisode (4).jpg

BeachEpisode (6).jpg

BeachEpisode (3).jpg

BeachEpisode (12).jpg

BeachEpisode (7).jpg

BeachEpisode (13).jpg

BeachEpisode (1).jpg

"Hey, Laura, what you said earlier about... letting me watch you do it with MC. I... I would really like that".

BeachEpisode (8).jpg

BeachEpisode (10).jpg

BeachEpisode (11).jpg

BeachEpisode (9).jpg
 

nickohoma

Member
Jan 11, 2022
299
1,862
On a side note, I've been watching a few YT on folks using Krita's AI interface to local (or remote) Comfy. Some artists are using models trained on their art style, while others don't. But all of them are basically using Krita for region and layer work -- almost like old school Kai Power Tools but 30 years later amped on generative AI.

Check out . With a good prompt, one could use Krita to make a few still frames by hand "animating" to guide the AI. Then stack them for the PNG/GIF loop. And all NSFW. ;D

I still need to sort out what my "beginners" workflow will be by trying out different tooling before I level up. XD. Gotta have goals.
That looks like a great interface for inpainting, and so much better than what I have been doing. I have no knowledge of actual image editing software, so I have been generating images, opening them in MSPaint or Pinta, saving, and reopening them in Stable Diffusion.

Any kind of inpainting is amazing when you get the hang of it. If you can't get a pose right just clumsily erase and redraw an arm somewhere else, then run it back through the AI and (sometimes) it comes out perfect.

You can also mess with ControlNet and Depth Maps if you want fine pose control, but I've messed only with it a little bit. Basically though, use DAZ or some 3d modeling software to pose a couple untextured figures, export a depth map, then feed that into Stable Diffusion along with a prompt. And it basically paints the image as a skin over the depth maps. It works surprisingly well. I just usually stuck with pure txt2img, then refining.
 

djvl666

New Member
Aug 20, 2021
5
44
nickohoma and others, idea just crossed my mind. I know some folks are toying around with AI video generation and such.

But, is it possible to include in a prompt for that each scene, to create multiple copies of it but each one a half degree off of the previous? With just a few or so of them, it should be possible to create a "3D" illusion by placing them in a back & forth loop.

It's a smidge different than full on animation as it's not prompted for character motion but a still shot from barely different angles (aimed at a fixed point of still scene).

Imagine psuedo-3D Tara's fit tits hanging there in the air with her legs spread OR Laura crowning Nathan with a load oozing out of her mouth. Those could be next level PNGs or GIFs here.
I'm busy right now with a SDXL/PONY workflow for video2video in ComfyUI. The positive side; this was my third run. Not ready yet.
The downside: 16GB on a RX6800 isn't nearly enough. To compile this video, took about 1 hour.

loop-count.gif

This took about one hour and the eye and face improvement via sam and seg just started. Picture by picture. It just costs a lot of time. Images are easy, video is a whole other level. Especially PONY.

Strange. I can't attach the file. For now just the compressed version instead of the full gif. . File too large..

1Pass__00012-compressed.gif



I know you want to see other scenes. I'm busy with Laura, not quit there yet, but this is a full PONY video:


1Pass-Hands__00005_1-compressed.gif
 
Last edited:

player two

Member
Dec 4, 2022
135
149
Some folks might not like the imperfect, trippy AI animations. I do. (keep up the amazing work. I'm still trying to figure out all this AI art stuff too.)

They remind me of the "live impressionist environments" in the movie "What Dreams May Come" - which was so cool, surreal, and unseen before. (So much it received awards for best visual effect).

But then throw a bunch of NSFW in the mix to get something like a dreamy, erotic, non-scary/surreal body horror-like of the most basic instinct of life. Like if H.P. Lovecraft or one of the "New Weird" authors wrote a erotic movie short.

It's a reminder that your sex organ isn't between your legs, but between your ears. (it's only via your skin can your implode your brain. XD)