Awesome.

Can you show a before adn after? so I ew see how much of a difference the img2img made?

I used to play with img2img 1 year ago but stopped.

Maybe it is time to do it again.

Above, before and after AI assist

damn the difference is massive.

Do you use just img2 img or you add soem strange controlnet to the stuff?

It is img2img. Controlnet doesnt like ponyxl right now. Hands especially are hard to have it get right. I also use the adetailer plugin

Its finnicky. Sometimes the ai makes a almost perfect conversion on the first attempt. Other times im spent render and rerendering again and again with no good results

>>1751
Please tell us how to use img to img.

First you need to download some stuff

Stable diffusion:https://github.com/AUTOMATIC1111/stable-diffusion-webui/releases/tag/v1.9.4

The the pony xl base model:
https://civitai.com/models/257749/pony-diffusion-v6-xl

also on civitai, do a search for any style or character LORA's you want.

Base model are placed here:
Stable Diffusion\stable-diffusion-webui\models\Stable-diffusion

LORA's go here:
Stable Diffusion\stable-diffusion-webui\models\Lora

A good style LORA is the modern anime screenshot
https://civitai.com/models/396772/modern-anime-screencap-style-pony-xl

>>1751
I've been using img2img recently as well. I'm a lot more of an armature when it comes to 3d modeling / koikatsu but I've had some decent results. This one isn't ryona or guro but it shows my results. I haven't tried guro img2img yet but it should work in the same way.

A bit tip I can give is to use inpainting with "only masked" instead of "whole picture". Regardless of the image size I set it to "only masked" and the resolution to 1024x1024. It will render out the masked section in 1024x1024 and then either downscale or upscale it back to fit exactly where it should be placed. Obviously you want it to downscale it, but it will work even if it is somehow upscaling it.

To give a rough run down of my workflow for img2img. I first do an img2img pass of the whole image with a low denoise, I use around 0.2~0.4 since ponyXL is aggressive. This is for the background, if I'm getting too many errors I place a mask excluding the characters. For the mask I re-render the same image with no background then use magic wand on photoshop to pickout the background to make the mask. I don't how this would work on ComfyUI but automatic1111 has mask uploading. I then go over each part of the character and inpaint them using "only masked". This can be done with either a low or high denoise. I typically section them off by face-body-any "complex behaviour". The danger of using "only masked" is that it can't understand the context of the whole image so skin tone for example may differ if you do the body in multiple "only masked" inpaints. If you need context I like to do a "whole picture" inpaint generation then go back to "only masked" with a low denoise just to fix up the details.

>>1789
*amature instead of armature.

I also do the hands in their own "inpaint" generation. Doing each section in their own inpaint generation may take longer but it means you're always moving closer to the desired result rather then just spamming the gacha. It may end up taking less time over all.

>>1751
Also just in case you don't know about it.

https://rentry.org/R-18G_Stable_Diffusion

This rentry contains a collection of guro Lora's including PonyXL lora's.

Yeah,ive mostly been spamming the gocha and hoping for the best. I'm the amateur when it comes to Stable diffusion

thanks for sharing
>>1761
my favorite pic itt so far

Working on a comic combining KK and AI

>>1971
Looks great so far. Rope on panel 2 of page 2 & the underwear on page 4 look a bit weird but should be easy to fix up with inpaint. Can't wait to see the finished comic.

>>1970
Amazing, I wonder if there will be a video of Asuka slowly fading on the noose.

So someone already made a lora for dave cheungs art style.(angel corps) and it works well with my img2img stuff

love that you are doing the angel corps. looking forward to see what you will accomplish with them

thanks for creating!

This is pretty ok, I feel like this technology is almost there.
It looks good, but still not perfect and it's distracting.
I like your stories!

Stunning work!! Will you take requests?

This is so good man.sooooo gooooood

So I don't take request as I just do this for funsies. And I'm not gonna be making a ton of full ai stuff most likely as that comic was a ton of work

Though I have found a good balance between Koikatsu rendering and using AI