~Supergirl~ by Ken1171_Designs
Close
Hover over top left image to zoom.
Click anywhere to exit.
Members remain the original copyright holder in all their materials here at Renderosity. Use of any of their material inconsistent with the terms and conditions set forth is prohibited and is considered an infringement of the copyrights of the respective holders unless specially stated otherwise.
Description
My homage to Kara, the Supergirl. Pose, body proportions, and camera framing established in Poser (DawnSE), rendered in Stable Diffusion, postwork in Paintshop.
Comments (3)
gavincas93
Sorry, I do not really get the point in putting a 3d model into Stable Diffusion when the program changes the graphics entirely. It's a waste of valuable time and effort. Don't really think it's worth the satisfaction.
Ken1171_Designs
Alright then, let me explain what the point is. This is a new workflow I have developed, so it makes sense people may not get it at first. The idea is to export 3D poses, which include body proportions, and camera framing to SD. Once loaded, we can then flesh out the character we want in AI, but now it will preserve all these 3 things we have created in Poser. These things are hard to control with AI alone.
ControlNet was created to help with it, but it uses AI to extract poses from images, which can often get it all wrong - especially hands. My plugin improves on THIS part in specific, extracting the poses from 3D models, which creates precise poses for body and hands. It replaces ControlNet's OpenPose AI preprocessor completely, by doing the job in Poser instead of AI. This provides an unprecedented level of posing control that AI alone cannot provide. For this, we only need a bare figure in Poser, since all we want is the pose. You may use morphs to change the body proportions, since the AI will use that too.
It is also possible to replicate an existing 3D character in AI, but that's a whole different process, and you don't need the plugin for that. Instead, what you do is to export the 3D render of the character into SD's Img2img, and render there with a low CFG value. I have done that in these 2 examples below:
https://www.renderosity.com/gallery/items/3124173/toonime-girl-iv https://www.renderosity.com/gallery/items/3123883/savages-ii
Hope this helps!
hborre
This is very interesting, something that may be worth exploring. Thanks.
Ken1171_Designs
You are very welcome! Watching the demo video from the product page can help understanding what the plugin does, and how, where I show all the workflow from Poser to SD. ^__^
gavincas93
no thanks anyway, I'd prefer the human approach in things, it doesn't feel right with me. Still nice try though.
Ken1171_Designs
That's alright, you don't have to like it. I am still creating my own 3D contents for both Poser and DS, and have been doing so since 2004. AI just expands what I can do with Poser - it doesn't replace me. It widens the horizons of my artistic expression to the next level, which led me to create this 3D-AI-2D workflow. AI is just another tool I have added to my arsenal. It's up to you what you want to use. Just remember you don't have to post negative comments on everything you dislike, unless you want to discourage other artists. ^___^