~Blue Lagoon II~ Another challenge putting 2 people together with specific poses. I have created a plugin to automate the bridging from Poser to Stable Diffusion, bypassing the OpenPose preprocessor, and this is the first result from the testing. The problem with the preprocessor is that it can misinterpret the input 3D render pose(s), so the plugin does that job straight from Poser. The part where the bodies intersect is still a problem with Stable Diffusion because of camera occlusion, and also because it's pretty bad with hands to start with. I know the holding hands are not great, but making a single hand is a challenge with AI generation, so imagine 2 hands interacting with each other. Maybe one day, but we are not there yet. 😋 For the ones in doubt, the woman is the one on the left. Poses and camera framing established in Poser, exported to SD using my plugin, rendered in Stable Diffusion, postwork in Paintshop.