MOSS Color & Depth Controlnet | 颜色深度控制 v1.0

MOSS Image LORA

This is a Model Patch and goes in your model patch folder!

A ControlNet for MOSS Image. The picture follows a depth map (pose, shape, front and back), a color region map (where each thing goes), or both.
Turn any photo into a pose for a new subject: the skater becomes an astronaut, the farmer a robot, parka girl a bearded man in a red raincoat. Same pose, same framing, everything else new.
One file. No separate LoRA: the adapter's input layer and its LoRA live inside it.

This model is darask0's work, not mine. Original upload, training code and examples [Here](https://huggingface.co/darask0/MOSS-Image-color-depth-control)
License: Apache-2.0 (same as MOSS Image). Re-shared here unchanged so CivitAI users can find it; only the filename is different moss-control.safetensors on Hugging Face).
Unofficial: a community adapter, not made or endorsed by OpenMOSS. Its training depth maps came from Depth Anything V2 Large (CC-BY-NC-4.0); the file is the adapter only.

This represents many hours of work. If you enjoy it, please 👍 like, 💬 comment, and feel free to ⚡ tip 😉