A quick implementation of hand refining with Microsoft's Meshgraphormer for InvokeAI.
Go to your nodes
folder in your root
directory and clone this repo
git clone https://github.com/blessedcoolant/invoke_meshgraphormer.git
You might have to manually install a couple of extra dependencies in your Invoke venv
.
pip install trimesh rtree yacs
The required models will be automatically downloaded by Invoke's Model Download API.
This extension consists of one node - Hand Depth w/ MeshGraphormer
.
- Image:
PIL.Image
- Any image with hands in it - Resolution:
int
- The size of the depth map to be generated - Mask Padding:
int
- The size of the padding that is to be applied around the hand region when generating the mask. - Offload:
bool
- Offload the model after generation
- Image:
PIL.Image
- A depth map with fixed hand shapes - Mask:
PIL.Image
- A bounding box mask around the areas of the hand - Width:
int
- Width of the output image - Height:
int
- Height of the output image
This extension consists of one node - Hand Depth w/ MeshGraphormer
. This node takes in an image and outputs a hand depth map and a mask for the hand area. Use this depth map with a depth ControlNet model and you can use the mask to create a Denoise Mask
for your hand-fix pass so only the hand are denoised.
Example workflow is provided in the workflow
folder.
Feel free to fix anything or improve it anyway and make a PR. Thank you!