Instructions to use georgefen/Face-Landmark-ControlNet with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use georgefen/Face-Landmark-ControlNet with Diffusers:
pip install -U diffusers transformers accelerate
from diffusers import ControlNetModel, StableDiffusionControlNetPipeline controlnet = ControlNetModel.from_pretrained("georgefen/Face-Landmark-ControlNet") pipe = StableDiffusionControlNetPipeline.from_pretrained( "fill-in-base-model", controlnet=controlnet ) - Notebooks
- Google Colab
- Kaggle
What preprocessor should I choose?
#8
by Orihane - opened
Hi!
I downloaded the model file and used the blender project file provided by toyxyz to produce the landmark picture and achieved very good results on sd-webui.
But if I want to generate a landmark directly by uploading a picture with conrtlnet,just like in the demo.what preprocessor of conrtlnet should I choose? Or do I need to do some other preparations?
Sorry, English is not my native language, so I used Google Translate,hope to express clearly.
I use dlib to generate landmark, you can check my implementation here :https://huggingface.co/georgefen/Face-Landmark-ControlNet/blob/bd3c61a26e74e42e3f43759f4ed39ebdee0e9092/gradio_landmark2image.py#L33