πͺ ControlNet modalities
ControlNet provides a greater degree of control over text-to-image generation by conditioning the model on additional inputs such as edge maps, depth maps, segmentation maps, and keypoints for pose detection.
With Scenario, you can use those modalities: canny,
depth,
illusion,
lineart,
lines,
normal-map,
pose,
scribble,
and seg
. Find below a quick description and examples of those modalities.
Modality | Description | Control image | Generated Image |
---|---|---|---|
canny | A monochrome image with white edges on a black background. | ||
depth | A grayscale image with black representing deep areas and white representing shallow areas. | ||
illusion | A range of grey shades from white to black. Can be used for ControlNet illusion | ||
lineart | A monochrome image with white soft edges on a black background. | ||
lines | A monochrome image composed only of white straight lines on a black background. | ||
normal-map | A normal mapped image. | ||
pose | A OpenPose bone image. | ||
scribble | A hand-drawn monochrome image with white outlines on a black background. | ||
seg | An ADE20K's segmentation protocol image. |
Sources: https://huggingface.co/lllyasviel/sd-controlnet-canny
Updated 8 months ago