Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

How do you use this? #3

Open
freshlesh3 opened this issue Jul 18, 2023 · 5 comments
Open

How do you use this? #3

freshlesh3 opened this issue Jul 18, 2023 · 5 comments

Comments

@freshlesh3
Copy link

I installed with comfyui manager and can't find openpose editor. Also can you explain how to use the patch.

@Duemellon
Copy link

Yes. Some details would be helpful please.

@andreszs
Copy link

andreszs commented Aug 17, 2023

Here's a guide on how to use Controlnet + Openpose in ComfyUI:
ComfyUI workflow sample with MultiAreaConditioning, Loras, Openpose and ControlNet

image

Replace the Load Image node with the OpenPose Editor node (right click workflow > Add Node > image > OpenPose Editor) and connect it to your ApplyControlNet image endpoint.

In this workflow openpose is applied twice as a fix for the MultiAreaConditioning plugin. Without this node you must apply it once for the entire image, simply connect the Apply ControlNet CONDITIONING to your KSampler positive directly, and forget about the MultiAreaConditioning node.

@shawn-j
Copy link

shawn-j commented Oct 1, 2023

is there any way to start with a reference image, load that into the openpose editor to tweak the result? that's how it works in A1111. it looks like with this one, you have to manually drag every point for every pose? seems like a bad way to do it?

@andreszs
Copy link

andreszs commented Oct 2, 2023

The OpenPose editor plugin is limited to manually dragging the points for every pose, just like you said.

Surely there are extensions to extract pose from existing images, however they may or may not work accurately or be compatible with Comfy. You could instead try using pose packs from civitai, or the predefined poses from posemy.art exported as skeleton PNG.

@shawn-j
Copy link

shawn-j commented Oct 2, 2023

That'a how openpose works in auto1111, you have the main openpose model which guesses the pose, then if it gets something wrong you can edit that in the editor.. so not sure what the point is of an addon where you drag every point manually, including fingers & faces?. Is ther a plan to update this one?. There could possibly be a way to save the image from openpose, then load that into the openpose editor?. Roundabout way but it might do it.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants