lllyasviel
Openpose has no effect on any picture.
open pose doesn't work neither on automatic1111 nor comfyUI.
already used both the 700 pruned model and the kohya pruned model as well. neither has any influence on my model. Also I click enable and also added the anotation files. Open pose simply doesnt work. neither the open pose editor can generate a picture that works with the open pose control net. Control picture just appears totally or totally black.
control net has not effect on text2image only on img2img
I wanna know if controlnets are an img2img mode only.
update the webui extension again. did the same for me today, now after update it works again
silimar problem on my webUI. how to fix?
Canny work well. Openpose, not working at all.
There is NO output annotator pic when i generate with openpose preprocessor..
some of the controlnet modes will require additional downloads apart those 5Gb models you have downloaded manually. The first time you chose a model and you hit Generate it will download files that are up to 1Gb in size. Some tiems after an update it will download them again (or maybe some updated ones) Check the console if there is any error about not being able to access an url. This can happen if some antivirus blocks the connection
never mind, I think is broken. This is what I get by always updating, lol.
TypeError: ControlParams.init() takes 9 positional arguments but 10 were given
go figure waht could be
third time the charm. enabled and disabled the extension a couple of times, restarting the Ui, restarting the entire thing. now it works. crisis adverted
im experiencing the same problem, updated webui to the latest, updated controlnet (which promptly broke my webui and made it become stuck on 'installing requirements', but regardless...) and openpose ends up having 0 effect on img gens
Set preprocessor to none works when you don't need to calculate the input. For example if you already have the depth image you want to use you can skip the creation of this image by using a preprocessor.
Try using with another browser, in my case I spent time trying to get it to work on bravo, but it didn't work. I used chrome and openpose and others worked.
Try using with another browser, in my case I spent time trying to get it to work on bravo, but it didn't work. I used chrome and openpose and others worked.
Yeah. This is worked for me, I switched to chrome and it's working.
The pre-processor is acting as annotator, used to prepare the raw images. If you already have a pose image (RGB colored stick) then its already been annotated. you just need to set it to none if you are using it. But if you are using a real photograph, or some sketch drawing, use the preprocessor to annotate the input image, for example: manga or sketch images can be annotated using canny or scribble.
Just changed model from control_v11p_sd15_openpose.pth to control_sd15_openpose [fef5e48e]
Work for me.
