mrhan1993
Is there any easier way for masking image on inpaint (Like CLIPSeg or etc) ?
Hello there,
I am currently using an API to generate images with text programmatically. However, I am facing difficulty in beautifying the face in the generated images. Presently, my process involves building a text-to-image and then an image-to-image with impainting. But on the Fooocus UI, I have to manually select the face in the image, which is a bit cumbersome. Is there an easier way to automatically recognize the face and use it as a mask? This would enable me to send an Inpaint request on the API. I recall using something called "clipseg" or a similar tool on ComfyUI. Do you have any ideas or suggestions, @konieshadow and @mrhan1993?
So far, Fooocus-API is just an API export of the original functionality of the Fooocus project. Your idea is very good, but I don't have the ability to implement it yet. It is recommended that you submit your request on the Fooocus project. If the author implements it, I will update it as soon as possible. Sorry for that.
So far, Fooocus-API is just an API export of the original functionality of the Fooocus project. Your idea is very good, but I don't have the ability to implement it yet. It is recommended that you submit your request on the Fooocus project. If the author implements it, I will update it as soon as possible. Sorry for that.
https://github.com/lllyasviel/Fooocus/issues/1516 created 👍🏼