You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
It would be nice if there was a seperate settings page where you could create IP adapter presets like how foooocus allows 4 images.
IP adapter inputs are mostly square so handling them as a layer seems a hassle and sometimes you want multiple images.
You could use clothes + face ip adapter in one preset for a specific character.
The text was updated successfully, but these errors were encountered:
Although they don't necessarily match the image dimensions, using layers as inputs has some advantages:
Can easily cut/crop the image to be square and show relevant content
Can easily use the actual image as input, this is quite useful for expand/outpaint, or just at low strength to be more style preserving
I don't think it's that much hassle to add an external image? I usually copy the image, press Ctrl+V in Krita (which inserts it as layer and makes it active), then click the add-control-layer button (which automatically uses the active layer). Not sure how to do it any faster tbh.
Regarding presets: I think this might fit into the general request to have inputs (prompts/control/etc) somehow attached to image areas or layers. Unfortunately diffusion isn't layer aware (and layer diffusion is like 50% there but missing img2img). Doing it manually via presets is also an option but more tedious than I'd like.
It would be nice if there was a seperate settings page where you could create IP adapter presets like how foooocus allows 4 images.
IP adapter inputs are mostly square so handling them as a layer seems a hassle and sometimes you want multiple images.
You could use clothes + face ip adapter in one preset for a specific character.
The text was updated successfully, but these errors were encountered: