When editing a image, area selection, face swap, clothes swap etc. is elements that would make the job easier. Currently if I want an image generated I need to generate 16 times to get the results I want because models are junk at quality -unless you go other cloud providers which then why am I here- and this results with fucked up image. If we could select faces, clothes, item people are holding and tell remove/add/change where models freedom limited to area —like photoshop— then we could generate less stuff while getting higher quality at same compute power. This way lower level model complaints and no properly uncensored model compliants —which I do have as well as 5 on the feed I just scrolled a second— could be slightly midigated as we can now at least moderatly manage the areas AI can fuck up and cant fuck up, parts we like and hate.
Please authenticate to join the conversation.
New Submission
Feature Requests
Image Editing
2 days ago

user
Get notified by email when there are changes.
New Submission
Feature Requests
Image Editing
2 days ago

user
Get notified by email when there are changes.