Skip to content

Cannot use ip-adapter-faceid-plusv2_sd15 ? #2679

Answered by pflky
sonewtogithub asked this question in Q&A
Discussion options

You must be logged in to vote

It takes a lot of tweaking to get good results, both on the lora weight and the controlnet weight, and also depends on the model you're using, the settings used, the face image itself, etc. Photomaker v2 is better, but it's a third party implementation that doesn't allow much customization or custom models. I can't recall but there may also be a trigger word required for the lora. It never was a super great method either way though. Sometimes it worked, sometimes it didn't, and was heavily dependent on the input image and tweaked settings and prompt. Sometimes even low resolution bad quality images of faces worked better than high resolution ones.

Replies: 2 comments 2 replies

Comment options

You must be logged in to vote
2 replies
@sonewtogithub
Comment options

@pflky
Comment options

Answer selected by sonewtogithub
Comment options

You must be logged in to vote
0 replies
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Category
Q&A
Labels
None yet
3 participants