search

huggingface / diffusers

🤗 Diffusers: State-of-the-art diffusion models for image and audio generation in PyTorch and FLAX.
https://huggingface.co/docs/diffusers
Apache License 2.0
25.39k stars 5.26k forks source link

ip adapter face id #6935

Open alexblattner opened 7 months ago

alexblattner commented 7 months ago

Is your feature request related to a problem? Please describe. Can't use ip adapter face id with ip adapter face if I use the community pipeline

Describe the solution you'd like. Add ip adapter face id to the official support

Describe alternatives you've considered. Spending a couple days and do something myself

Additional context. none really

yiyixuxu commented 7 months ago

I think ip-adapter face has a different attention processor so it can't be used in multi-ip-adapter anyway

cc @fabiorigano here

fabiorigano commented 7 months ago

exactly, the community pipeline implementation has many differences and currently can be only used alone

attention processors in ip-adapter face id only support the single-ip-adapter (face id) inference

in addition, this adapter uses image embeddings extracted by third-party software and not images, which means that using ip adapter face id and ip adapter face is not as straightforward as using ip adapter plus and ip adapter face

github-actions[bot] commented 6 months ago

This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.

Please note that issues that do not follow the contributing guidelines are likely to be ignored.