Can we get an improved version of IP-Adapter-Face(non-FaceID) model?

tencent-ailab / IP-Adapter

The image prompt adapter is designed to enable a pretrained text-to-image diffusion model to generate images with image prompt.

Apache License 2.0

4.5k stars 296 forks source link

Can we get an improved version of IP-Adapter-Face(non-FaceID) model? #267

Open ynie opened 5 months ago

ynie commented 5 months ago

Hello, thanks for this research project. We would like to use it in our commercial project, but we can’t use the latest and greatest FaceID based models. Could you also improved the non-faceID based models? Thank you!

xiaohu2015 commented 5 months ago

FaceID using non-commercial insightface. In the future, we may use some free models