[Feature]: Face Model with multiple images

[A-2-H]

Feature description

I saw a request about mixed face feature that was approved here as a future option. I don't know if its the same or not, but It would be great to have option to make face model with multiple images so the results will have better similarity to the person that we want to create. It is the feature in Roop Unleashed. You can create face model with multiple images and the results are much better in compare to one image model.

Also while making the model of a face, when it extract the face from an image to make it possible to choose the correct face that was extracted and delete the wrong one. Sometimes on image there are multiple people and it can affect the face model so to not make extra step in external program to cropout faces that we don't want in our model it would be great that we have the option to delete the faces that we don't want in our model.

Right now we can create face model with one images, I don't see the point to it as it is the same as to put one image as a source.

[Gourieff]

Yes, it will be implemented

[agarzon]

I was about to ask the same 😀