search

EvelynFan / FaceFormer

[CVPR 2022] FaceFormer: Speech-Driven 3D Facial Animation with Transformers
MIT License
778 stars 133 forks source link

How to group into 51 blendshapes and driven new charactor? #38

Open lucasjinreal opened 1 year ago

lucasjinreal commented 1 year ago

How to group into 51 blendshapes and driven new charactor?

ujjawalcse commented 1 year ago

Hey @jinfagang , Did you get the blendshape mapping? if yes, Please share how you did. Thanks.

EvelynFan commented 1 year ago

Hi, for the current model, the output representation is vertex positions. I guess some other techniques based on inverse graphics may be required to convert the vertex positions into 51 blendshape coefficients.

ujjawalcse commented 1 year ago

Thanks @EvelynFan for your suggestion.