About shape and expression parameters clipping

[Andyzzz]

“The parameters clipping accelerates the training and reconstruction, but degrades the accuracy especially the details like closing eyes. Below is an image, with parameters dimension 40+10, 60+29 and 199+29 (the original one). Compared to shape, expression clipping has more effect on reconstruction accuracy when emotion is involved. Therefore, you can choose a tradeoff between the speed/parameter-size and the accuracy. A recommendation of clipping trade-off is 60+29.”

Is there a pre-trained model with 60+29 or 199+29 dimensional parameters available? If I want to use 199+29 dimensional models, what should I do?

[cleardusk]

Nope now, but this repo may be updated in the near future.

[zaverichintan5]

@cleardusk any updates on the pre-trained model with 60+29 or 199+29 dimensional parameters ?