face-vid2vid icon indicating copy to clipboard operation
face-vid2vid copied to clipboard

about the nework

Open marvin-nj opened this issue 3 years ago • 3 comments

Hello, zhengkw18, thank you for your contribution!

the output “delta” of the the HPE_EDE model should be the expression of the persion, not head pose, right ? but , when i frozen the yaw,pitch and roll matrixs, and only extract delta feature from HPE model of driving person , the source persion still have a head movtion. so , what's wrong with me?

I want to transfer one person's expression from another, with no head movtion. how shoud i do.

marvin-nj avatar Oct 26 '21 09:10 marvin-nj

My pretrained model is insufficiently trained so the decorrelation between the components is not guaranteed.

zhengkw18 avatar Oct 26 '21 10:10 zhengkw18

thanks! so , the expression feature can be decomposed if model trained sufficiently, right ?

marvin-nj avatar Oct 27 '21 01:10 marvin-nj

Theoretically it is. You can see the NVlab's demo, which demonstrates quite impressive results.

zhengkw18 avatar Oct 27 '21 05:10 zhengkw18