Why don't you use detach function about embedding vector in generator?

grey-eye / talking-heads

Our implementation of "Few-Shot Adversarial Learning of Realistic Neural Talking Head Models" (Egor Zakharov et al.)

GNU General Public License v3.0

593 stars 110 forks source link

Why don't you use detach function about embedding vector in generator? #46

Open Longseabear opened 4 years ago

Longseabear commented 4 years ago

G(rand_mark, embedding; psi, P)

In paper formula, the generator's training parameters are only psi and projection function. Is there any special reason not to detach () the embedding vector?