search

andrerochow / fsrt

Official implementation of the CVPR 2024 paper "FSRT: Facial Scene Representation Transformer for Face Reenactment from Factorized Appearance, Head-pose, and Facial Expression Features"
https://andrerochow.github.io/fsrt
67 stars 4 forks source link

Does it support only 256 x 256 only? #7

Closed nitinmukesh closed 3 weeks ago

nitinmukesh commented 3 weeks ago

As the title says

In code resize(img, (256, 256)

Can't we use high resolution images?

[EDIT] The models are trained for 256 x 256 vox256.pt

Inferencer commented 3 weeks ago

As the title says

In code resize(img, (256, 256)

Can't we use high resolution images?

[EDIT] The models are trained for 256 x 256 vox256.pt

use liveportrait repo its current sota and 512