search

michaildoukas / head2head

PyTorch implementation for Head2Head and Head2Head++. It can be used to fully transfer the head pose, facial expression and eye movements from a source video to a target identity.
MIT License
310 stars 44 forks source link

Create dataset without LSFM #3

Closed thanisornsr closed 4 years ago

thanisornsr commented 4 years ago

Hello,

I have a question about creating my own dataset. In order to create my own dataset, are there any alternatives to LSFM model since I can't request for the LSFM. I am only master student which is not pass their qualification. Would you mind giving me some recommendation?

All the best, Thanisorn

michaildoukas commented 4 years ago

Unfortunately this Head2Head implementation currently supports only the LSFM model. In principle other 3D Morphable Models can be used to construct the NMFC conditional input images, but I would suggest requesting access to LSFM files.

Kindly, Michail

thanisornsr commented 4 years ago

Thank you for your kindly reply. I will ask my faculty for the requesting. Anyway, I will try to use other 3D Morphable Models as well just in case I cannot get the LSFM.

Big thanks, Thanisorn

ketan4373 commented 3 years ago

Hello @thanisornsr, Kindly provide way to use 3D Morphable Models if you have got success. Also if you can provide quality of results if you have got any. Thanks.