Closed zhang-haojie closed 5 months ago
you may find AniPortrait more relevant - they enhanced MediaPipe lip detection - and landmark detection is SOTA. https://github.com/Zejun-Yang/AniPortrait/blob/cb86caa741d6ab1e119ea7ac2554eb28aabc631b/src/utils/face_landmark.py#L123
For landmarks
we used face-alignment. But feel free to use anything for that.
audio_emb_dir
is a directory containing audio embeddings. The folder structure should be the same as data's (as indicated by file list). You can find information on how to compute them in README.
The landmarks
is 3D facial landmarks or 2D facial landmarks? Thank you
They are 2D.
Dear auther, how to set param
landmarks
andaudio_emb_dir
and their corresponding data? Could you provide more information about them?