cleardusk / 3DDFA_V2

The official PyTorch implementation of Towards Fast, Accurate and Stable 3D Dense Face Alignment, ECCV 2020.
MIT License
2.9k stars 514 forks source link

Camera matrix #57

Open pietro-ux opened 3 years ago

pietro-ux commented 3 years ago

Hi, what does the "camera matrix" P in utils/pose.py represent? Is it actually describing the position of the camera or the transformation of the face?

cleardusk commented 3 years ago

The transformation of the face. 3DDFA_V2 adopts the scale orthographic projection.

abcliguanxi commented 3 years ago

The transformation of the face. 3DDFA_V2 adopts the scale orthographic projection.

what is the difference between the scale orthographic projection and the weak perspective projection?

cleardusk commented 3 years ago

@abcliguanxi The same.