3D landmark dataset to 2D 68 points

Hi,
Thank you for sharing your work.
I there anyway to convert the 3D landmark dataset to 2D 68 point landmark (without occluded) as shown in the **Detect 2D facial landmarks in pictures** part in https://github.com/1adrianb/face-alignment.

Thank you.
tsly

The issue asks about converting a 3D landmark dataset to 2D 68-point landmarks, referencing the 'Detect 2D facial landmarks in pictures' section of the repository. Start by examining the existing 2D landmark detection code in the repository, likely in the face_alignment module. Understand how the current 2D projection or conversion works, and research the dataset structure for 3D landmarks. 'Done' would be a script or method that takes 3D landmarks and outputs the corresponding 2D 68-point set without occluded points.

Vom Indexierungsmodell aus dem Issue-Text verfasst.