Batch Inference of Facial Landmarks

[mibrahimy]

• face_recognition version: 1.2.3
• Python version: 2.7 and 3.5.2
• Operating System: Ubuntu 16.04

I am trying to use 5 pt shape predictor for batch inference. I have a batch of face crops let says 50, and I want to get their landmarks returned in a single go.

What I Did

I tried passing the dlib.shape_predictor() function the list of images as first argument and list of respective detections as second. But it returned the following error:

TypeError: __call__(): incompatible function arguments. The following argument types are supported:
    1. (self: dlib.shape_predictor, image: array, box: dlib.rectangle) -> dlib.full_object_detection

[maheshwarkuchana]

@ibrahimsherdil

The function dlib.shape_predictor() accepts

• shape_predictor_face_landmarks.dat file

• Only one image as numpy ndarray

as its arguments.

But we cannot pass a list of images to predict the landmarks.

For more explanation in documentation I prefer you to read this link