Batch processing for face detection (MTCNN)

Question

Is it possible to detect faces with mtcnn performing batch processing? I need to detect faces from 1 million images.

If it's possible, please share the ways to achieve that.

Answer 1

According to this:

batch_boxes, batch_points = detect_face(
    img, self.min_face_size,
    self.pnet, self.rnet, self.onet,
    self.thresholds, self.factor,
    self.device
)

would do the batch-inference