0

There is a confirmation from perhaps Google engineer here that MLKit cannot determine whether two detected faces are of the same person. However, according to Face detection document:

Track faces across video frames Get an identifier for each unique detected face. The identifier is consistent across invocations, so you can perform image manipulation on a particular person in a video stream.

I am wonder why it doesn't work on a list of photos as video is just combination of photo frames. It seems currently only MLKit is the on-device library for face detection that works without API, it would be great if MLKit supports face recognizer as well.

thanhbinh84
  • 17,876
  • 6
  • 62
  • 69

1 Answers1

2

MLKit face tracking is done via the position and motion of the face in the video sequence. Face recognition is not currently supported and there is no plan to support this going forward.

Hoi
  • 599
  • 2
  • 7