The ML Kit's face detection API allows for the detection of faces in an image or video. Once the face is detected, we can apply the following refinements:

• Landmark detection: Determining specific points of interest (landmarks) within the face, such as the eyes
• Classification: Classifying the face based on certain characteristics, such as whether the eyes are open or closed
• Face tracking: Recognizing and tracking the same face (in various positions) across different frames of video

Face detection can be done only on-device and in real time. There may be many use cases for mobile device applications, in which the camera captures an image and manipulates it based on landmarks or classifications, to produce selfies, avatars, and so on.