Face Recognition

The Face Recognition video analytic offers facial recognition, event triggering, and alarms.

The Face Recognition video analytic uses profiles to represent individuals. Each profile can include multiple images of an individual's face. Use the Symphony Client interface to create profiles, add images of faces, and set whether the profile is blocked or allowed. When the Face Recognition video analytic identifies a face in video, it can trigger events on the Symphony Server.

The Face Recognition video analytic can detect faces when there is greater than 29 pixels between the eyes and it works best when the resolution of the camera is greater than or equal to 640 x 480.

The Face Recognition video analytic includes spoof attack detection. The video analytic can detect the difference between pictures of faces and real faces. You can use spoof attack detection to generate alarms when the video analytic detects spoofed faces.

Note: The Face Recognition video analytic cannot generate alarms on real faces and spoofed faces at the same time. You must create two rules to generate alarms on both.

In the Symphony Client interface, a blue square surrounds a tracked face and a red square surrounds a tracked face that triggers an alarm.