Real-time Face Identification

To monitor the real-time face identification, go to the Events tab. The system can identify faces in both live video streams and archived videos. Besides monitoring, the Events tab also allows you to access the history of identification events.

Tip

Search for faces through the event database and dossier database on the Search tab.

In this chapter:

View Identification Events

Once a face detected, you will see a notification in the event list.

events_en

A notification can feature different pieces of information, depending on whether a detected face has a match in the database:

  • Match not found: a normalized face image, detection date and time, the name of a camera group.
  • Match found: a normalized face image, the photo from a dossier, the name of a person, similarity between faces, the comment from a dossier, the name of a dossier list, detection date and time, the name of a camera group.

Note

You can configure the system in such a way that you will get notifications only for the faces with a match.

Important

In order to pause the notifications thread, click pause above the list of events.

When working with events, the following default filters may come in handy:

  • Dossier: display events only for a selected dossier.

  • Watch lists: display events only for a selected dossier category (watch list).

    Note

    To view only unmatched faces in the event list, select Unmatched in this filter.

  • Matches: display events only with/without matches, or all events.

  • Acknowledged: display only acknowledged/unacknowledged events, or all events.

  • Cameras: display only events from a selected camera.

  • Camera groups: display only events from a selected group of cameras.

  • Start, End: display only events occurred within a certain time period.

  • id: display an event with a given ID.

Face Liveness and Face Features Recognition

Depending on the system settings, you can see an estimation of face liveness and/or a result of such face features recognition as gender, age, emotions, glasses and/or beard.

The face liveness detector automatically spots fake faces and prevents photo attacks by distinguishing a live face from a face image.

The face feature recognition result is in the following format:

Face feature Result format Example
Age Feature: age: number of years age: 33
Gender Result: male/female (feature: gender): algorithm confidence in result female (gender): 0.95
Emotions Result: angry/disgust/fear/happy/sad/surprise/neutral (feature: emotions): algorithm confidence in result happy (emotions): 0.99
Glasses Result: eye/sun/none (feature: glasses): algorithm confidence in result none (glasses): 0.87
Beard Result: beard/none (feature: beard): algorithm confidence in result none (beard): 0.91

features_en

Filter events by face features and liveness when needed.

filters_en

Event Ticket. Acknowledging Event

In order to navigate to an event ticket from the list of events, click on the face recognition result in a notification (No matches or the name of a matching person).

An event ticket contains the same data as a relevant notification. It also allows for acknowledging the event. To do so, click Not accepted to change the event aknowledgment status. Click Save.

accept_en

Tip

If a detected face has a match in the dossiers, you can navigate into a relevant one by clicking on the person’s name in the event ticket.

Tip

In order to acknowledge all the events, click check above the list of events.

Note

Event acknowledgment can be automated for selected watch lists.