Metadata is logical information extracted from video streams for easier camera management and retrieval of object-related content.

In Axxon Next, metadata can be obtained in two ways:

  1. by analyzing video streams on Server;
  2. by receiving metadata from edge devices (see Embedded Detection Tools). 

The following tools are used for server-side analysis and metadata generation:  

Attention!

To extract metadata from video, you have to de-compress and analyze the video stream which, in its turn, increases the Server's workload, thus limiting the number of available camera channels.

  1. Object Tracker.

  2. Neural Tracker.

  3. VMD.

    Note

    Object tracker and Neural Tracker generates metadata containing the following information about moving objects in scene: object type, position, size and color, motion speed and direction, etc.

    VMD generates less accurate data; it does not detect object type and color.

  4. Face detection tool.

    Note

    Face detection metadata contain facial bounding boxes and their positions, as well as facial vectors.

  5. Automatic Number Plate Recognition (LPR/ANPR) tools.

    Note

    ANPR metadata contain license plate bounding boxes and their positions, as well as vehicle registration numbers.

  6. Pose detection.

    Note

    Metadata from pose detection tools contains information on positions and postures of all persons in FOV. 

The metadata is used for the following system options:

Option

Required source of metadata

Scene Analytics

Object Tracker, Neural Tracker, built-in detection tool, or VMD.

Forensic search MomentQuest

Object Tracker, Neural Tracker, built-in detection tool, or VMD.

Face search

Face detection tool.

LPR search

Automatic Number Plate Recognition (LPR/ANPR).

TimeCompressor

Object Tracker, Neural Tracker, built-in detection tool, or VMD.

Tag&Track Pro

Object Tracker, Neural Tracker, built-in detection tool, or VMD.

Object tracking

Any

Autozoom

Any

Attention!

If a camera uses several sources of metadata, the required source is selected automatically, except for MomentQuest.

To perform facial/license number searches, only metadata from corresponding detection tools is used.

By default, metadata files are stored in Server's object trajectory database: C:\Program Files\AxxonSoft\AxxonNext\Metadata\vmda_db\VMDA_DB.0\vmda_schema; if necessary, you can place them on any available network storage (see Configuring storage of the system log and metadata).

  • No labels