Skip to main content

Video object detection tracker

The video object detection module tracks objects detected by AI models. This data is used by the Fusion Tracker module for a more complete understanding of the scene.

Instances​

A device is preconfigured with a fixed number of instances of this module.

In general, each physical video channel has one instance of this module. On some multisensor cameras, several image sensors produce a stitched panorama, and the module has one instance for that full view.

  • Devices with one image sensor will have one tracker instance.
  • Multidirectional cameras will have one tracker instance for each image sensor.
  • Multisensor cameras that produce a panorama stitched from several image sensors have one tracker instance for the full view area.

Function​

The video object detection tracker tracks objects in the scene and estimates their positions and attributes.

Configuration​

Privacy mask​

You can enable privacy masks through the camera's video-analytics - v1 API. You can also use this API to configure how analytics behave when objects appear behind privacy masks.

You can access the API at:

http://camera-ip/config/web-ui/swagger-ui/?url=/config/discover/apis/video-analytics/v1/openapi.json

Output protocols​

This module's output isn't directly accessible. For more information, see the Computer Vision SDK examples.