Introduction

Similar to how a world map represents the globe, an image from a camera is a two-dimensional view of three-dimensional space. Media Server can use perspective when processing video of a scene. This allows Media Server to:

  • convert the position of an object in the scene into real-world 3D coordinates.
  • reduce the number of false positive results, by checking whether the size of an object, in pixels, corresponds to the real-world size of such an object. Media Server takes into account the position of the object in the scene and its distance from the camera.

The analysis engines that support perspective features are object class recognition and vehicle model recognition.