Video: Target Tracking in Video Data

This video contains an example of Eonfusion's capabilities for analysis and visualization of video data.

Video data can be considered as a special case of a 3D raster data set. Eonfusion's powerful and flexible high-dimensional raster handling can be seamlessly applied to video data. This main data set consists of more than an hour of video data shot from a stationary camera.

  • The video data is analyzed to detect moving targets.
  • Targets are converted to vector point data and tracked.
  • All detected tracks are consolidated into a six-dimensional (6D) matrix that defines "normal" object behavior.
  • The matrix is used to identify tracks that correspond to objects that behave "abnormally".
  • A non-spatial visualization allows the full data set to be summarized by a single, browsable timeline.

There are two downloads below: each contains the same video in either .wmv (Windows Media) or .mov (Apple Quicktime) format.