Real-time Human Motion Analysis by Image Skeletonization

  • Authors:
  • Hironobu Fujiyoshi;Alan J. Lipton

  • Affiliations:
  • -;-

  • Venue:
  • WACV '98 Proceedings of the 4th IEEE Workshop on Applications of Computer Vision (WACV'98)
  • Year:
  • 1998

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper, a process is described for analysing themotion of a human target in a video stream. Moving targetsare detected and their boundaries extracted. From these,a "star" skeleton is produced. Two motion cues are determinedfrom this skeletonization: body posture, and cyclicmotion of skeleton segments. These cues are used to determinehuman activities such as walking or running, andeven potentially, the target's gait. Unlike other methods,this does not require an a priorihuman model, or a largenumber of "pixels on target". Furthermore, it is computationallyinexpensive, and thus ideal for real-world videoapplications such as outdoor video surveillance.