Detection of Changes in Surveillance Videos

In this paper we provide theoretical and experimental results that dimension of video trajectories is a useful tool to access mid-level content of videos, like, appearance or disappearance of an object, and changes in velocity and in direction of moving objects. Moreover, the amount of changes is proportional to the size of objects involved and their speed. All this is achieved by a robust technique of dimensionality computation of video trajectories based on eigenvalues.

• Type: Conference
• Year: 2003
• Where: AVSS
• Authors: Longin Jan Latecki, Xiangdong Wen, Nilesh Ghubade
• Keywords: Access Mid-level Content, AVSS 2003, Signal Processing, Useful Tool, Video Trajectories