Human movement analysis around a view circle using time-order similarity distributions

• We propose a novel scheme for view-changeable action event analysis.
• A view alignment scheme is proposed for action analysis around a view circle.
• A property of mirror symmetry is proposed for reducing the whole view space.
• A novel time-order similarity distribution matrix is proposed for robust event analysis.

This paper presents a new behavior classification system to analyze human movements around a view circle using time-order similarity distributions. To maintain the view in-variance, an action is represented not only from its spatial domain but also its temporal domain. After that, a novel alignment scheme is proposed for aligning each action to a fixed view. With the best view, the task of behavior analysis becomes a string matching problem. One novel idea proposed in this paper is to code a posture using not only its best matched key posture but also other unmatched key postures to form various similarity distributions. Then, recognition of two actions becomes a problem of matching two time-order distributions which can be very effectively solved by comparing their KL distance via a dynamic programming scheme.