Recognizing human actions based on silhouette energy image and global motion description

In this paper, we propose a spatio-temporal silhouette, called silhouette energy image (SEI), and models, to characterize motion and shape properties automatic recognition of human actions in daily . To address the variability in the recognition of human, several parameters, such as anthropometry of the, speed of the action, phase (starting and ending state an action), camera observations (distance from camera, motion, and rotation of human body), and view are proposed. We construct the variability models on SEI and the variability parameters. The global based motions express the spatio-temporal properties SEI and variability models. Our construction of the model for each action and view is based on the vectors of motion descriptions of combined action . We recognize different daily human actions of different successfully in the indoor and outdoor environment. experimental results show that the proposed of human action recognition is robust, flexible and efficient.