Human Motion Analysis

Abstract

Human motion analysis in the video has its vast application. The recognition of the newlinehuman action is the most prominent application of human motion analysis. In this newlineresearch we analyzed different methodologies for modeling human action. We also newlinediscussed challenges and methodologies which are used to handle them. These newlinemethodologies are divided into two categories. One is global feature descriptor and other newlineis local feature descriptors. The disadvantage of the global feature descriptor is that they newlinecan only give the shape information but fails to give motion information. The local newlinefeature descriptors are used to find out the motion information of the action video. The newlinedisadvantage is that it cannot give the shape or structure information of the action video. newlineThe hybrid descriptors are used to solve these problems but these descriptors suffer from newlinehigh dimensionality features. In this research we proposed new feature descriptors which newlineare capable to deal with these issues in the following manner: newline1) We proposed a new local descriptor evaluated from the Finite Element Analysis newlinefor human action recognition. This local descriptor represents the distinctive newlinehuman poses in the form of the stiffness matrix. This stiffness matrix gives the newlineinformation of motion as well as shape change of the human body while newlineperforming an action. Initially, the human body is represented in the silhouette newlineform. Most prominent points of the silhouette are then selected. This silhouette is newlinediscretized into several finite small triangle faces (elements) where the prominent newlinepoints of the boundaries are the vertices of the triangles. The stiffness matrix of newlineeach triangle is then calculated. The feature vector representing the action video newlineiii newlineframe is constructed by combining all stiffness matrices of all possible triangles. newlineThese feature vectors are given to the Radial Basis Function-Support Vector newlineMachine (RBF-SVM) classifier.