Evaluation of Similarity Measuring Method of Human Body Movement Based on 3D Chain Code

Abstract

This paper experimentally evaluates the method for measuring human body movement similarity considering the human body motion as a set of 3-dimensional curves representing the changes of direction. After the direction sequences describing the motion paths are derived, an improved method inspired by the dissimilarity measure of the 3D curves is used for measuring the similarity of human body movements. The tests for statistical significance is used in order to evaluate the efficiency of measure in the experiments. The experimental results show that the proposed method successfully measures the similarity of human body movements for the low complexity dataset and the similar motion types.