Obtaining the Similarity Value of Human Body Motions Through Their Sub Motions

Obtaining the Similarity Value of Human Body Motions Through Their Sub Motions

Truong Hong Ngan Pham (Kyoto Institute of Technology, Kyoto, Japan), Teruhisa Hochin (Graduate School of Information Science, Kyoto Institute of Technology, Kyoto, Japan) and Hiroki Nomiya (Graduate School of Information Science, Kyoto Institute of Technology, Kyoto, Japan)
Copyright: © 2020 |Pages: 19
DOI: 10.4018/IJSI.2020100105

Abstract

This article proposes a method of measuring human body motions by using the similarity measure through a 3D chain code. The method of deriving pauses in human body motions and splitting a motion into sub motions by pauses are also introduced. The requirements of sub motions to represent the similar parts between two motions are clearly explained. The pause waveform of basic human body motions is obtained and is used to split motions into sub motions. Two options of the proposed method are evaluated by experiments. The tests for statistical significance are used in evaluating the effectiveness of the method. The result shows that the first option of the method is effective in enhancing the performing time. It is shown that the proposed method is effective in measuring the similarity of human body motions.
Article Preview
Top

Chain Code Method for Encoding the Line-Drawing

The chain code method is a popular technique in computer vision, image processing, pattern recognition, and geographic information systems. The chain code method representing digital curves was introduced by Freeman (Freeman, 1961). A 2-dimensional curve is needed to be quantized and encoded to obtain its chain code. Firstly, a uniform mesh is superimposed on the line drawing. The approximate lines represent the curve defined by mesh nodes, which are closest to intersections of the curve and the standardized mesh. After that, the approximate line segments are encoded according to their directions. Any direction of a line segment in the approximation can be represented by the eight direction codes as shown in Figure 1. In other words, a sequence of direction codes, a chain of codes, represents a 2D line drawing.

3D Chain Code Method for Encoding 3D Digital Object

A 3D chain code method is proposed to represent 3D digital curves by extending the quantization structures of chain coding scheme from a 2D mesh to a cubic lattice (Herbert Freeman, 1974). The number of possible directions was increased from eight directions to 26 directions. A 3D chain code technique using orthogonal direction changes for digitizing a 3D continuous curve is introduced (Bribiesca, 2000). Any 3D discrete curve can be represented by a sequence of five possible direction change values – five codes. Sanchez-Cruz, Lopez-Valdez, and Cuevas (2014) proposed another approach using three main kind vectors for obtaining chain code elements. The relative changes of directions in the 3D Euclidean space are utilized to obtain a simple path in a grid of 26 connected components. An example of a chain code representing a discrete curve using six directions is shown in Figure 2.

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 9: 4 Issues (2021): Forthcoming, Available for Pre-Order
Volume 8: 4 Issues (2020)
Volume 7: 4 Issues (2019)
Volume 6: 4 Issues (2018)
Volume 5: 4 Issues (2017)
Volume 4: 4 Issues (2016)
Volume 3: 4 Issues (2015)
Volume 2: 4 Issues (2014)
Volume 1: 4 Issues (2013)
View Complete Journal Contents Listing