摘要
The purpose of this paper was to develop a reliable body shape analysis approach based on cluster analysis, k. nearestneighbor( KNN), and multi-class support vector machine( MSVM). Firstly,a total of 357 Chinese men were selected to make a dataset. Secondly, the experiences of these data were not accumulated to build general models. Five body angles were extracted as independent variables. Four clusters were the most efficient cluster number for our study. Finally,the accuracy of body classifications is compared between KNN and MSVM. In this study,the body classification framework was studied to transfer the body feature data to intuitive types. Moreover,the adaptive made-tomeasure( MTM) framework based on body classification was studied. The case demonstration and analysis show the effectiveness of the study.
The purpose of this paper was to develop a reliable body shape analysis approach based on cluster analysis, k. nearestneighbor（ KNN）, and multi-class support vector machine（ MSVM）. Firstly,a total of 357 Chinese men were selected to make a dataset. Secondly, the experiences of these data were not accumulated to build general models. Five body angles were extracted as independent variables. Four clusters were the most efficient cluster number for our study. Finally,the accuracy of body classifications is compared between KNN and MSVM. In this study,the body classification framework was studied to transfer the body feature data to intuitive types. Moreover,the adaptive made-tomeasure（ MTM） framework based on body classification was studied. The case demonstration and analysis show the effectiveness of the study.
基金
Talent Project of Xiamen University of Technology,China(No.90030617)
