Parametric human body modelling system for virtual garment fitting

Abstract

Modelling customer-tailored 3D human models is an important problem in virtual garment fitting systems. However, manual generation of such models often requires high cost and is considerably difficult for common users lacking the knowledge of computer graphics. To solve this, we propose a parametric human modelling method. In our study, the 3D body scan data of 250 different individuals are gathered, and for each model, 128 feature points are specified. The modelling parameters are then extracted by analysing the feature point statistics. Using these parameters, we generate a new set of feature points satisfying the input body sizes, and interpolate those points with a surface. As a result, a 3D human model satisfying the given constraints is obtained. Since our method uses only a few parameters, even a novice user can generate or modify the model easily. Further, since the method involves only simple calculations, the computational cost is low. Copyright © 2013 Inderscience Enterprises Ltd.