An application of virtual twins for fit prediction: case study of women jackets

Abstract

PurposeThe purpose of this paper is to present a fit prediction case study of virtual twins of women jacket. For this reason, several basic principles of clothes fit prediction were developed and verified.Design/methodology/approachTo develop the principles of fit prediction, the women's jacket sleeve was selected as the study object. The study objects were categorized into three types: the patterns and two virtual sleeves generating on full avatar from Clo3D and on dummy (arm partly removed). Through series of subjective and objective evaluation experiments, the relationship between the similar indexes of the patterns and the virtual sleeves was built, including fit criteria range, the categorization of the indexes in terms of its sensitiveness, and the linear regressions to predict several indexes of virtual sleeves after its pattern parameterization. The results obtained were verified in case study by the virtual and real sleeves generating.FindingsThe proposed principles of clothing fit prediction based on parallel research of the flat patterns and its virtual 3D shapes. The principles include the choosing of virtual twins of human body for virtual try-on, the establishing of indexes common schedule for patterns and virtual sleeves, the creation of criteria and its ranges for perfect fit and poor fit evaluation, and the application of existing relations between the patterns and the sleeves for predicting indexes responsible for fit.Research limitations/implicationsThe authors propose and verify the validity of the principles to predict several parameters of virtual 3D sleeve of women's jacket which are forming the level of fit. The result of this study can be used to convenient fit prediction and to find the misfit reasons.Practical implicationsThis study developed basic principles for predicting the fit of the clothing through the virtual simulation and the statistical analysis. Through studying the jacket sleeve, the several related ranges, the row of more sensitive indexes, and the equations were presented and verified, which certified the validity of proposed principles.Social implicationsThe results can effectively predict the sleeve fit before sewing, which reduce the time and materials cost and the operator's skill requirements.Originality/valueThe authors propose and verify the validity of the principles to predict several parameters of virtual 3D sleeve of women's jacket which form the level of fit. The result of this study can be used for convenient fit prediction and to find the misfit reasons.