Intelligent clothing size and fit recommendations based on human model customisation technology

Abstract

E-commerce and mobile commerce have become part of our lives; the growth of online sales has continuously outpaced that of through traditional retail channels. However, even though nearly everyone now shops online, it is still challenging to sell clothes online. One of the major reasons is that size selection and clothing fit are still difficult to address within the current online platforms; this is reflected by the alarming return rate, between 20% and 40%, of online purchased clothes. New technology recently becomes available to create accurate 3D human models for individual customers by taking two photos using smart phone. The resulting models have accurate size and detailed local shape characteristics. The accuracy of the resulting models is comparable to scan, meeting the size tolerance of the clothing industry. Extending from this photo-based precise human modelling technology, we propose new solutions in this paper to advise suitable clothing sizes for individual customers, and to visualise virtual try-on effects of clothing on customised human models for online fit and size evaluation. We verify the size and fit recommendation solutions by experiment on clothing items of dress and shirt in this paper. With these novel sizing and fitting solutions, fashion brands can better serve their customers by attracting, engaging and delighting them with a brand new shopping experience.