Real-time virtual clothes try-on system

Abstract

Virtual clothes try-on has recently come into the researchers' focus owing to its commercial potential. It can be used as an efficient tool in online shopping, allowing users to narrow down their choices to a few designs and sizes. In this study, a real 3D virtual clothes try-on system is proposed, which makes it possible for users to see themselves on a virtual monitor, wearing virtual clothes without taking off their real clothes. The tool also allows users to choose various clothes to try on. The most important point in this paper is that by using a motion and skeleton capture sensor, the user's body measurements are computed. By comparing the computed body sizes with the measurements in the body size standards, the user's body size can be estimated via an expert system and these data are used for proper clothes fitting and clothes simulation in our virtual try-on system. Clothes are selected by hand motion from the clothing list on the screen and users can thus view the clothing of their choice virtually displayed on their body. This study contributes to the current literature by offering an original approach, which brings a new perspective on body scanning, body determination and virtual clothing simulation in the clothing sector.