SNAPPER: Fashion Coordinate Image Retrieval System
Shinya Miura, Toshihiko Yamasaki, Kiyoharu Aizawa · 2013
Fashion is one of the most important factors in our lives. A lot of people enjoy their fashion. On the other hand, they are not confident about their fashion coordination. Especially, it is a very worrisome problem to purchase a new fashion item that suits to their clothing. In this paper, we propose an image-retrieval-based system for supporting a person's fashion coordination and purchase of new fashion items. This system provides users an efficient way to know and purchase the fashion items which suit to those in their closets. The user uploads the image of his or her clothing and specifies the part which he or she puts it on. Then, our system retrieves similar items and recommends fashion coordinate from our fashion coordinate database that has been collected from fashion websites.