Automatic Virtual Makeup System Using User-Preference Information

Abstract

We present an automatic virtual makeup system with an interactive evolutionary computation (IEC) method using user preference information. While purchasing cosmetic items, users try each cosmetic item and check their appearance. Consequently, users feel burdened to try numerous cosmetic products before making a choice. To avoid this burden, some companies have developed virtual makeup applications for cosmetic displays using tablet devices. However, users who try numerous cosmetic items to choose from become confused and burdened with making the correct selection. Therefore, we propose an automatic virtual makeup system to generate users’ favorite makeup patterns using their preferences. The system employs the IEC method to generate a user preferred makeup. The IEC method is a probability retrieval technique adopted to generate designs that satisfy user preference. Moreover, the proposed system contains direct manipulation operations for choosing a user’s favorite makeup part design. We performed evaluation experiments to investigate the effectiveness of the proposed system from the viewpoints of the IEC with paired comparison applications. The subjects were female students in their twenties. From the results, we confirmed that the proposed system was useful from the viewpoint of applying the IEC method to a virtual makeup system.