Virtual Makeover and Makeup Recommendation Based on Personal Trait Analysis

Abstract

The utilization of facial makeup is an important attribute in modern society as a means of self-expression, and as a method to feel more confident during social interactions. A makeover has become a necessity when attending divergent functions, and makeup used on diverse occasions varies in style. Choosing the perfect makeup that best suits a person is challenging unless they have years of expertise with cosmetics. This paper proposes a "Virtual Makeover and Makeup Recommendation System" to eliminate the need to be concerned about the appearance after applying makeup. The proposed system enables real-time makeup simulation in an Augmented Reality (AR) environment and recommends makeup styles considering skin tone, colour of clothing and hair, type of clothing and occasion to be attended with makeup. The personal traits of a user would be automatically detected and processed to generate recommendations for makeup products, namely lipstick, foundation and eyeshadow. Complications of wasting makeup products and time, and cleaning makeup can be mitigated by using a real-time makeup simulation system. Recommendations generated by the application assist the users to decide on makeup styles and provides a better user experience. The proposed system is developed with the aid of Deep Learning (DL) algorithms.