This publication describes systems and techniques for makeup detection on an electronic device that uses an image of a user’s desired look as an input to detect makeup. The detected makeup is mapped to a virtual makeup library and saved as virtual makeup in a corresponding user profile. The user can retrieve and apply the virtual makeup to their face in another image to achieve a desired look. The mapped virtual makeup can be displayed as a filter over an image to digitally create an appearance of the user wearing makeup. The user is able to adjust the strength, color, and/or style of the virtual makeup. Further, the user may be presented with one or more recommendations for virtual makeup based on attributes of an image (e.g., a background of an image, the user’s clothing, hairstyle, etc.). The recommendations for virtual makeup may be based on results of a machine-learned model that received training from a professional source (e.g., stylist, makeup artist, etc.). The recommendations for virtual makeup may display on a captured image or they may display in real time on a display of the electronic device. The user is able to adjust the strength, color, and/or style of the recommendations for virtual makeup to achieve a desired look.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Lin, Shengyi; Masharani, Nisha; Renn, Marius; and Shaw, Leslie, "Applying Virtual Makeup Using Makeup Detection and Recommendations", Technical Disclosure Commons, (May 05, 2021)