Design and Realization of Real-time Face Beauty System based on Android

With the popularity of mobile terminals and the speed-up of network, short videos are favored by all platforms, fans and capital. People often record and share their lives through short videos. For most users, it is vitally important to show their best self to the audience. Therefore, short video software needs to provide users with a variety of real-time beauty effects, such as face thinning, whitening, mapping and filters, to meet users’ demand for beauty. This paper mainly focuses on the design of a real-time facial beauty system based on Android. The system uses key point detection technology based on deep learning to realize detection of key facial feature point, adopts OpenGL ES to realize image deformation algorithm, and adjusts the face shape according to the detected key points, so as to achieve the effects of thin face and big eyes. The system mainly includes three functional modules: camera module, detection module of key facial feature point and beauty filter module. To meet individual needs, every function parameter in the system is adjustable.