Research on Image-based Automatic Modification Algorithm of Eyebrows

To a certain extent, modern aesthetics force people to change the shape and color of eyebrows according to the current popular form, and achieve the aesthetic effect of the overall effect. How to match the appropriate eyebrows according to different face types has always been pursued and desired by people. However, hand animation eyebrows require certain skills. Not everyone can draw beautiful and clean eyebrows. In order to meet the requirements of automation and real-time in virtual makeup technology, this paper proposes an algorithm for automatically modifying eyebrows by using digital image processing technique. Firstly, the Haar classifier and the Dajin threshold method combined with the Haar-Like feature and the AdaBoost algorithm are used to realize the detection, segmentation and replacement of the eyebrows. At the same time, the color transformation can be performed to achieve further modification effects. Through the experimental results, it can be found that the proposed automatic modification algorithm not only realize the automatic modification of eyebrows and the transformation of different colors, but also the makeup effect is natural and beautiful.