ACGAN: Age-Compensated Makeup Transfer Based on Homologous Continuity Generative Adversarial Network Model

The authors focus on the makeup transformation problem, which refers to the transfer of makeup from a reference face to a source face image while maintaining the source makeup-free face image. In recent years, makeup transformation has become a hot issue and a lot of research has been conducted on this basis, but there are some limitations in the existing methods, mainly due to the lack of consideration of age factor, which makes the final generated face makeup images appear not natural and lack appearance attractiveness. In order to further solve this problem, an age-compensated makeup transformation framework based on homology continuity is proposed. In order to achieve a stable and controllable age-compensation effect, the authors design a new coding module that can map the face makeup semantic vector into the higher feature space and achieve age compensation by adjusting the direction of the semantic vector. Finally, in order to comprehensively evaluate the effectiveness of the authors’ proposed method, a large number of qualitative and quantitative experiments have been conducted, and the experimental results show that the authors’ proposed framework outperforms existing methods.