Applications of Generative Adversarial Networks in Hairstyle Transfer

Abstract

Recently, Generative Adversarial Networks (GANs) extensively studied, and a range of GANs architectures and related methods proposed. The evolution of GANs makes a significant impact in the areas of computer vision; thus, significant advances have been made, for instance, plausible image generation, image-to-image translation, facial attribute manipulation, and related domains. Most recent works have focused on many domains instead of focusing on a single domain of GANs. In this works, we want to focus on the applications in a single area named facial attribute manipulation, an exciting area of GANs. We apply the state-of-the-art architectures and methods of GANs in facial attribute manipulation for editing the hairstyle of a given facial image.