Generative adversarial networks and their application to 3D face generation: A survey

Abstract

Generative adversarial networks (GANs) have been extensively studied in recent years and have been used to address several problems in the fields of image generation and computer vision. Despite significant advancements in computer vision, applying GANs to real-world problems such as 3D face generation remains a challenge. Owing to the proliferation of fake images generated by GANs, it is important to analyze and build a taxonomy for providing an overall view of GANs. This, in turn, would facilitate many interesting applications, including virtual reality, augmented reality, computer games, teleconferencing, virtual try-on, special effects in movies, and 3D avatars. This paper reviews and discusses GANs and their application to 3D face generation. We aim to compare existing GANs methods in terms of their application to 3D face generation, investigate the related theoretical issues, and highlight the open research problems. Authors provided both qualitative and quantitative evaluations of the proposed approach. They claimed their results show the higher quality of the synthesized data compared to state-of-the-art ones.