MeshGAN: Non-linear 3D Morphable Models of Faces
This work addresses the challenge of generating realistic 3D faces for applications in computer graphics and vision, representing an incremental improvement by adapting GANs to mesh-based representations.
The paper tackles the problem of generating realistic 3D faces using GANs, which have been less successful for 3D objects compared to images, by proposing MeshGAN, an intrinsic GAN architecture operating directly on 3D meshes, resulting in high-fidelity 3D faces with rich identities and expressions as shown through quantitative and qualitative results.
Generative Adversarial Networks (GANs) are currently the method of choice for generating visual data. Certain GAN architectures and training methods have demonstrated exceptional performance in generating realistic synthetic images (in particular, of human faces). However, for 3D object, GANs still fall short of the success they have had with images. One of the reasons is due to the fact that so far GANs have been applied as 3D convolutional architectures to discrete volumetric representations of 3D objects. In this paper, we propose the first intrinsic GANs architecture operating directly on 3D meshes (named as MeshGAN). Both quantitative and qualitative results are provided to show that MeshGAN can be used to generate high-fidelity 3D face with rich identities and expressions.