MeshGAN: Non-linear 3D Morphable Models of Faces
arXiv:1903.10384
Abstract
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.
References in corpus (16)
- Semi-Supervised Classification with Graph Convolutional Networks
- Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
- GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
- Inductive Representation Learning on Large Graphs
- Geometric deep learning: going beyond Euclidean data
- Spectral Networks and Locally Connected Networks on Graphs
- Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling
- Progressive Growing of GANs for Improved Quality, Stability, and Variation
- Deep Convolutional Networks on Graph-Structured Data
- BEGAN: Boundary Equilibrium Generative Adversarial Networks
- Representation Learning on Graphs: Methods and Applications
- Review of Statistical Shape Spaces for 3D Data with Comparative Analysis for Human Faces
- Generating 3D faces using Convolutional Mesh Autoencoders
- Nonlinear 3D Face Morphable Model
- Inverting The Generator Of A Generative Adversarial Network (II)
- A Simple Approach to Intrinsic Correspondence Learning on Unstructured 3D Meshes
Cited by in corpus (9)
- MeshSDF: Differentiable Iso-Surface Extraction
- Single Image 3D Hand Reconstruction with Mesh Convolutions
- Deep Deformable 3D Caricatures with Learned Shape Control
- FaceScape: a Large-scale High Quality 3D Face Dataset and Detailed Riggable 3D Face Prediction
- Latent Disentanglement in Mesh Variational Autoencoders Improves the Diagnosis of Craniofacial Syndromes and Aids Surgical Planning
- Disentangled Face Identity Representations for joint 3D Face Recognition and Expression Neutralisation
- Unsupervised Geometric Disentanglement for Surfaces via CFAN-VAE
- Shape registration in the time of transformers
- Dynamic Facial Asset and Rig Generation from a Single Scan