Corpus ID: 211296328

PolyGen: An Autoregressive Generative Model of 3D Meshes

@article{Nash2020PolyGenAA,
  title={PolyGen: An Autoregressive Generative Model of 3D Meshes},
  author={Charlie Nash and Yaroslav Ganin and S. Eslami and P. Battaglia},
  journal={ArXiv},
  year={2020},
  volume={abs/2002.10880}
}
Polygon meshes are an efficient representation of 3D geometry, and are of central importance in computer graphics, robotics and games development. Existing learning-based approaches have avoided the challenges of working with 3D meshes, instead using alternative object representations that are more compatible with neural architectures and training approaches. We present an approach which models the mesh directly, predicting mesh vertices and faces sequentially using a Transformer-based… Expand
26 Citations
DSM-Net: Disentangled Structured Mesh Net for Controllable Generation of Fine Geometry
  • 5
  • PDF
A survey on deep geometry learning: From a representation perspective
  • 12
  • PDF
Learning to Reconstruct and Segment 3D Objects
  • B. Yang
  • Computer Science, Engineering
  • ArXiv
  • 2020
  • PDF
REIN: Flexible Mesh Generation from Point Clouds
  • 1
  • PDF
Building-GAN: Graph-Conditioned Architectural Volumetric Design Generation
  • PDF
ShapeFlow: Learnable Deformations Among 3D Shapes
  • 7
  • PDF
SketchGraphs: A Large-Scale Dataset for Modeling Relational Geometry in Computer-Aided Design
  • 2
  • PDF
Engineering Sketch Generation for Computer-Aided Design
  • Highly Influenced
  • PDF
...
1
2
3
...

References

SHOWING 1-10 OF 41 REFERENCES
A Papier-Mache Approach to Learning 3D Surface Generation
  • 266
  • Highly Influential
  • PDF
Generating 3D faces using Convolutional Mesh Autoencoders
  • 183
  • PDF
Occupancy Networks: Learning 3D Reconstruction in Function Space
  • 439
  • Highly Influential
  • PDF
The shape variational autoencoder: A deep generative model of part‐segmented 3D objects
  • 78
  • PDF
DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation
  • 482
  • PDF
3D-R2N2: A Unified Approach for Single and Multi-view 3D Object Reconstruction
  • 795
  • PDF
ShapeNet: An Information-Rich 3D Model Repository
  • 1,797
  • Highly Influential
  • PDF
Octree Generating Networks: Efficient Convolutional Architectures for High-resolution 3D Outputs
  • 423
  • PDF
Unsupervised Learning of 3D Structure from Images
  • 281
  • PDF
PointFlow: 3D Point Cloud Generation With Continuous Normalizing Flows
  • 131
  • PDF
...
1
2
3
4
5
...