A Papier-Mache Approach to Learning 3D Surface Generation (2018-02-15T00:00:00.000000Z)

TL;DR

This work introduces a method for learning to generate the surface of 3D shapes as a collection of parametric surface elements and, in contrast to methods generating voxel grids or point clouds, naturally infers a surface representation of the shape.

Abstract

We introduce a method for learning to generate the surface of 3D shapes. Our approach represents a 3D shape as a collection of parametric surface elements and, in contrast to methods generating voxel grids or point clouds, naturally infers a surface representation of the shape. Beyond its novelty, our new shape generation framework, AtlasNet, comes with significant advantages, such as improved precision and generalization capabilities, and the possibility to generate a shape of arbitrary resolution without memory issues. We demonstrate these benefits and compare to strong baselines on the ShapeNet benchmark for two applications: (i) autoencoding shapes, and (ii) single-view reconstruction from a still image. We also provide results showing its potential for other applications, such as morphing, parametrization, super-resolution, matching, and co-segmentation.

Authors

Bryan C. Russell

13 papers

Matthew Fisher

6 papers

Mathieu Aubry

9 papers

TL;DR

Abstract

Authors

References34 items

High-Resolution Shape Completion Using Deep Neural Networks for Global Structure and Local Geometry Inference

Convolutional neural networks on surfaces via seamless toric covers

PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space

OctNetFusion: Learning Depth Fusion from Data

Hierarchical Surface Prediction for 3D Object Reconstruction

Octree Generating Networks: Efficient Convolutional Architectures for High-resolution 3D Outputs

SurfNet: Generating 3D Shape Surfaces Using Deep Residual Networks

PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation

A Point Set Generation Network for 3D Object Reconstruction from a Single Image

Learning Shape Abstractions by Assembling Volumetric Primitives

Image-to-Image Translation with Conditional Adversarial Networks

Deep Learning 3D Shape Surfaces Using Geometry Images

Accelerated quadratic proxy for geometric optimization

Learning shape correspondence with anisotropic convolutional neural networks

Marr Revisited: 2D-3D Alignment via Surface Normal Prediction

3D-R2N2: A Unified Approach for Single and Multi-view 3D Object Reconstruction

Learning a Predictable and Generative Vector Representation for Objects

Deep Residual Learning for Image Recognition

ShapeNet: An Information-Rich 3D Model Repository

Deep Exemplar 2D-3D Detection by Adapting from Real to Rendered Views

Joint embeddings of shapes and images via CNN image purification

Geodesic Convolutional Neural Networks on Riemannian Manifolds

FAUST: Dataset and Evaluation for 3D Mesh Registration

3D ShapeNets: A deep representation for volumetric shapes

Screened poisson surface reconstruction

Et al

Mesh parameterization: theory and practice

Multi-Chart Geometry Images

Geometry images

Metro: Measuring Error on Simplified Surfaces

A Method for Registration of 3-D Shapes

Approximation capabilities of multilayer feedforward networks

Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks

Supplementary material (appendix) for the paper https

Field of Study

Journal Information

Name

Page