Learning Manifold Patch-Based Representations of Man-Made Shapes
arXiv:1906.12337
Abstract
Choosing the right representation for geometry is crucial for making 3D models compatible with existing applications. Focusing on piecewise-smooth man-made shapes, we propose a new representation that is usable in conventional CAD modeling pipelines and can also be learned by deep neural networks. We demonstrate its benefits by applying it to the task of sketch-based modeling. Given a raster image, our system infers a set of parametric surfaces that realize the input in 3D. To capture piecewise smooth geometry, we learn a special shape representation: a deformable parametric template composed of Coons patches. Naively training such a system, however, is hampered by non-manifold artifacts in the parametric shapes and by a lack of data. To address this, we introduce loss functions that bias the network to output non-self-intersecting shapes and implement them as part of a fully self-supervised system, automatically generating both shape templates and synthetic training data. We develop a testbed for sketch-based modeling, demonstrate shape interpolation, and provide comparison to related work.
arXiv commentAccepted to ICLR 2021
References in corpus (14)
- Adam: A Method for Stochastic Optimization
- Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)
- Perspective Transformer Nets: Learning Single-View 3D Object Reconstruction without 3D Supervision
- MarrNet: 3D Shape Reconstruction via 2.5D Sketches
- 3D Sketching using Multi-View Deep Volumetric Prediction
- Learning to Reconstruct Shapes from Unseen Classes
- Robust Watertight Manifold Surface Generation Method for ShapeNet Models
- Learning to Infer and Execute 3D Shape Programs
- PolyGen: An Autoregressive Generative Model of 3D Meshes
- DeepSpline: Data-Driven Reconstruction of Parametric Curves and Surfaces
- 3D-LMNet: Latent Embedding Matching for Accurate and Diverse 3D Point Cloud Reconstruction from a Single Image
- 3D-PRNN: Generating Shape Primitives with Recurrent Neural Networks
- 3DN: 3D Deformation Network
- Better Patch Stitching for Parametric Surface Reconstruction