Neural Marching Cubes

Abstract

We introduce Neural Marching Cubes (NMC), a data-driven approach for extracting a triangle mesh from a discretized implicit field. Classical MC is defined by coarse tessellation templates isolated to individual cubes. While more refined tessellations have been proposed, they all make heuristic assumptions, such as trilinearity, when determining the vertex positions and local mesh topologies in each cube. In principle, none of these approaches can reconstruct geometric features that reveal coherence or dependencies between nearby cubes (e.g., a sharp edge), as such information is unaccounted for, resulting in poor estimates of the true underlying implicit field. To tackle these challenges, we re-cast MC from a deep learning perspective, by designing tessellation templates more apt at preserving geometric features, and learning the vertex positions and mesh topologies from training meshes, to account for contextual information from nearby cubes. We develop a compact per-cube parameterization to represent the output triangle mesh, while being compatible with neural processing, so that a simple 3D convolutional network can be employed for the training. We show that all topological cases in each cube that are applicable to our design can be easily derived using our representation, and the resulting tessellations can also be obtained naturally and efficiently by following a few design guidelines. In addition, our network learns local features with limited receptive fields, hence it generalizes well to new shapes and new datasets. We evaluate our neural MC approach by quantitative and qualitative comparisons to all well-known MC variants. In particular, we demonstrate the ability of our network to recover sharp features such as edges and corners, a long-standing issue of MC and its variants. Our network also reconstructs local mesh topologies more accurately than previous approaches.

arXiv commentAccepted to SIGGRAPH Asia 2021. Code: https://github.com/czq142857/NMC, supplementary material: https://www.sfu.ca/~zhiqinc/nmc/NMC_supplementary.zip

References in corpus (9)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• scikit-image: Image processing in Python
• Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling
• Learning Representations and Generative Models for 3D Point Clouds
• Thingi10K: A Dataset of 10,000 3D-Printing Models
• Learning Deformable Tetrahedral Meshes for 3D Reconstruction
• LOGAN: Unpaired Shape Transform in Latent Overcomplete Space
• SDM-NET: Deep Generative Network for Structured Deformable Mesh
• Analytic Marching: An Analytic Meshing Solution from Deep Implicit Surface Networks

Cited by in corpus (12)

• Neural Dual Contouring
• Iterative Poisson Surface Reconstruction (iPSR) for Unoriented Points
• Robust Zero Level-Set Extraction from Unsigned Distance Fields Based on Double Covering
• Implicit Conversion of Manifold B-Rep Solids by Neural Halfspace Representation
• LFS-Aware Surface Reconstruction from Unoriented 3D Point Clouds
• NeuralMeshing: Differentiable Meshing of Implicit Neural Representations
• Occupancy-Based Dual Contouring
• McGrids: Monte Carlo-Driven Adaptive Grids for Iso-Surface Extraction
• Estimation and Visualization of Isosurface Uncertainty from Linear and High-Order Interpolation Methods
• Local Surface Parameterizations via Smoothed Geodesic Splines
• PaMO: Parallel Mesh Optimization for Intersection-Free Low-Poly Modeling on the GPU
• Point2Quad: Generating Quad Meshes from Point Clouds via Face Prediction