Split-and-Fit: Learning B-Reps via Structure-Aware Voronoi Partitioning
- URL: http://arxiv.org/abs/2406.05261v1
- Date: Fri, 7 Jun 2024 21:07:49 GMT
- Title: Split-and-Fit: Learning B-Reps via Structure-Aware Voronoi Partitioning
- Authors: Yilin Liu, Jiale Chen, Shanshan Pan, Daniel Cohen-Or, Hao Zhang, Hui Huang,
- Abstract summary: We introduce a novel method for acquiring boundary representations (B-Reps) of 3D CAD models.
We apply a spatial partitioning to derive a single primitive within each partition.
We show that our network, coined NVD-Net for neural Voronoi diagrams, can effectively learn Voronoi partitions for CAD models from training data.
- Score: 50.684254969269546
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: We introduce a novel method for acquiring boundary representations (B-Reps) of 3D CAD models which involves a two-step process: it first applies a spatial partitioning, referred to as the ``split``, followed by a ``fit`` operation to derive a single primitive within each partition. Specifically, our partitioning aims to produce the classical Voronoi diagram of the set of ground-truth (GT) B-Rep primitives. In contrast to prior B-Rep constructions which were bottom-up, either via direct primitive fitting or point clustering, our Split-and-Fit approach is top-down and structure-aware, since a Voronoi partition explicitly reveals both the number of and the connections between the primitives. We design a neural network to predict the Voronoi diagram from an input point cloud or distance field via a binary classification. We show that our network, coined NVD-Net for neural Voronoi diagrams, can effectively learn Voronoi partitions for CAD models from training data and exhibits superior generalization capabilities. Extensive experiments and evaluation demonstrate that the resulting B-Reps, consisting of parametric surfaces, curves, and vertices, are more plausible than those obtained by existing alternatives, with significant improvements in reconstruction quality. Code will be released on https://github.com/yilinliu77/NVDNet.
Related papers
- Neural Vector Fields: Generalizing Distance Vector Fields by Codebooks
and Zero-Curl Regularization [73.3605319281966]
We propose a novel 3D representation, Neural Vector Fields (NVF), which adopts the explicit learning process to manipulate meshes and implicit unsigned distance function (UDF) representation to break the barriers in resolution and topology.
We evaluate both NVFs on four surface reconstruction scenarios, including watertight vs non-watertight shapes, category-agnostic reconstruction vs category-unseen reconstruction, category-specific, and cross-domain reconstruction.
arXiv Detail & Related papers (2023-09-04T10:42:56Z) - BPNet: B\'ezier Primitive Segmentation on 3D Point Clouds [17.133617027574353]
BPNet is a novel end-to-end deep learning framework to learn B'ezier primitive segmentation on 3D point clouds.
A joint optimization framework is proposed to learn B'ezier primitive segmentation and geometric fitting simultaneously.
Experiments show superior performance over previous work in terms of segmentation, with a substantially faster inference speed.
arXiv Detail & Related papers (2023-07-08T16:46:01Z) - Neural Vector Fields: Implicit Representation by Explicit Learning [63.337294707047036]
We propose a novel 3D representation method, Neural Vector Fields (NVF)
It not only adopts the explicit learning process to manipulate meshes directly, but also the implicit representation of unsigned distance functions (UDFs)
Our method first predicts displacement queries towards the surface and models shapes as text reconstructions.
arXiv Detail & Related papers (2023-03-08T02:36:09Z) - NCP: Neural Correspondence Prior for Effective Unsupervised Shape
Matching [31.61255365182462]
We present Neural Correspondence Prior (NCP), a new paradigm for computing correspondences between 3D shapes.
Our approach is fully unsupervised and can lead to high-quality correspondences even in challenging cases.
We show that NCP is data-efficient, fast, and state-of-the-art results on many tasks.
arXiv Detail & Related papers (2023-01-14T07:22:18Z) - Learning Mesh Representations via Binary Space Partitioning Tree
Networks [28.962866472806812]
We present BSP-Net, a network that learns to represent a 3D shape via convex decomposition without supervision.
The network is trained to reconstruct a shape using a set of convexes obtained from a BSP-tree built over a set of planes, where the planes and convexes are both defined by learned network weights.
The generated meshes are watertight, compact (i.e., low-poly), and well suited to represent sharp geometry.
arXiv Detail & Related papers (2021-06-27T16:37:54Z) - Learning Compositional Shape Priors for Few-Shot 3D Reconstruction [36.40776735291117]
We show that complex encoder-decoder architectures exploit large amounts of per-category data.
We propose three ways to learn a class-specific global shape prior, directly from data.
Experiments on the popular ShapeNet dataset show that our method outperforms a zero-shot baseline by over 40%.
arXiv Detail & Related papers (2021-06-11T14:55:49Z) - Explanation-Guided Training for Cross-Domain Few-Shot Classification [96.12873073444091]
Cross-domain few-shot classification task (CD-FSC) combines few-shot classification with the requirement to generalize across domains represented by datasets.
We introduce a novel training approach for existing FSC models.
We show that explanation-guided training effectively improves the model generalization.
arXiv Detail & Related papers (2020-07-17T07:28:08Z) - PIE-NET: Parametric Inference of Point Cloud Edges [40.27043782820615]
We introduce an end-to-end learnable technique to robustly identify feature edges in 3D point cloud data.
Our deep neural network, coined PIE-NET, is trained for parametric inference of edges.
arXiv Detail & Related papers (2020-07-09T15:35:10Z) - Monocular Human Pose and Shape Reconstruction using Part Differentiable
Rendering [53.16864661460889]
Recent works succeed in regression-based methods which estimate parametric models directly through a deep neural network supervised by 3D ground truth.
In this paper, we introduce body segmentation as critical supervision.
To improve the reconstruction with part segmentation, we propose a part-level differentiable part that enables part-based models to be supervised by part segmentation.
arXiv Detail & Related papers (2020-03-24T14:25:46Z) - Learning Nonparametric Human Mesh Reconstruction from a Single Image
without Ground Truth Meshes [56.27436157101251]
We propose a novel approach to learn human mesh reconstruction without any ground truth meshes.
This is made possible by introducing two new terms into the loss function of a graph convolutional neural network (Graph CNN)
arXiv Detail & Related papers (2020-02-28T20:30:07Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.