UV-Net: Learning from Boundary Representations
This work addresses a problem for industries using CAD modeling by enabling machine learning on complex B-rep data, though it appears incremental as it builds on existing image and graph convolutional networks.
The authors tackled the challenge of applying machine learning to Boundary representation (B-rep) data from 3D CAD models by introducing UV-Net, a neural network architecture that operates directly on B-reps and outperforms alternative 3D shape representations like point clouds, voxels, and meshes on supervised and unsupervised tasks across five datasets.
We introduce UV-Net, a novel neural network architecture and representation designed to operate directly on Boundary representation (B-rep) data from 3D CAD models. The B-rep format is widely used in the design, simulation and manufacturing industries to enable sophisticated and precise CAD modeling operations. However, B-rep data presents some unique challenges when used with modern machine learning due to the complexity of the data structure and its support for both continuous non-Euclidean geometric entities and discrete topological entities. In this paper, we propose a unified representation for B-rep data that exploits the U and V parameter domain of curves and surfaces to model geometry, and an adjacency graph to explicitly model topology. This leads to a unique and efficient network architecture, UV-Net, that couples image and graph convolutional neural networks in a compute and memory-efficient manner. To aid in future research we present a synthetic labelled B-rep dataset, SolidLetters, derived from human designed fonts with variations in both geometry and topology. Finally we demonstrate that UV-Net can generalize to supervised and unsupervised tasks on five datasets, while outperforming alternate 3D shape representations such as point clouds, voxels, and meshes.