Subdivision-Based Mesh Convolution Networks

Abstract

Convolutional neural networks (CNNs) have made great breakthroughs in 2D computer vision. However, their irregular structure makes it hard to harness the potential of CNNs directly on meshes. A subdivision surface provides a hierarchical multi-resolution structure, in which each face in a closed 2-manifold triangle mesh is exactly adjacent to three faces. Motivated by these two observations, this paper presents SubdivNet, an innovative and versatile CNN framework for 3D triangle meshes with Loop subdivision sequence connectivity. Making an analogy between mesh faces and pixels in a 2D image allows us to present a mesh convolution operator to aggregate local features from nearby faces. By exploiting face neighborhoods, this convolution can support standard 2D convolutional network concepts, e.g. variable kernel size, stride, and dilation. Based on the multi-resolution hierarchy, we make use of pooling layers which uniformly merge four faces into one and an upsampling method which splits one face into four. Thereby, many popular 2D CNN architectures can be easily adapted to process 3D meshes. Meshes with arbitrary connectivity can be remeshed to have Loop subdivision sequence connectivity via self-parameterization, making SubdivNet a general approach. Extensive evaluation and various applications demonstrate SubdivNet's effectiveness and efficiency.

Codes are available in https://github.com/lzhengning/SubdivNet

References in corpus (4)

• PCT: Point cloud transformer
• O-CNN: Octree-based Convolutional Neural Networks for 3D Shape Analysis
• DNF-Net: a Deep Normal Filtering Network for Mesh Denoising
• Deep Geometric Texture Synthesis

Cited by in corpus (10)

• Laplacian2Mesh: Laplacian-Based Mesh Understanding
• Generating Procedural Materials from Text or Image Prompts
• Learning the Geodesic Embedding with Graph Neural Networks
• Bringing Attention to CAD: Boundary Representation Learning via Transformer
• 3D Adversarial Attacks Beyond Point Cloud
• iSeg: Interactive 3D Segmentation via Interactive Attention
• Self-Parameterization Based Multi-Resolution Mesh Convolution Networks
• CageNet: A Meta-Framework for Learning on Wild Meshes
• A 3D mesh convolution-based autoencoder for geometry compression
• Deep 3D Mesh Watermarking with Self-Adaptive Robustness