MeshConv3D: Efficient convolution and pooling operators for triangular 3D meshes

TL;DR

MeshConv3D introduces specialized convolution and pooling operators for 3D meshes, enabling effective deep learning directly on irregular mesh data without re-meshing, achieving high classification accuracy efficiently.

Contribution

It presents MeshConv3D, a novel method for applying CNNs directly to 3D meshes with arbitrary topology, using face collapse pooling and specialized convolution operators.

Findings

Achieves comparable or better classification accuracy than existing methods.

Reduces memory footprint and computational load.

Operates directly on meshes without re-meshing.

Abstract

Convolutional neural networks (CNNs) have been pivotal in various 2D image analysis tasks, including computer vision, image indexing and retrieval or semantic classification. Extending CNNs to 3D data such as point clouds and 3D meshes raises significant challenges since the very basic convolution and pooling operators need to be completely re-visited and re-defined in an appropriate manner to tackle irregular connectivity issues. In this paper, we introduce MeshConv3D, a 3D mesh-dedicated methodology integrating specialized convolution and face collapse-based pooling operators. MeshConv3D operates directly on meshes of arbitrary topology, without any need of prior re-meshing/conversion techniques. In order to validate our approach, we have considered a semantic classification task. The experimental results obtained on three distinct benchmark datasets show that the proposed approach makes it possible to achieve equivalent or superior classification results, while minimizing the related memory footprint and computational load.