Synthetic Data Generation and Deep Learning for the Topological Analysis of 3D Data

TL;DR

This paper explores using deep learning on synthetically generated 3D point cloud data to estimate manifold topology, demonstrating neural networks' ability to learn topological features and outperform traditional methods.

Contribution

It introduces a novel synthetic dataset with topological labels and compares neural network architectures for topological analysis of 3D data.

Findings

Deep learning models can effectively learn topological features from synthetic 3D data.

Neural networks outperform traditional topological data analysis tools in this context.

Synthetic data enables neural networks to perform segmentation-based topological analysis.

Abstract

This research uses deep learning to estimate the topology of manifolds represented by sparse, unordered point cloud scenes in 3D. A new labelled dataset was synthesised to train neural networks and evaluate their ability to estimate the genus of these manifolds. This data used random homeomorphic deformations to provoke the learning of visual topological features. We demonstrate that deep learning models could extract these features and discuss some advantages over existing topological data analysis tools that are based on persistent homology. Semantic segmentation was used to provide additional geometric information in conjunction with topological labels. Common point cloud multi-layer perceptron and transformer networks were both used to compare the viability of these methods. The experimental results of this pilot study support the hypothesis that, with the aid of sophisticated synthetic data generation, neural networks can perform segmentation-based topological data analysis. While our study focused on simulated data, the accuracy achieved suggests a potential for future applications using real data.