ChartPointFlow for Topology-Aware 3D Point Cloud Generation

TL;DR

ChartPointFlow is a novel flow-based generative model for 3D point clouds that preserves topological structures and enables unsupervised segmentation by using multiple labels and chart-like mappings.

Contribution

It introduces a topology-aware generative model with multiple latent labels and chart-based mappings for improved 3D point cloud generation and segmentation.

Findings

Achieves state-of-the-art generation and reconstruction performance.

Effectively preserves topological structures in generated point clouds.

Enables unsupervised segmentation of 3D objects into semantic subparts.

Abstract

A point cloud serves as a representation of the surface of a three-dimensional (3D) shape. Deep generative models have been adapted to model their variations typically using a map from a ball-like set of latent variables. However, previous approaches did not pay much attention to the topological structure of a point cloud, despite that a continuous map cannot express the varying numbers of holes and intersections. Moreover, a point cloud is often composed of multiple subparts, and it is also difficult to express. In this study, we propose ChartPointFlow, a flow-based generative model with multiple latent labels for 3D point clouds. Each label is assigned to points in an unsupervised manner. Then, a map conditioned on a label is assigned to a continuous subset of a point cloud, similar to a chart of a manifold. This enables our proposed model to preserve the topological structure with clear boundaries, whereas previous approaches tend to generate blurry point clouds and fail to generate holes. The experimental results demonstrate that ChartPointFlow achieves state-of-the-art performance in terms of generation and reconstruction compared with other point cloud generators. Moreover, ChartPointFlow divides an object into semantic subparts using charts, and it demonstrates superior performance in case of unsupervised segmentation.