Multiresolution Tree Networks for 3D Point Cloud Processing

TL;DR

This paper introduces multiresolution tree-structured networks for efficient 3D point cloud processing, enabling shape classification, generation, and unsupervised learning with improved performance and lower memory usage.

Contribution

It proposes a novel multiresolution tree architecture that enhances 3D shape understanding and generation, outperforming existing methods in classification and shape inference tasks.

Findings

Outperforms existing point-based architectures in shape classification.

Generates higher-quality shapes using a variational autoencoder.

Enables faster training with lower memory footprint.

Abstract

We present multiresolution tree-structured networks to process point clouds for 3D shape understanding and generation tasks. Our network represents a 3D shape as a set of locality-preserving 1D ordered list of points at multiple resolutions. This allows efficient feed-forward processing through 1D convolutions, coarse-to-fine analysis through a multi-grid architecture, and it leads to faster convergence and small memory footprint during training. The proposed tree-structured encoders can be used to classify shapes and outperform existing point-based architectures on shape classification benchmarks, while tree-structured decoders can be used for generating point clouds directly and they outperform existing approaches for image-to-shape inference tasks learned using the ShapeNet dataset. Our model also allows unsupervised learning of point-cloud based shapes by using a variational autoencoder, leading to higher-quality generated shapes.