Rotation-Invariant Local-to-Global Representation Learning for 3D Point Cloud

TL;DR

This paper introduces a rotation-invariant local-to-global representation learning method for 3D point clouds, effectively handling geometric transformations without explicit data augmentation, and achieves state-of-the-art results in recognition and segmentation tasks.

Contribution

The paper presents a novel multi-level graph convolutional neural network that encodes rotation-invariant features from 3D point clouds without data augmentation.

Findings

Achieves state-of-the-art performance on rotation-augmented 3D recognition benchmarks.

Demonstrates robustness of learned representations to geometric variations.

Provides comprehensive analysis through ablative experiments.

Abstract

We propose a local-to-global representation learning algorithm for 3D point cloud data, which is appropriate to handle various geometric transformations, especially rotation, without explicit data augmentation with respect to the transformations. Our model takes advantage of multi-level abstraction based on graph convolutional neural networks, which constructs a descriptor hierarchy to encode rotation-invariant shape information of an input object in a bottom-up manner. The descriptors in each level are obtained from a neural network based on a graph via stochastic sampling of 3D points, which is effective in making the learned representations robust to the variations of input data. The proposed algorithm presents the state-of-the-art performance on the rotation-augmented 3D object recognition and segmentation benchmarks, and we further analyze its characteristics through comprehensive ablative experiments.