RIConv++: Effective Rotation Invariant Convolutions for 3D Point Clouds Deep Learning

TL;DR

This paper introduces RIConv++, a novel rotation-invariant convolution operator for 3D point clouds that significantly improves feature distinction and achieves state-of-the-art results in rotation-robust 3D scene understanding tasks.

Contribution

The paper proposes a simple yet effective rotation-invariant convolution method that enhances feature distinction by considering local relationships, improving 3D point cloud analysis under rotations.

Findings

Achieves state-of-the-art accuracy on synthetic and real-world datasets.

Effectively captures local and global context in point cloud analysis.

Demonstrates robustness to challenging rotations.

Abstract

3D point clouds deep learning is a promising field of research that allows a neural network to learn features of point clouds directly, making it a robust tool for solving 3D scene understanding tasks. While recent works show that point cloud convolutions can be invariant to translation and point permutation, investigations of the rotation invariance property for point cloud convolution has been so far scarce. Some existing methods perform point cloud convolutions with rotation-invariant features, existing methods generally do not perform as well as translation-invariant only counterpart. In this work, we argue that a key reason is that compared to point coordinates, rotation-invariant features consumed by point cloud convolution are not as distinctive. To address this problem, we propose a simple yet effective convolution operator that enhances feature distinction by designing powerful rotation invariant features from the local regions. We consider the relationship between the point of interest and its neighbors as well as the internal relationship of the neighbors to largely improve the feature descriptiveness. Our network architecture can capture both local and global context by simply tuning the neighborhood size in each convolution layer. We conduct several experiments on synthetic and real-world point cloud classifications, part segmentation, and shape retrieval to evaluate our method, which achieves the state-of-the-art accuracy under challenging rotations.