Pointwise Rotation-Invariant Network with Adaptive Sampling and 3D Spherical Voxel Convolution

Abstract

Point cloud analysis without pose priors is very challenging in real applications, as the orientations of point clouds are often unknown. In this paper, we propose a brand new point-set learning framework PRIN, namely, Pointwise Rotation-Invariant Network, focusing on rotation-invariant feature extraction in point clouds analysis. We construct spherical signals by Density Aware Adaptive Sampling to deal with distorted point distributions in spherical space. In addition, we propose Spherical Voxel Convolution and Point Re-sampling to extract rotation-invariant features for each point. Our network can be applied to tasks ranging from object classification, part segmentation, to 3D feature matching and label alignment. We show that, on the dataset with randomly rotated point clouds, PRIN demonstrates better performance than state-of-the-art methods without any data augmentation. We also provide theoretical analysis for the rotation-invariance achieved by our methods.

8 pages, to appear on AAAI 2020

References in corpus (9)

• Geometric deep learning: going beyond Euclidean data
• PCPNET: Learning Local Shape Properties from Raw Point Clouds
• PointSIFT: A SIFT-like Network Module for 3D Point Cloud Semantic Segmentation
• Spherical CNNs
• Linked Dynamic Graph CNN: Learning on Point Cloud via Linking Hierarchical Features
• PointCNN: Convolution On $\mathcal{X}$-Transformed Points
• SO-Net: Self-Organizing Network for Point Cloud Analysis
• SyncSpecCNN: Synchronized Spectral CNN for 3D Shape Segmentation
• Geometric deep learning on graphs and manifolds using mixture model CNNs

Cited by in corpus (10)

• Detailed 2D-3D Joint Representation for Human-Object Interaction
• KeypointNet: A Large-scale 3D Keypoint Dataset Aggregated from Numerous Human Annotations
• 3D-Rotation-Equivariant Quaternion Neural Networks
• TubeTK: Adopting Tubes to Track Multi-Object in a One-Step Training Model
• Disentangling Geometric Deformation Spaces in Generative Latent Shape Models
• SRINet: Learning Strictly Rotation-Invariant Representations for Point Cloud Classification and Segmentation
• RobustPointSet: A Dataset for Benchmarking Robustness of Point Cloud Classifiers
• Learning Rotation-Invariant Representations of Point Clouds Using Aligned Edge Convolutional Neural Networks
• Multi-scale Receptive Fields Graph Attention Network for Point Cloud Classification
• REST: Performance Improvement of a Black Box Model via RL-based Spatial Transformation