Quaternion Product Units for Deep Learning on 3D Rotation Groups

Abstract

We propose a novel quaternion product unit (QPU) to represent data on 3D rotation groups. The QPU leverages quaternion algebra and the law of 3D rotation group, representing 3D rotation data as quaternions and merging them via a weighted chain of Hamilton products. We prove that the representations derived by the proposed QPU can be disentangled into "rotation-invariant" features and "rotation-equivariant" features, respectively, which supports the rationality and the efficiency of the QPU in theory. We design quaternion neural networks based on our QPUs and make our models compatible with existing deep learning models. Experiments on both synthetic and real-world data show that the proposed QPU is beneficial for the learning tasks requiring rotation robustness.

arXiv commentCVPR 2020

References in corpus (8)

• Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition
• Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point clouds
• NTU RGB+D: A Large Scale Dataset for 3D Human Activity Analysis
• 3D Steerable CNNs: Learning Rotationally Equivariant Features in Volumetric Data
• QuaterNet: A Quaternion-based Recurrent Model for Human Motion
• Spherical CNNs
• An Attention Enhanced Graph Convolutional LSTM Network for Skeleton-Based Action Recognition
• First-Person Hand Action Benchmark with RGB-D Videos and 3D Hand Pose Annotations

Cited by in corpus (1)

• RISA-Net: Rotation-Invariant Structure-Aware Network for Fine-Grained 3D Shape Retrieval