DRINet: A Dual-Representation Iterative Learning Network for Point Cloud Segmentation

TL;DR

DRINet introduces a dual-representation iterative learning architecture for point cloud segmentation, effectively propagating features between point and voxel representations, achieving state-of-the-art results with high efficiency.

Contribution

The paper proposes a flexible dual-representation learning network with iterative feature transfer and a multi-scale pooling layer, enhancing segmentation accuracy and efficiency for large-scale point clouds.

Findings

Achieves state-of-the-art segmentation accuracy on multiple datasets.

Operates in real-time with 62ms per frame for outdoor scenarios.

Outperforms existing methods in large-scale point cloud processing.

Abstract

We present a novel and flexible architecture for point cloud segmentation with dual-representation iterative learning. In point cloud processing, different representations have their own pros and cons. Thus, finding suitable ways to represent point cloud data structure while keeping its own internal physical property such as permutation and scale-invariant is a fundamental problem. Therefore, we propose our work, DRINet, which serves as the basic network structure for dual-representation learning with great flexibility at feature transferring and less computation cost, especially for large-scale point clouds. DRINet mainly consists of two modules called Sparse Point-Voxel Feature Extraction and Sparse Voxel-Point Feature Extraction. By utilizing these two modules iteratively, features can be propagated between two different representations. We further propose a novel multi-scale pooling layer for pointwise locality learning to improve context information propagation. Our network achieves state-of-the-art results for point cloud classification and segmentation tasks on several datasets while maintaining high runtime efficiency. For large-scale outdoor scenarios, our method outperforms state-of-the-art methods with a real-time inference speed of 62ms per frame.