Learning point embedding for 3D data processing

TL;DR

This paper introduces PE-Net, a novel point cloud processing architecture that learns high-dimensional representations, enabling effective classification and segmentation with adaptability to varying input sizes, achieving state-of-the-art results.

Contribution

PE-Net is the first to encode point clouds into high-dimensional features suitable for standard CNNs, overcoming fixed input size limitations of previous methods.

Findings

PE-Net achieves state-of-the-art accuracy on ModelNet and ShapeNetPart datasets.

The method adapts to varying numbers of input points.

PE-Net outperforms existing point-based approaches in classification and segmentation tasks.

Abstract

Among 2D convolutional networks on point clouds, point-based approaches consume point clouds of fixed size directly. By analysis of PointNet, a pioneer in introducing deep learning into point sets, we reveal that current point-based methods are essentially spatial relationship processing networks. In this paper, we take a different approach. Our architecture, named PE-Net, learns the representation of point clouds in high-dimensional space, and encodes the unordered input points to feature vectors, which standard 2D CNNs can be applied to. The recommended network can adapt to changes in the number of input points which is the limit of current methods. Experiments show that in the tasks of classification and part segmentation, PE-Net achieves the state-of-the-art performance in multiple challenging datasets, such as ModelNet and ShapeNetPart.