Point Cloud Recognition with Position-to-Structure Attention Transformers

TL;DR

This paper introduces PS-Former, a Transformer-based model for 3D point cloud recognition that automatically extracts features and enriches structural information without extensive feature engineering.

Contribution

The novel PS-Former architecture integrates a learnable condensation layer and a Position-to-Structure Attention mechanism for improved 3D point cloud analysis.

Findings

Competitive results on classification, part segmentation, and scene segmentation tasks.

Less reliance on heuristic feature engineering.

Effective structural information enrichment.

Abstract

In this paper, we present Position-to-Structure Attention Transformers (PS-Former), a Transformer-based algorithm for 3D point cloud recognition. PS-Former deals with the challenge in 3D point cloud representation where points are not positioned in a fixed grid structure and have limited feature description (only 3D coordinates ($x, y, z$) for scattered points). Existing Transformer-based architectures in this domain often require a pre-specified feature engineering step to extract point features. Here, we introduce two new aspects in PS-Former: 1) a learnable condensation layer that performs point downsampling and feature extraction; and 2) a Position-to-Structure Attention mechanism that recursively enriches the structural information with the position attention branch. Compared with the competing methods, while being generic with less heuristics feature designs, PS-Former demonstrates competitive experimental results on three 3D point cloud tasks including classification, part segmentation, and scene segmentation.