PCAN: 3D Attention Map Learning Using Contextual Information for Point Cloud Based Retrieval

TL;DR

This paper introduces PCAN, a novel neural network that leverages contextual information to learn 3D attention maps for improved point cloud retrieval, enhancing feature discrimination.

Contribution

The paper proposes a new Point Contextual Attention Network (PCAN) that predicts point significance based on context, improving local feature aggregation for place recognition.

Findings

Outperforms state-of-the-art methods on benchmark datasets

Effectively highlights task-relevant features in point clouds

Demonstrates robustness across various scenarios

Abstract

Point cloud based retrieval for place recognition is an emerging problem in vision field. The main challenge is how to find an efficient way to encode the local features into a discriminative global descriptor. In this paper, we propose a Point Contextual Attention Network (PCAN), which can predict the significance of each local point feature based on point context. Our network makes it possible to pay more attention to the task-relevent features when aggregating local features. Experiments on various benchmark datasets show that the proposed network can provide outperformance than current state-of-the-art approaches.