CASSPR: Cross Attention Single Scan Place Recognition

TL;DR

CASSPR introduces a novel cross attention transformer method that fuses point-based and voxel-based LiDAR data to improve fine-grained place recognition accuracy in autonomous navigation.

Contribution

It is the first to combine point and voxel approaches with cross attention transformers for LiDAR place recognition, enhancing fine-grained matching capabilities.

Findings

Surpasses state-of-the-art on multiple datasets

Achieves AR@1 of 85.6% on TUM dataset

Significantly improves fine-grained matching accuracy

Abstract

Place recognition based on point clouds (LiDAR) is an important component for autonomous robots or self-driving vehicles. Current SOTA performance is achieved on accumulated LiDAR submaps using either point-based or voxel-based structures. While voxel-based approaches nicely integrate spatial context across multiple scales, they do not exhibit the local precision of point-based methods. As a result, existing methods struggle with fine-grained matching of subtle geometric features in sparse single-shot Li- DAR scans. To overcome these limitations, we propose CASSPR as a method to fuse point-based and voxel-based approaches using cross attention transformers. CASSPR leverages a sparse voxel branch for extracting and aggregating information at lower resolution and a point-wise branch for obtaining fine-grained local information. CASSPR uses queries from one branch to try to match structures in the other branch, ensuring that both extract self-contained descriptors of the point cloud (rather than one branch dominating), but using both to inform the output global descriptor of the point cloud. Extensive experiments show that CASSPR surpasses the state-of-the-art by a large margin on several datasets (Oxford RobotCar, TUM, USyd). For instance, it achieves AR@1 of 85.6% on the TUM dataset, surpassing the strongest prior model by ~15%. Our code is publicly available.