GeoAdapt: Self-Supervised Test-Time Adaptation in LiDAR Place Recognition Using Geometric Priors

TL;DR

GeoAdapt is a self-supervised method that enhances LiDAR place recognition across different environments by leveraging geometric priors for pseudo-labeling, reducing the need for costly ground truth data.

Contribution

It introduces a novel auxiliary classification head that uses geometric consistency for self-supervised test-time adaptation in LiDAR place recognition.

Findings

Significantly improves recognition performance under domain shifts.

Competitive with fully supervised test-time adaptation methods.

Effective in complex or GPS-deprived environments.

Abstract

LiDAR place recognition approaches based on deep learning suffer from significant performance degradation when there is a shift between the distribution of training and test datasets, often requiring re-training the networks to achieve peak performance. However, obtaining accurate ground truth data for new training data can be prohibitively expensive, especially in complex or GPS-deprived environments. To address this issue we propose GeoAdapt, which introduces a novel auxiliary classification head to generate pseudo-labels for re-training on unseen environments in a self-supervised manner. GeoAdapt uses geometric consistency as a prior to improve the robustness of our generated pseudo-labels against domain shift, improving the performance and reliability of our Test-Time Adaptation approach. Comprehensive experiments show that GeoAdapt significantly boosts place recognition performance across moderate to severe domain shifts, and is competitive with fully supervised test-time adaptation approaches. Our code is available at https://github.com/csiro-robotics/GeoAdapt.