AirLoop: Lifelong Loop Closure Detection

TL;DR

AirLoop introduces a lifelong learning approach for loop closure detection in SLAM systems, enabling incremental updates without catastrophic forgetting, demonstrated effectively on multiple datasets.

Contribution

The paper presents AirLoop, a novel lifelong learning method for deep loop closure detection that maintains performance over time as new data is incorporated.

Findings

Effective reduction of catastrophic forgetting in loop closure models

Successful incremental learning demonstrated on diverse datasets

First application of lifelong learning to deep loop closure detection

Abstract

Loop closure detection is an important building block that ensures the accuracy and robustness of simultaneous localization and mapping (SLAM) systems. Due to their generalization ability, CNN-based approaches have received increasing attention. Although they normally benefit from training on datasets that are diverse and reflective of the environments, new environments often emerge after the model is deployed. It is therefore desirable to incorporate the data newly collected during operation for incremental learning. Nevertheless, simply finetuning the model on new data is infeasible since it may cause the model's performance on previously learned data to degrade over time, which is also known as the problem of catastrophic forgetting. In this paper, we present AirLoop, a method that leverages techniques from lifelong learning to minimize forgetting when training loop closure detection models incrementally. We experimentally demonstrate the effectiveness of AirLoop on TartanAir, Nordland, and RobotCar datasets. To the best of our knowledge, AirLoop is one of the first works to achieve lifelong learning of deep loop closure detectors.