Loop Closure using AnyLoc Visual Place Recognition in DPV-SLAM

TL;DR

This paper introduces an improved loop closure method for DPV-SLAM using AnyLoc, a deep learning-based visual place recognition technique, which enhances robustness and accuracy across diverse environments without manual parameter tuning.

Contribution

It replaces the classical BoVW approach with AnyLoc in DPV-SLAM and introduces an adaptive threshold mechanism for better loop detection performance.

Findings

Significantly improves loop closure accuracy and robustness.

Outperforms original DPV-SLAM in diverse datasets.

Eliminates manual threshold tuning through adaptive mechanism.

Abstract

Loop closure is crucial for maintaining the accuracy and consistency of visual SLAM. We propose a method to improve loop closure performance in DPV-SLAM. Our approach integrates AnyLoc, a learning-based visual place recognition technique, as a replacement for the classical Bag of Visual Words (BoVW) loop detection method. In contrast to BoVW, which relies on handcrafted features, AnyLoc utilizes deep feature representations, enabling more robust image retrieval across diverse viewpoints and lighting conditions. Furthermore, we propose an adaptive mechanism that dynamically adjusts similarity threshold based on environmental conditions, removing the need for manual tuning. Experiments on both indoor and outdoor datasets demonstrate that our method significantly outperforms the original DPV-SLAM in terms of loop closure accuracy and robustness. The proposed method offers a practical and scalable solution for enhancing loop closure performance in modern SLAM systems.