Register assisted aggregation for Visual Place Recognition

TL;DR

This paper introduces a register-assisted feature aggregation method for Visual Place Recognition that improves the discrimination of place features by separating stable from unstable features, outperforming existing methods.

Contribution

A novel register-based approach to enhance feature aggregation in VPR, effectively distinguishing useful features and improving recognition accuracy.

Findings

Registers help separate stable from unstable features.

The method outperforms state-of-the-art VPR techniques.

Experimental results validate the effectiveness of the approach.

Abstract

Visual Place Recognition (VPR) refers to the process of using computer vision to recognize the position of the current query image. Due to the significant changes in appearance caused by season, lighting, and time spans between query images and database images for retrieval, these differences increase the difficulty of place recognition. Previous methods often discarded useless features (such as sky, road, vehicles) while uncontrolled discarding features that help improve recognition accuracy (such as buildings, trees). To preserve these useful features, we propose a new feature aggregation method to address this issue. Specifically, in order to obtain global and local features that contain discriminative place information, we added some registers on top of the original image tokens to assist in model training. After reallocating attention weights, these registers were discarded. The experimental results show that these registers surprisingly separate unstable features from the original image representation and outperform state-of-the-art methods.