Adaptive Thresholding for Visual Place Recognition using Negative Gaussian Mixture Statistics

TL;DR

This paper proposes an automatic thresholding method for visual place recognition that uses negative Gaussian mixture statistics to improve match selection across diverse visual scenarios.

Contribution

It introduces a novel thresholding approach based on negative Gaussian mixture models, enhancing robustness in VPR without manual parameter tuning.

Findings

Thresholds derived from negative Gaussian mixtures improve recognition accuracy.

The method is effective across multiple image databases and descriptors.

Automatic thresholding reduces manual effort and enhances robustness.

Abstract

Visual place recognition (VPR) is an important component technology for camera-based mapping and navigation applications. This is a challenging problem because images of the same place may appear quite different for reasons including seasonal changes, weather illumination, structural changes to the environment, as well as transient pedestrian or vehicle traffic. Papers focusing on generating image descriptors for VPR report their results using metrics such as recall@K and ROC curves. However, for a robot implementation, determining which matches are sufficiently good is often reduced to a manually set threshold. And it is difficult to manually select a threshold that will work for a variety of visual scenarios. This paper addresses the problem of automatically selecting a threshold for VPR by looking at the 'negative' Gaussian mixture statistics for a place - image statistics indicating not this place. We show that this approach can be used to select thresholds that work well for a variety of image databases and image descriptors.