On Model-Free Re-ranking for Visual Place Recognition with Deep Learned Local Features

TL;DR

This paper introduces three novel model-free re-ranking methods using deep-learned local features for visual place recognition, demonstrating their effectiveness and robustness in long-term autonomous systems through extensive experiments.

Contribution

The paper proposes three new model-free re-ranking algorithms tailored for deep-learned local features, enhancing long-term visual place recognition performance.

Findings

Methods achieve results comparable to state-of-the-art approaches.

Model-free re-ranking is effective with deep-learned local features.

Robustness to appearance changes is confirmed in experiments.

Abstract

Re-ranking is the second stage of a visual place recognition task, in which the system chooses the best-matching images from a pre-selected subset of candidates. Model-free approaches compute the image pair similarity based on a spatial comparison of corresponding local visual features, eliminating the need for computationally expensive estimation of a model describing transformation between images. The article focuses on model-free re-ranking based on standard local visual features and their applicability in long-term autonomy systems. It introduces three new model-free re-ranking methods that were designed primarily for deep-learned local visual features. These features evince high robustness to various appearance changes, which stands as a crucial property for use with long-term autonomy systems. All the introduced methods were employed in a new visual place recognition system together with the D2-net feature detector (Dusmanu, 2019) and experimentally tested with diverse, challenging public datasets. The obtained results are on par with current state-of-the-art methods, affirming that model-free approaches are a viable and worthwhile path for long-term visual place recognition.