X-View: Graph-Based Semantic Multi-View Localization

TL;DR

X-View introduces a semantic graph matching approach for robust multi-view global localization in environments with significant viewpoint changes, outperforming traditional appearance-based methods.

Contribution

The paper presents a novel semantic graph descriptor matching system for multi-view localization, demonstrating effectiveness across diverse datasets and view-point variations.

Findings

Achieves up to 85% localization accuracy in multi-view scenarios.

Outperforms appearance-based methods by up to 10%.

Effective in aerial-to-ground and ground-to-ground localization.

Abstract

Global registration of multi-view robot data is a challenging task. Appearance-based global localization approaches often fail under drastic view-point changes, as representations have limited view-point invariance. This work is based on the idea that human-made environments contain rich semantics which can be used to disambiguate global localization. Here, we present X-View, a Multi-View Semantic Global Localization system. X-View leverages semantic graph descriptor matching for global localization, enabling localization under drastically different view-points. While the approach is general in terms of the semantic input data, we present and evaluate an implementation on visual data. We demonstrate the system in experiments on the publicly available SYNTHIA dataset, on a realistic urban dataset recorded with a simulator, and on real-world StreetView data. Our findings show that X-View is able to globally localize aerial-to-ground, and ground-to-ground robot data of drastically different view-points. Our approach achieves an accuracy of up to 85 % on global localizations in the multi-view case, while the benchmarked baseline appearance-based methods reach up to 75 %.