Self-localization Using Visual Experience Across Domains

TL;DR

This paper presents a novel approach for robot self-localization across different domains by mining cross-domain visual experiences and matching visual patterns, outperforming traditional bag-of-words methods in varied conditions.

Contribution

The study introduces a cross-domain visual experience mining method that improves scene retrieval for robot localization across different seasons and routes.

Findings

Achieves promising localization results across domains

Uses image-to-class distance and spatial pyramid matching effectively

Outperforms traditional bag-of-words approaches in cross-domain scenarios

Abstract

In this study, we aim to solve the single-view robot self-localization problem by using visual experience across domains. Although the bag-of-words method constitutes a popular approach to single-view localization, it fails badly when it's visual vocabulary is learned and tested in different domains. Further, we are interested in using a cross-domain setting, in which the visual vocabulary is learned in different seasons and routes from the input query/database scenes. Our strategy is to mine a cross-domain visual experience, a library of raw visual images collected in different domains, to discover the relevant visual patterns that effectively explain the input scene, and use them for scene retrieval. In particular, we show that the appearance and the pose of the mined visual patterns of a query scene can be efficiently and discriminatively matched against those of the database scenes by employing image-to-class distance and spatial pyramid matching. Experimental results obtained using a novel cross-domain dataset show that our system achieves promising results despite our visual vocabulary being learned and tested in different domains.