Paper ID: 2112.04701

Unsupervised Complementary-aware Multi-process Fusion for Visual Place Recognition

Stephen Hausler, Tobias Fischer, Michael Milford

A recent approach to the Visual Place Recognition (VPR) problem has been to fuse the place recognition estimates of multiple complementary VPR techniques simultaneously. However, selecting the optimal set of techniques to use in a specific deployment environment a-priori is a difficult and unresolved challenge. Further, to the best of our knowledge, no method exists which can select a set of techniques on a frame-by-frame basis in response to image-to-image variations. In this work, we propose an unsupervised algorithm that finds the most robust set of VPR techniques to use in the current deployment environment, on a frame-by-frame basis. The selection of techniques is determined by an analysis of the similarity scores between the current query image and the collection of database images and does not require ground-truth information. We demonstrate our approach on a wide variety of datasets and VPR techniques and show that the proposed dynamic multi-process fusion (Dyn-MPF) has superior VPR performance compared to a variety of challenging competitive methods, some of which are given an unfair advantage through access to the ground-truth information.

Submitted: Dec 9, 2021