Change Detection under Global Viewpoint Uncertainty

TL;DR

This paper presents a scalable change detection method for large maps under global viewpoint uncertainty, utilizing a novel scene model, motion prior, and scene retrieval techniques, validated on dynamic urban datasets.

Contribution

It introduces a new approach combining motion priors and scene retrieval with a compact BoW model for change detection under GPS-denied conditions.

Findings

Effective change detection in highly dynamic scenes

Successful retrieval of reference images using BoLCF model

Validated on challenging Malaga dataset

Abstract

This paper addresses the problem of change detection from a novel perspective of long-term map learning. We are particularly interested in designing an approach that can scale to large maps and that can function under global uncertainty in the viewpoint (i.e., GPS-denied situations). Our approach, which utilizes a compact bag-of-words (BoW) scene model, makes several contributions to the problem: 1) Two kinds of prior information are extracted from the view sequence map and used for change detection. Further, we propose a novel type of prior, called motion prior, to predict the relative motions of stationary objects and anomaly ego-motion detection. The proposed prior is also useful for distinguishing stationary from non-stationary objects. 2) A small set of good reference images (e.g., 10) are efficiently retrieved from the view sequence map by employing the recently developed Bag-of-Local-Convolutional-Features (BoLCF) scene model. 3) Change detection is reformulated as a scene retrieval over these reference images to find changed objects using a novel spatial Bag-of-Words (SBoW) scene model. Evaluations conducted of individual techniques and also their combinations on a challenging dataset of highly dynamic scenes in the publicly available Malaga dataset verify their efficacy.