Detection-by-Localization: Maintenance-Free Change Object Detector

TL;DR

This paper introduces a novel object-level change detection method that leverages self-localization performance as a likelihood-of-change measure, enabling maintenance-free and modality-fused change detection without relevance scores.

Contribution

It generalizes detection-by-localization to object-level change detection, modeling self-localization as a ranking function and enabling unsupervised multi-modal rank fusion.

Findings

Effective object-level change detection without relevance scores

Generalizes to various self-localization systems and modalities

Enables unsupervised fusion of multi-modal localization results

Abstract

Recent researches demonstrate that self-localization performance is a very useful measure of likelihood-of-change (LoC) for change detection. In this paper, this "detection-by-localization" scheme is studied in a novel generalized task of object-level change detection. In our framework, a given query image is segmented into object-level subimages (termed "scene parts"), which are then converted to subimage-level pixel-wise LoC maps via the detection-by-localization scheme. Our approach models a self-localization system as a ranking function, outputting a ranked list of reference images, without requiring relevance score. Thanks to this new setting, we can generalize our approach to a broad class of self-localization systems. Our ranking based self-localization model allows to fuse self-localization results from different modalities via an unsupervised rank fusion derived from a field of multi-modal information retrieval (MMR).