Fault-Diagnosing Monocular-SLAM for Scale-Aware Change Detection

In this paper, we present a new fault diagnosis (FD) -based approach for image change detection (ICD) that can detect significant changes as inconsistencies between different visual experiences of monocular-SLAM. Unlike classical change detection approaches such as pairwise image comparison (PC) and anomaly detection (AD), neither the memorization of each map image nor the maintenance of up-to-date placespecific anomaly detectors are required in this FD approach. A significant challenge that is encountered when incorporating different visual experiences into FD involves dealing with the varying scales of changed objects. To address this issue, we reconsider the bag-of-words (BoW) image representation, and focus on the state-of-the-art BoW-based SLAM paradigm. As a key advantage, the local feature -based representation enables to re-organize the BoW into any different scales without modifying the database entries (i.e., the map). Furthermore, it enables to control discriminative power and expected inconsistencies of local features. Experiments on challenging cross-season ICD using publicly available NCLT dataset, and comparison against state-of-the-art ICD algorithms validate the efficacy of the proposed FD approach with/without combining AD and/or PC.