We present a novel approach to place recognition well-suited to environments with many dynamic objects--objects that may or may not be present in an agent's subsequent visits. By incorporating an object-detecting preprocessing step, our approach yields high-quality place representations that incorporate object information. Not only does this result in significantly improved place recognition in dynamic environments, it also significantly reduces memory/storage requirements, which may increase the effectiveness of mobile agents with limited resources.
@article{arxiv.2002.04698,
title = {Improving Place Recognition Using Dynamic Object Detection},
author = {Juan Pablo Munoz and Scott Dexter},
journal= {arXiv preprint arXiv:2002.04698},
year = {2020}
}