CoHOG: A Light-Weight, Compute-Efficient, and Training-Free Visual Place Recognition Technique for Changing Environments

Mubariz Zaffar,Klaus Mcdonald-Maier,Shoaib Ehsan,Michael Milford

doi:10.1109/lra.2020.2969917

Abstract

This letter presents a novel, compute-efficient and training-free approach based on Histogram-of-Oriented-Gradients (HOG) descriptor for achieving state-of-the-art performance-per-compute-unit in Visual Place Recognition (VPR). The inspiration for this approach (namely CoHOG) is based on the convolutional scanning and regions-based feature extraction employed by Convolutional Neural Networks (CNNs). By using image entropy to extract regions-of-interest (ROI) and regional-convolutional descriptor matching, our technique performs successful place recognition in changing environments. We use viewpoint- and appearance-variant public VPR datasets to report this matching performance, at lower RAM commitment, zero training requirements and 20 times lesser feature encoding time compared to state-of-the-art neural networks. We also discuss the image retrieval time of CoHOG and the effect of CoHOG's parametric variation on its place matching performance and encoding time.

Highlights

F OR A ROBOT to operate autonomously, it needs to be able to remember previously visited places
We propose a novel technique based on handcrafted feature descriptors delivering state-of-the-art Visual Place Recognition (VPR) performance with no training requirements compared to Convolutional Neural Networks (CNNs)
This section first discusses the experimental setup used in our analysis including the VPR datasets, VPR techniques and evaluation metric used for assessing CoHOG’s performance

Summary

Introduction

F OR A ROBOT to operate autonomously, it needs to be able to remember previously visited places. This ability to remember places has been discussed and widely researched (surveyed by Lowry et al [1]) as the sub-domain of visualSLAM (Simultaneous Localization and Mapping), namely Visual Place Recognition (VPR). VPR is a well-defined, albeit a highly challenging problem since places change their appearance rapidly due to varying viewpoints and conditions. Texture-less and low-informative scenes pose difficulty to place matching. The task of a VPR system is to retrieve the best matched image of the same place

Methods

Results

Conclusion