Research Repository

Visual Place Recognition for Aerial Robotics: Exploring Accuracy-Computation Trade-off for Local Image Descriptors

Ferrarini, Bruno and Waheed, Maria and Waheed, Sania and Ehsan, Shoaib and Milford, Michael and McDonald-Maier, Klaus D (2019) Visual Place Recognition for Aerial Robotics: Exploring Accuracy-Computation Trade-off for Local Image Descriptors. In: 2019 NASA/ESA Conference on Adaptive Hardware and Systems (AHS), 2019-07-22 - 2019-07-24, Colchester.

[img]
Preview
Text
1908.00258v1.pdf - Accepted Version

Download (1MB) | Preview

Abstract

Visual Place Recognition (VPR) is a fundamental yet challenging task for small Unmanned Aerial Vehicle (UAV). The core reasons are the extreme viewpoint changes, and limited computational power onboard a UAV which restricts the applicability of robust but computation intensive state-ofthe-art VPR methods. In this context, a viable approach is to use local image descriptors for performing VPR as these can be computed relatively efficiently without the need of any special hardware, such as a GPU. However, the choice of a local feature descriptor is not trivial and calls for a detailed investigation as there is a trade-off between VPR accuracy and the required computational effort. To fill this research gap, this paper examines the performance of several state-of-the-art local feature descriptors, both from accuracy and computational perspectives, specifically for VPR application utilizing standard aerial datasets. The presented results confirm that a trade-off between accuracy and computational effort is inevitable while executing VPR on resource-constrained hardware.

Item Type: Conference or Workshop Item (Paper)
Additional Information: Published proceedings: 2019 NASA/ESA Conference on Adaptive Hardware and Systems (AHS)
Divisions: Faculty of Science and Health > Computer Science and Electronic Engineering, School of
Depositing User: Elements
Date Deposited: 27 Feb 2020 10:46
Last Modified: 27 Feb 2020 10:47
URI: http://repository.essex.ac.uk/id/eprint/26958

Actions (login required)

View Item View Item