Enhanced Gradient-Based Local Feature Descriptors by Saliency Map for Egocentric Action Recognition

Zuo, Zheming, Wei, Bo, Chao, Fei, Qu, Yanpeng, Peng, Yonghong and Yang, Longzhi (2019) Enhanced Gradient-Based Local Feature Descriptors by Saliency Map for Egocentric Action Recognition. Applied System Innovation, 2 (1). ISSN 2571-5577

[img]
Preview
Text (Full text)
Zuo et al - Enhanced Gradient-Based Local Feature Descriptors OA.pdf - Published Version
Available under License Creative Commons Attribution 4.0.

Download (8MB) | Preview
Official URL: https://www.mdpi.com/2571-5577/2/1/7

Abstract

Egocentric video analysis is an important tool in healthcare that serves a variety of purposes, such as memory aid systems and physical rehabilitation, and feature extraction is an indispensable process for such analysis. Local feature descriptors have been widely applied due to their simple implementation and reasonable efficiency and performance in applications. This paper proposes an enhanced spatial and temporal local feature descriptor extraction method to boost the performance of action classification. The approach allows local feature descriptors to take advantage of saliency maps, which provide insights into visual attention. The effectiveness of the proposed method was validated and evaluated by a comparative study, whose results demonstrated an improved accuracy of around 2%.

Item Type: Article
Uncontrolled Keywords: saliency map; local feature descriptors; egocentric action recognition; HOG; HMG; HOF; MBH
Subjects: B800 Medical Technology
G400 Computer Science
Department: Faculties > Engineering and Environment > Computer and Information Sciences
Depositing User: Paul Burns
Date Deposited: 19 Feb 2019 11:36
Last Modified: 11 Oct 2019 06:41
URI: http://nrl.northumbria.ac.uk/id/eprint/38109

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year

View more statistics


Policies: NRL Policies | NRL University Deposit Policy | NRL Deposit Licence