Space-time flexible kernel for recognizing activities from wearable cameras
- PDF / 1,493,324 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 90 Downloads / 254 Views
INDUSTRIAL AND COMMERCIAL APPLICATION
Space‑time flexible kernel for recognizing activities from wearable cameras Mario Rodriguez1 · Carlos Orrite2 · Carlos Medrano3 Received: 19 February 2018 / Accepted: 9 November 2020 © Springer-Verlag London Ltd., part of Springer Nature 2020
Abstract Recognizing activities of daily living is useful for ambient assisted living. In this regard, the use of wearable cameras is a promising technology. In this paper, we propose a novel approach for recognizing activities of daily living using egocentric viewpoint video clips. First, in every frame, the appearing objects are detected and labelled depending if they are being used or not by the subject. Later, the video clip is divided into spatiotemporal bins created with an object-centric cut. Finally, a support vector machine classifier is computed using a spatiotemporal flexible kernel between video clips. The validity of the proposed method has been proved by conducting experiments in the ADL dataset. Results confirm the suitability of using the space-time location of objects as information for the classification of activities using an egocentric viewpoint. Keywords Activities of daily living · Ambient assisted living · Wearable cameras · Activity recognition
1 Introduction Thanks to advances in medicine, life expectancy is increasing and so is the number of dependent people, including not only the elderly but also the disabled ones. The number of people aged 65 or over in Europe and the USA will almost double between 2015 and 2060 [1, 2]. Furthermore, the total fertility rate (TFR) is low in many developed countries. The Statistical Office of the European Communities (EUROSTAT) projects that, by 2060, the ratio between working and retired people will have passed from four-to-one to twoto-one in the EU. The increase in dependent people and the, at least, stagnation of the TFR make it impossible to rely on younger generations for future assistance of the dependent population, and this quite certain future makes the development in assisted living technologies a very important issue. * Carlos Orrite [email protected] Mario Rodriguez [email protected] Carlos Medrano [email protected] 1
ITCL, Burgos, Spain
2
CVLab, I3A, Zaragoza University, Zaragoza, Spain
3
EduQTech, IIS, Zaragoza University, Teruel, Spain
Ambient assisted living (AAL) systems aim at improving the quality of life and supporting independent and healthy living of older or/and impaired people by using information and communication technologies at home, at the workplace and in public spaces. AAL environments are embedded with a variety of sensors, either located in the environment or worn by the user, that acquire data on the state of both the environment and the individual and/or allow person-environment interaction. These data are processed using more or less advanced intelligent systems in order to provide services, such as monitoring of daily living activities, prevention and management of chronic conditions, frailty detection and mobility a
Data Loading...
