Home // SPWID 2020, The Sixth International Conference on Smart Portable, Wearable, Implantable and Disability-oriented Devices and Systems // View article

Mood Detection for Improving Lifestyle of Older Adults in Ambient Assisted Living Contexts

Authors:
Andrea Caroppo
Alessandro Leone
Pietro Siciliano

Keywords: Mood Detection; Ambient Assisted Living; Older Adults; Disability; Lifestyle Improvements

Abstract:
Ambient Assisted Living (AAL) has the ambitious goal of improving the lifestyle or quality of life of elderly or vulnerable people through the use of technology. In this research, area considerable efforts have been made for the design and development of automatic solutions for the recognition of mood trough patterns of facial expressions, even using low cost and commercial vision sensors. However, there are still some open issues to be faced like the age or any situation of disability of the observed subject, the pose of the face and the environment illumination conditions. A lot of progress has been made in this topic with the emergence of deep learning methods. In the proposed work, the performance of two recent deep convolutional neural networks models are evaluated on the CIFE and FER-2013 datasets that include also facial expressions of older adults performed in uncontrolled conditions. A thorough session of experiments focused on the concept of “Transfer Learning” was carried out. The results obtained demonstrated that both deep architectures reach levels of accuracy higher than 67.8% grouping expressions of older adults into 3 categories: positive, negative and neutral. The latter classification may be sufficient for a mood detection module that could be the input of a future e-coaching platform to be integrated in the AAL context.

Pages: 1 to 6

Copyright: Copyright (c) IARIA, 2020

Publication date: September 27, 2020

Published in: conference

ISSN: 2519-8440

ISBN: 978-1-61208-809-9

Location: Lisbon, Portugal

Dates: from September 27, 2020 to October 1, 2020