Home // SIGNAL 2018, The Third International Conference on Advances in Signal, Image and Video Processing // View article

Analysis of Emotions From Body Postures Based on Digital Imaging

Authors:
Bruno Barbosa
António Neves
Sandra Soares
Isabel Dimas

Keywords: Pose Estimation; Digital Image; Emotions; Skeleton Detection.

Abstract:
In this paper, we present a state of the art regarding computer vision systems for the detection and classification of human body postures. Although emotions conveyed by human body postures are an important means of socio-communicative functions in several contexts, a surprising lack of systems that enable the recognition and classification of postures for different emotional signatures, is acknowledged. The neglect of such systems is most likely due to the complexity of the emotions reflected in body postures, and to the wide range of variations in postures, particularly when assessing groups of individuals. Despite the existence of several sensors, allowing to obtain images of various types, from color images to thermal images, no one was yet used for this purpose. We propose the use of a recently developed algorithm, which has presented optimal results in several domains, to allow for the recognition of emotions from human body postures.

Pages: 73 to 78

Copyright: Copyright (c) IARIA, 2018

Publication date: May 20, 2018

Published in: conference

ISSN: 2519-8432

ISBN: 978-1-61208-638-5

Location: Nice, France

Dates: from May 20, 2018 to May 24, 2018