Please use this identifier to cite or link to this item: https://hdl.handle.net/10955/1312
Title: A machine learning system for developing a Human-Robot interface for automatic facial emotions and hand gestures recognition
Authors: Renteria Bustamante, Leonardo Fabian
Pantano, Pietro
Keywords: Fisica matematica
Appendimento automatico
Issue Date: 13-Feb-2017
Series/Report no.: MAT/07;
Abstract: Emotions are an essential part of people's lives because they help us to make decisions, communicate and somehow to understand each other. Additionally, facial emotion perception plays an important role in various fields of psychology, neuroscience, computational intelligence and robotics. In the last years, robots have stopped being used as simple machines dedicated to carry out repetitive, difficult and dangerous jobs. Thanks to rapid technological advances and the emergence of more powerful computers, robotic applications have extended beyond the commercial and industrial scope becoming more popular in the fields of education and medicine. With the development of humanoid and interactive robots such as ASIMO from Honda, AIBO from Sony and currently NAO from Aldebaran, it has emerged a new line of research in robotics. Worldwide, research groups have increased their work on topics such as bipedal locomotion, manipulation, audio-visual systems, human-robot interaction, adaptive control, and learning tasks. This work shows the development of a Machine Learning system to equip the robot NAO with the capacity not only to recognize the people but also to recognize people’s facial expressions and hand gestures. An analysis of the facial symmetry and hand contour geometry was carried out in order to improve the facial expressions and hand gestures recognition rate. In addition, the robot was programmed to express emotions through movement, sound and colors, making the interaction with people more friendly and eye-catching.
Description: Dottorato "Archimede" in Scienze, Comunicazione e Tecnologie, Ciclo XXVIII,a.a. 2015-2016
URI: http://hdl.handle.net/10955/1313
https://doi.org/10.13126/unical.it/dottorati/1312
Appears in Collections:Dipartimento di Fisica - Tesi di Dottorato

Files in This Item:
File Description SizeFormat 
thesis_leonardo_renteria.pdf6,47 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.