Bradford Scholars -how to deposit your paper
Overview
Copyright check• Check if your publisher allows submission to a repository.• Use the Sherpa RoMEO database if you are not sure about your publisher's position or email openaccess@bradford.ac.uk.
Automatic gender classification has become a topic of great interest to the visual computing research community in recent times. This is due to the fact that computer-based automatic gender recognition has multiple applications including, but not limited to, face perception, age, ethnicity, identity analysis, video surveillance and smart human computer interaction. In this paper, we discuss a machine learning approach for efficient identification of gender purely from the dynamics of a person's smile. Thus, we show that the complex dynamics of a smile on someone's face bear much relation to the person's gender. To do this, we first formulate a computational framework that captures the dynamic characteristics of a smile. Our dynamic framework measures changes in the face during a smile using a set of spatial features on the overall face, the area of the mouth, the geometric flow around prominent parts of the face and a set of intrinsic features based on the dynamic geometry of the face. This enables us to extract 210 distinct dynamic smile parameters which form as the contributing features for machine learning. For machine classification, we have utilised both the Support Vector Machine and the k-Nearest Neighbour algorithms. To verify the accuracy of our approach, we have tested our algorithms on two databases, namely the CK+ and the MUG, consisting of a total of 109 subjects. As a result, using the k-NN algorithm, along with tenfold cross validation, for example, we achieve an accurate gender classification rate of over 85%. Hence, through the methodology we present here, we establish proof of the existence of strong indicators of gender dimorphism, purely in the dynamics of a person's smile.
translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.