Cargando…

A hierarchically trained generative network for robust facial symmetrization

Face symmetrization has extensive applications in both medical and academic fields, such as facial disorder diagnosis. Human face possesses an important characteristic, which is as known as symmetry. However, in many scenarios, the perfect symmetry doesn’t exist in human faces, which yields a large...

Descripción completa

Detalles Bibliográficos
Autores principales: Zhang, Shu, Wang, Ting, Peng, Yanjun, Dong, Junyu
Formato: Online Artículo Texto
Lenguaje:English
Publicado: IOS Press 2019
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6598010/
https://www.ncbi.nlm.nih.gov/pubmed/31045541
http://dx.doi.org/10.3233/THC-199021
_version_ 1783430684997582848
author Zhang, Shu
Wang, Ting
Peng, Yanjun
Dong, Junyu
author_facet Zhang, Shu
Wang, Ting
Peng, Yanjun
Dong, Junyu
author_sort Zhang, Shu
collection PubMed
description Face symmetrization has extensive applications in both medical and academic fields, such as facial disorder diagnosis. Human face possesses an important characteristic, which is as known as symmetry. However, in many scenarios, the perfect symmetry doesn’t exist in human faces, which yields a large number of studies around this topic. For example, facial palsy evaluation, facial beauty evaluation based on facial symmetry analysis, and many among others. Currently, there are still very limited researches dedicated for automatic facial symmetrization. Most of the existing studies only utilized their own implantations for facial symmetrization to assist their interdisciplinary academic researches. Limitations thus can be noticed in their methods, such as the requirements for manual interventions. Furthermore, most existing methods utilize facial landmark detection algorithms for automatic facial symmetrization. Though accuracies of the landmark detection algorithms are promising, the uncontrolled conditions in the facial images can still negatively impact the performance of the symmetrical face production. To this end, this paper presents a joint-loss enhanced deep generative network model for automatic facial symmetrization, which is achieved by a full facial image analysis. The joint-loss consists of a pair of adversarial losses and an identity loss. The adversarial losses try to make the generated symmetrical face as realistic as possible, while the identity loss helps to constrain the output to have the same identity of the person in the original input as much as possible. Rather than an end-to-end learning strategy, the proposed model is constructed by a multi-stage training process, which avoids the demand for a large size of the symmetrical face as training data. Experiments are conducted with comparisons with several existing methods based on some of the most popular facial landmark detection algorithms. Competitive results of the proposed method are demonstrated.
format Online
Article
Text
id pubmed-6598010
institution National Center for Biotechnology Information
language English
publishDate 2019
publisher IOS Press
record_format MEDLINE/PubMed
spelling pubmed-65980102019-07-01 A hierarchically trained generative network for robust facial symmetrization Zhang, Shu Wang, Ting Peng, Yanjun Dong, Junyu Technol Health Care Research Article Face symmetrization has extensive applications in both medical and academic fields, such as facial disorder diagnosis. Human face possesses an important characteristic, which is as known as symmetry. However, in many scenarios, the perfect symmetry doesn’t exist in human faces, which yields a large number of studies around this topic. For example, facial palsy evaluation, facial beauty evaluation based on facial symmetry analysis, and many among others. Currently, there are still very limited researches dedicated for automatic facial symmetrization. Most of the existing studies only utilized their own implantations for facial symmetrization to assist their interdisciplinary academic researches. Limitations thus can be noticed in their methods, such as the requirements for manual interventions. Furthermore, most existing methods utilize facial landmark detection algorithms for automatic facial symmetrization. Though accuracies of the landmark detection algorithms are promising, the uncontrolled conditions in the facial images can still negatively impact the performance of the symmetrical face production. To this end, this paper presents a joint-loss enhanced deep generative network model for automatic facial symmetrization, which is achieved by a full facial image analysis. The joint-loss consists of a pair of adversarial losses and an identity loss. The adversarial losses try to make the generated symmetrical face as realistic as possible, while the identity loss helps to constrain the output to have the same identity of the person in the original input as much as possible. Rather than an end-to-end learning strategy, the proposed model is constructed by a multi-stage training process, which avoids the demand for a large size of the symmetrical face as training data. Experiments are conducted with comparisons with several existing methods based on some of the most popular facial landmark detection algorithms. Competitive results of the proposed method are demonstrated. IOS Press 2019-06-18 /pmc/articles/PMC6598010/ /pubmed/31045541 http://dx.doi.org/10.3233/THC-199021 Text en © 2019 – IOS Press and the authors. All rights reserved https://creativecommons.org/licenses/by-nc/4.0/ This article is published online with Open Access and distributed under the terms of the Creative Commons Attribution Non-Commercial License (CC BY-NC 4.0).
spellingShingle Research Article
Zhang, Shu
Wang, Ting
Peng, Yanjun
Dong, Junyu
A hierarchically trained generative network for robust facial symmetrization
title A hierarchically trained generative network for robust facial symmetrization
title_full A hierarchically trained generative network for robust facial symmetrization
title_fullStr A hierarchically trained generative network for robust facial symmetrization
title_full_unstemmed A hierarchically trained generative network for robust facial symmetrization
title_short A hierarchically trained generative network for robust facial symmetrization
title_sort hierarchically trained generative network for robust facial symmetrization
topic Research Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6598010/
https://www.ncbi.nlm.nih.gov/pubmed/31045541
http://dx.doi.org/10.3233/THC-199021
work_keys_str_mv AT zhangshu ahierarchicallytrainedgenerativenetworkforrobustfacialsymmetrization
AT wangting ahierarchicallytrainedgenerativenetworkforrobustfacialsymmetrization
AT pengyanjun ahierarchicallytrainedgenerativenetworkforrobustfacialsymmetrization
AT dongjunyu ahierarchicallytrainedgenerativenetworkforrobustfacialsymmetrization
AT zhangshu hierarchicallytrainedgenerativenetworkforrobustfacialsymmetrization
AT wangting hierarchicallytrainedgenerativenetworkforrobustfacialsymmetrization
AT pengyanjun hierarchicallytrainedgenerativenetworkforrobustfacialsymmetrization
AT dongjunyu hierarchicallytrainedgenerativenetworkforrobustfacialsymmetrization