GAN-based Face Mask Removal using Facial Landmarks and Pixel Errors in Masked Region

Hitoshi Yoshihashi; Naoto Ienaga; Maki Sugimoto

doi:10.5220/0010827500003124

GAN-based Face Mask Removal using Facial Landmarks and Pixel Errors in Masked Region

Hitoshi Yoshihashi
, Naoto Ienaga
, Maki Sugimoto

情報工学科

研究成果: Conference article › 査読

4 被引用数 (Scopus)

抄録

In 2020 and beyond, the opportunities to communicate with others while wearing a face mask have increased. A mask hides the mouth and facial muscles, making it difficult to convey facial expressions to others. In this study, we propose to use generative adversarial networks (GAN) to complete the facial region hidden by the mask. We defined custom loss functions that focus on the errors of the feature point coordinates of the face and the pixels in the masked region. As a result, we were able to generate images with higher quality than existing methods.

本文言語	English
ページ（範囲）	125-133
ページ数	9
ジャーナル	Proceedings of the International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
巻	5
DOI	https://doi.org/10.5220/0010827500003124
出版ステータス	Published - 2022
イベント	17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2022 - Virtual, Online 継続期間: 2022 2月 6 → 2022 2月 8

ASJC Scopus subject areas

コンピュータグラフィックスおよびコンピュータ支援設計
コンピュータビジョンおよびパターン認識
人間とコンピュータの相互作用

文献へのアクセス

10.5220/0010827500003124

他のファイルとリンク

引用スタイル

@article{eb274732728e419fb9409d3256e268b0,

title = "GAN-based Face Mask Removal using Facial Landmarks and Pixel Errors in Masked Region",

abstract = "In 2020 and beyond, the opportunities to communicate with others while wearing a face mask have increased. A mask hides the mouth and facial muscles, making it difficult to convey facial expressions to others. In this study, we propose to use generative adversarial networks (GAN) to complete the facial region hidden by the mask. We defined custom loss functions that focus on the errors of the feature point coordinates of the face and the pixels in the masked region. As a result, we were able to generate images with higher quality than existing methods.",

keywords = "COVID-19, Face Completion, Face Mask, Generative Adversarial Networks, Inpainting",

author = "Hitoshi Yoshihashi and Naoto Ienaga and Maki Sugimoto",

note = "Publisher Copyright: {\textcopyright} 2022 by SCITEPRESS-Science and Technology Publications, Lda.; 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2022 ; Conference date: 06-02-2022 Through 08-02-2022",

year = "2022",

doi = "10.5220/0010827500003124",

language = "English",

volume = "5",

pages = "125--133",

journal = "Proceedings of the International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications",

issn = "2184-5921",

publisher = "Science and Technology Publications, Lda",

}

TY - JOUR

T1 - GAN-based Face Mask Removal using Facial Landmarks and Pixel Errors in Masked Region

AU - Yoshihashi, Hitoshi

AU - Ienaga, Naoto

AU - Sugimoto, Maki

PY - 2022

Y1 - 2022

N2 - In 2020 and beyond, the opportunities to communicate with others while wearing a face mask have increased. A mask hides the mouth and facial muscles, making it difficult to convey facial expressions to others. In this study, we propose to use generative adversarial networks (GAN) to complete the facial region hidden by the mask. We defined custom loss functions that focus on the errors of the feature point coordinates of the face and the pixels in the masked region. As a result, we were able to generate images with higher quality than existing methods.

AB - In 2020 and beyond, the opportunities to communicate with others while wearing a face mask have increased. A mask hides the mouth and facial muscles, making it difficult to convey facial expressions to others. In this study, we propose to use generative adversarial networks (GAN) to complete the facial region hidden by the mask. We defined custom loss functions that focus on the errors of the feature point coordinates of the face and the pixels in the masked region. As a result, we were able to generate images with higher quality than existing methods.

KW - COVID-19

KW - Face Completion

KW - Face Mask

KW - Generative Adversarial Networks

KW - Inpainting

UR - https://www.scopus.com/pages/publications/85175994469

UR - https://www.scopus.com/pages/publications/85175994469#tab=citedBy

U2 - 10.5220/0010827500003124

DO - 10.5220/0010827500003124

M3 - Conference article

AN - SCOPUS:85175994469

SN - 2184-5921

VL - 5

SP - 125

EP - 133

JO - Proceedings of the International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications

JF - Proceedings of the International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications

T2 - 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2022

Y2 - 6 February 2022 through 8 February 2022

ER -

GAN-based Face Mask Removal using Facial Landmarks and Pixel Errors in Masked Region

抄録

ASJC Scopus subject areas

文献へのアクセス

他のファイルとリンク

フィンガープリント

引用スタイル