3-D Face Reconstruction Method Using Deep Learning Based Simulated Annealing

Fei Fei Chen; Bing Guan; Sumi Kim; Jaeho Choi

doi:10.1007/978-3-031-39777-6_26

3-D Face Reconstruction Method Using Deep Learning Based Simulated Annealing

Fei Fei Chen
, Bing Guan
, Sumi Kim
, Jaeho Choi^*

^*Corresponding author for this work

Division of Electronic Engineering

JBNU
Huizhou University
Seoyoung University

Research output: Contribution to conference › Conference paper › peer-review

Abstract

Face reconstruction becomes more accurate and popular in the age of artificial intelligence and deep learning. Particularly, 3-D face reconstruction can be one of the key technologies in meta-world and virtual reality applications. In recent years, studies on realistic 3-D face reconstruction draw much attention among the researchers. For realizing accurate shape and facial textures, the neural network models for deep learning are under investigation and several schemes are present-ed. In this paper, a deep learning based simulated annealing algorithm is proposed for 3-D face reconstruction. Face labeling, feature extraction, and 3-D reconstruction are three major elements investigated in this study. A set of computer simulation is performed by using the CelebFaces Attributes and Labeled Faces in the Wild data sets. The system performance is evaluated in terms of reconstruction accuracy and the results show us that the proposed method can be a successful alternative for providing accurate and robust 3-D face reconstruction.

Original language	English
Title of host publication	Intelligent and Fuzzy Systems - Intelligence and Sustainable Future Proceedings of the INFUS 2023 Conference
Editors	Cengiz Kahraman, Irem Ucal Sari, Basar Oztaysi, Sezi Cevik Onar, Selcuk Cebi, A. Çağrı Tolga
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	215-221
Number of pages	7
ISBN (Print)	9783031397769
DOIs	https://doi.org/10.1007/978-3-031-39777-6_26
State	Published - 2023
Event	Intelligent and Fuzzy Systems - Intelligence and Sustainable Future Proceedings of the INFUS 2023 Conference - Istanbul, Turkey Duration: 2023.08.22 → 2023.08.24

Publication series

Name	Lecture Notes in Networks and Systems
Volume	759 LNNS
ISSN (Print)	2367-3370
ISSN (Electronic)	2367-3389

Conference

Conference	Intelligent and Fuzzy Systems - Intelligence and Sustainable Future Proceedings of the INFUS 2023 Conference
Country/Territory	Turkey
City	Istanbul
Period	23.08.22 → 23.08.24

Keywords

3-D face reconstruction
Deep learning
Face labeling
Face textures
Feature extraction
Simulated annealing

Access to Document

10.1007/978-3-031-39777-6_26

Cite this

Chen, F. F., Guan, B., Kim, S., & Choi, J. (2023). 3-D Face Reconstruction Method Using Deep Learning Based Simulated Annealing. In C. Kahraman, I. U. Sari, B. Oztaysi, S. Cevik Onar, S. Cebi, & A. Ç. Tolga (Eds.), Intelligent and Fuzzy Systems - Intelligence and Sustainable Future Proceedings of the INFUS 2023 Conference (pp. 215-221). (Lecture Notes in Networks and Systems; Vol. 759 LNNS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-39777-6_26

Chen, Fei Fei ; Guan, Bing ; Kim, Sumi et al. / 3-D Face Reconstruction Method Using Deep Learning Based Simulated Annealing. Intelligent and Fuzzy Systems - Intelligence and Sustainable Future Proceedings of the INFUS 2023 Conference. editor / Cengiz Kahraman ; Irem Ucal Sari ; Basar Oztaysi ; Sezi Cevik Onar ; Selcuk Cebi ; A. Çağrı Tolga. Springer Science and Business Media Deutschland GmbH, 2023. pp. 215-221 (Lecture Notes in Networks and Systems).

@inproceedings{6899ef6d615e43a9ae226385e1a6fb4c,

title = "3-D Face Reconstruction Method Using Deep Learning Based Simulated Annealing",

abstract = "Face reconstruction becomes more accurate and popular in the age of artificial intelligence and deep learning. Particularly, 3-D face reconstruction can be one of the key technologies in meta-world and virtual reality applications. In recent years, studies on realistic 3-D face reconstruction draw much attention among the researchers. For realizing accurate shape and facial textures, the neural network models for deep learning are under investigation and several schemes are present-ed. In this paper, a deep learning based simulated annealing algorithm is proposed for 3-D face reconstruction. Face labeling, feature extraction, and 3-D reconstruction are three major elements investigated in this study. A set of computer simulation is performed by using the CelebFaces Attributes and Labeled Faces in the Wild data sets. The system performance is evaluated in terms of reconstruction accuracy and the results show us that the proposed method can be a successful alternative for providing accurate and robust 3-D face reconstruction.",

keywords = "3-D face reconstruction, Deep learning, Face labeling, Face textures, Feature extraction, Simulated annealing",

author = "Chen, \{Fei Fei\} and Bing Guan and Sumi Kim and Jaeho Choi",

note = "Publisher Copyright: {\textcopyright} 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.; Intelligent and Fuzzy Systems - Intelligence and Sustainable Future Proceedings of the INFUS 2023 Conference ; Conference date: 22-08-2023 Through 24-08-2023",

year = "2023",

doi = "10.1007/978-3-031-39777-6\_26",

language = "English",

isbn = "9783031397769",

series = "Lecture Notes in Networks and Systems",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "215--221",

editor = "Cengiz Kahraman and Sari, \{Irem Ucal\} and Basar Oztaysi and \{Cevik Onar\}, Sezi and Selcuk Cebi and Tolga, \{A. {\c C}ağrı\}",

booktitle = "Intelligent and Fuzzy Systems - Intelligence and Sustainable Future Proceedings of the INFUS 2023 Conference",

}

Chen, FF, Guan, B, Kim, S & Choi, J 2023, 3-D Face Reconstruction Method Using Deep Learning Based Simulated Annealing. in C Kahraman, IU Sari, B Oztaysi, S Cevik Onar, S Cebi & AÇ Tolga (eds), Intelligent and Fuzzy Systems - Intelligence and Sustainable Future Proceedings of the INFUS 2023 Conference. Lecture Notes in Networks and Systems, vol. 759 LNNS, Springer Science and Business Media Deutschland GmbH, pp. 215-221, Intelligent and Fuzzy Systems - Intelligence and Sustainable Future Proceedings of the INFUS 2023 Conference, Istanbul, Turkey, 23.08.22. https://doi.org/10.1007/978-3-031-39777-6_26

3-D Face Reconstruction Method Using Deep Learning Based Simulated Annealing. / Chen, Fei Fei; Guan, Bing; Kim, Sumi et al.
Intelligent and Fuzzy Systems - Intelligence and Sustainable Future Proceedings of the INFUS 2023 Conference. ed. / Cengiz Kahraman; Irem Ucal Sari; Basar Oztaysi; Sezi Cevik Onar; Selcuk Cebi; A. Çağrı Tolga. Springer Science and Business Media Deutschland GmbH, 2023. p. 215-221 (Lecture Notes in Networks and Systems; Vol. 759 LNNS).

Research output: Contribution to conference › Conference paper › peer-review

TY - GEN

T1 - 3-D Face Reconstruction Method Using Deep Learning Based Simulated Annealing

AU - Chen, Fei Fei

AU - Guan, Bing

AU - Kim, Sumi

AU - Choi, Jaeho

PY - 2023

Y1 - 2023

N2 - Face reconstruction becomes more accurate and popular in the age of artificial intelligence and deep learning. Particularly, 3-D face reconstruction can be one of the key technologies in meta-world and virtual reality applications. In recent years, studies on realistic 3-D face reconstruction draw much attention among the researchers. For realizing accurate shape and facial textures, the neural network models for deep learning are under investigation and several schemes are present-ed. In this paper, a deep learning based simulated annealing algorithm is proposed for 3-D face reconstruction. Face labeling, feature extraction, and 3-D reconstruction are three major elements investigated in this study. A set of computer simulation is performed by using the CelebFaces Attributes and Labeled Faces in the Wild data sets. The system performance is evaluated in terms of reconstruction accuracy and the results show us that the proposed method can be a successful alternative for providing accurate and robust 3-D face reconstruction.

AB - Face reconstruction becomes more accurate and popular in the age of artificial intelligence and deep learning. Particularly, 3-D face reconstruction can be one of the key technologies in meta-world and virtual reality applications. In recent years, studies on realistic 3-D face reconstruction draw much attention among the researchers. For realizing accurate shape and facial textures, the neural network models for deep learning are under investigation and several schemes are present-ed. In this paper, a deep learning based simulated annealing algorithm is proposed for 3-D face reconstruction. Face labeling, feature extraction, and 3-D reconstruction are three major elements investigated in this study. A set of computer simulation is performed by using the CelebFaces Attributes and Labeled Faces in the Wild data sets. The system performance is evaluated in terms of reconstruction accuracy and the results show us that the proposed method can be a successful alternative for providing accurate and robust 3-D face reconstruction.

KW - 3-D face reconstruction

KW - Deep learning

KW - Face labeling

KW - Face textures

KW - Feature extraction

KW - Simulated annealing

UR - https://www.scopus.com/pages/publications/85172734209

U2 - 10.1007/978-3-031-39777-6_26

DO - 10.1007/978-3-031-39777-6_26

M3 - Conference paper

AN - SCOPUS:85172734209

SN - 9783031397769

T3 - Lecture Notes in Networks and Systems

SP - 215

EP - 221

BT - Intelligent and Fuzzy Systems - Intelligence and Sustainable Future Proceedings of the INFUS 2023 Conference

A2 - Kahraman, Cengiz

A2 - Sari, Irem Ucal

A2 - Oztaysi, Basar

A2 - Cevik Onar, Sezi

A2 - Cebi, Selcuk

A2 - Tolga, A. Çağrı

PB - Springer Science and Business Media Deutschland GmbH

T2 - Intelligent and Fuzzy Systems - Intelligence and Sustainable Future Proceedings of the INFUS 2023 Conference

Y2 - 22 August 2023 through 24 August 2023

ER -

Chen FF, Guan B, Kim S, Choi J. 3-D Face Reconstruction Method Using Deep Learning Based Simulated Annealing. In Kahraman C, Sari IU, Oztaysi B, Cevik Onar S, Cebi S, Tolga AÇ, editors, Intelligent and Fuzzy Systems - Intelligence and Sustainable Future Proceedings of the INFUS 2023 Conference. Springer Science and Business Media Deutschland GmbH. 2023. p. 215-221. (Lecture Notes in Networks and Systems). doi: 10.1007/978-3-031-39777-6_26

3-D Face Reconstruction Method Using Deep Learning Based Simulated Annealing

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this