Publication Details

Facial Image Reconstruction and its Influence to Face Recognition

PLEŠKO Filip, GOLDMANN Tomáš and MALINKA Kamil. Facial Image Reconstruction and its Influence to Face Recognition. In: 2023 International Conference of the Biometrics Special Interest Group (BIOSIG). Darmstadt: Society for Informatics, 2023, pp. 1-4. ISBN 979-8-3503-3655-9. Available from: https://ieeexplore.ieee.org/document/10346000
Czech title
Rekonstrukce snímku obličeje a její vliv na rozpoznávání
Type
conference paper
Language
english
Authors
Pleško Filip, Ing. (FIT BUT)
Goldmann Tomáš, Ing. (DITS FIT BUT)
Malinka Kamil, Mgr., Ph.D. (DITS FIT BUT)
URL
Keywords

face recognition, neural network, face reconstruction, generativní adversariální síť, SFace, ArcFace, QMagFace

Abstract

This paper focuses on reconstructing damaged facial images using GAN neural networks. In addition, the effect of generating the missing part of the face on face recognition is investigated. The main objective of this work is to observe whether it is possible to increase the accuracy of face recognition by generating missing parts while maintaining a low false accept rate (FAR). A new model for generating the missing parts of a face has been proposed. For face-based recognition, state-of-the-art solutions from the DeepFace library and the QMagFace solution have been used.

Published
2023
Pages
1-4
Proceedings
2023 International Conference of the Biometrics Special Interest Group (BIOSIG)
Conference
22nd International Conference of the Biometrics Special Interest Group, Darmstadt, DE
ISBN
979-8-3503-3655-9
Publisher
Society for Informatics
Place
Darmstadt, DE
DOI
BibTeX
@INPROCEEDINGS{FITPUB13034,
   author = "Filip Ple\v{s}ko and Tom\'{a}\v{s} Goldmann and Kamil Malinka",
   title = "Facial Image Reconstruction and its Influence to Face Recognition",
   pages = "1--4",
   booktitle = "2023 International Conference of the Biometrics Special Interest Group (BIOSIG)",
   year = 2023,
   location = "Darmstadt, DE",
   publisher = "Society for Informatics",
   ISBN = "979-8-3503-3655-9",
   doi = "10.1109/BIOSIG58226.2023",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/13034"
}
Back to top