Thesis Details
Research in the field of biometric detection and recognition of individuals using facial image data
Biometric recognition has long since become a common concern in various fields of study, including forensics, anthropometry, biometrics, and computer science. This thesis focuses on the development of an approach to create datasets for the evaluation of face recognition algorithms, with an emphasis on the preservation of facial features. Such datasets open up new possibilities for the evaluation of face recognition algorithms, which were previously hindered by the limited sample size of the datasets usually used. Through extensive research in the field of face recognition algorithms and modern neural network techniques, algorithms for face detection and recognition on embedded devices have been developed. These algorithms are based on the EfficientNet feature extractor.
face, face detection, face recognition, neural network, Neural Processing Unit, generovaný dataset, EfficientNet, RetinaFace, ArcFace, MagFace
@phdthesis{FITPT988,
author = "Tom\'{a}\v{s} Goldmann",
type = "Ph.D. thesis",
title = "Research in the field of biometric detection and recognition of individuals using facial image data",
school = "Brno University of Technology, Faculty of Information Technology",
year = 2024,
location = "Brno, CZ",
language = "english",
url = "https://www.fit.vut.cz/study/phd-thesis/988/"
}