期刊
IEEE ACCESS
卷 9, 期 -, 页码 80511-80523出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/ACCESS.2021.3085423
关键词
Face recognition; Generative adversarial networks; Deep learning; Generators; Databases; Data models; Training; Generative adversarial networks; StyleGAN2; thermal face recognition; deep learning
资金
- Fondo Nacional de Desarrollo Cientifico y Tecnologico (FONDECYT) [1191188, 1181943]
- Pontificia Universidad Catolica de Valparaiso (PUCV) DI-Consolidado [039.381/2021]
This article utilizes StyleGAN2 and generative adversarial networks (GANs) to create high-quality synthetic thermal images and build thermal face recognition models using deep learning. By training with thermal databases and pretrained deep learning models, the synthetic thermal database achieved 99.98% accuracy in classifying thermal face images.
This article proposes the use of generative adversarial networks (GANs) via StyleGAN2 to create high-quality synthetic thermal images and obtain training data to build thermal face recognition models using deep learning. We employed different variants of StyleGAN2, incorporating the new improved version of StyleGAN that uses adaptive discriminator augmentation (ADA). In addition, three different thermal databases from the literature were employed to train a thermal face detector based on YOLOv3 and to train StyleGAN2 and its variants, evaluating different metrics. The synthetic thermal database was built using GANSpace to manipulate the intermediate latent space w of StyleGAN2 and obtain images with different characteristics, such as eyeglasses, rotation, beards, etc. We carried out the training of 6 pretrained deep learning models for face recognition to validate the use of our synthetic thermal database, obtaining 99.98% accuracy for classifying synthetic thermal face images.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据