MaskedFace-Net
收藏资源简介:
MaskedFace-Net是一个包含137,016张图像的大型数据集,由诺曼底大学等机构合作创建。该数据集分为正确佩戴口罩(CMFD)和错误佩戴口罩(IMFD)两大类,旨在通过深度学习模型检测和分析人脸是否正确佩戴口罩。数据集的创建过程涉及使用FFHQ数据集作为基础,通过特定的mask-to-face变形模型生成图像。MaskedFace-Net的应用领域包括机场门户和人群监控,旨在解决疫情期间口罩佩戴的合规性问题。
MaskedFace-Net is a large-scale dataset consisting of 137,016 images, co-created by institutions such as the University of Normandy and other collaborating organizations. The dataset is categorized into two major classes: Correctly Masked Face Dataset (CMFD) and Incorrectly Masked Face Dataset (IMFD), with its core objective being to detect and analyze whether faces wear masks correctly using deep learning models. The construction of MaskedFace-Net takes the FFHQ dataset as the foundation, generating images via a specialized mask-to-face deformation model. Application scenarios of MaskedFace-Net include airport access control and crowd monitoring, aiming to resolve mask-wearing compliance issues during the COVID-19 pandemic.




