Masked Face Detection Dataset (MFDD), Real-world Masked Face Recognition Dataset (RMFRD), Simulated Masked Face Recognition Dataset (SMFRD)

本研究提出了三个类型的蒙面人脸数据集，包括Masked Face Detection Dataset (MFDD), Real-world Masked Face Recognition Dataset (RMFRD) 和 Simulated Masked Face Recognition Dataset (SMFRD)。其中，RMFRD是目前全球最大的真实蒙面人脸数据集。这些数据集主要用于蒙面人脸的检测和识别，数据来源包括互联网爬取和现有研究样本，通过人工标注和半自动工具处理，确保数据质量。数据集的应用领域主要集中在社区出入管理、车站安检等需要非接触式身份验证的场景，旨在解决因佩戴口罩导致传统人脸识别技术失效的问题。

This study proposes three types of masked face datasets, namely Masked Face Detection Dataset (MFDD), Real-world Masked Face Recognition Dataset (RMFRD), and Simulated Masked Face Recognition Dataset (SMFRD). Among these, RMFRD is currently the world's largest real-world masked face dataset. These datasets are primarily intended for masked face detection and recognition. Their data sources include web-crawled data and existing research samples, which are processed via manual annotation and semi-automatic tools to ensure data quality. The application scenarios of these datasets mainly focus on non-contact identity verification scenarios such as community access control and station security inspection, aiming to address the failure of traditional face recognition technologies caused by mask wearing.