MLFW: A Database for Face Recognition on Masked Faces

TL;DR

This paper introduces MLFW, a new masked face database created to evaluate face recognition systems' performance degradation caused by masks, highlighting the need for more robust models during pandemics.

Contribution

The authors develop a method to generate masked faces from unmasked images and construct the MLFW database, enabling assessment of face recognition under masked conditions.

Findings

Recognition accuracy drops 5%-16% on MLFW compared to unmasked images.

Generated masks have high visual consistency with original faces.

Diverse mask styles are included to simulate real-world scenarios.

Abstract

As more and more people begin to wear masks due to current COVID-19 pandemic, existing face recognition systems may encounter severe performance degradation when recognizing masked faces. To figure out the impact of masks on face recognition model, we build a simple but effective tool to generate masked faces from unmasked faces automatically, and construct a new database called Masked LFW (MLFW) based on Cross-Age LFW (CALFW) database. The mask on the masked face generated by our method has good visual consistency with the original face. Moreover, we collect various mask templates, covering most of the common styles appeared in the daily life, to achieve diverse generation effects. Considering realistic scenarios, we design three kinds of combinations of face pairs. The recognition accuracy of SOTA models declines 5%-16% on MLFW database compared with the accuracy on the original images. MLFW database can be viewed and downloaded at \url{http://whdeng.cn/mlfw}.