Facial recognition software can accurately identify people who are wearing face masks.

Is it true that facial recognition software can accurately identify people who are wearing face masks?

Partly. A NIST preliminary study found pre-pandemic commercial algorithms had 5–50% error rates matching masked faces, while several 2021–2025 papers describe masked-face recognition algorithms that improve accuracy on masked datasets. The evidence covers only partial, often preliminary or dataset-specific results on both sides.

Verdict: CONTESTED PARTIAL · 3 sources for, 2 against · Last checked 3 Oct 2026

The private library was unavailable for some searches in this check. Missing library results are not evidence against the claim.

Online Attendance System Using CCTV. (2023). doi.org. https://doi.org/10.26562/ijiris.2023.v0902.07

Ventura, M., Palmisano, A., Innamorato, F., Tedesco, G., Manippa, V., Caffò, A. O., & Rivolta, D. (2022). Face memory and facial expression recognition are both affected by wearing disposable surgical face masks. doi.org. https://doi.org/10.1007/s10339-022-01112-2

MFCosface: A Masked-Face Recognition Algorithm Based on Large Margin Cosine Loss. (n.d.). cir.nii.ac.jp. https://doi.org/10.3390/app11167310

JA, A., KS, P., R, A., SK, S., S, V., & K, A. (2025). A generative adversarial network-based accurate masked face recognition model using dual scale adaptive efficient attention network.. doi.org. https://doi.org/10.1038/s41598-025-02144-2

NIST Launches Studies into Masks’ Effect on Face Recognition Software | NISTLock. (n.d.). nist.gov. https://www.nist.gov/news-events/news/2020/07/nist-launches-studies-masks-effect-face-recognition-software