The DOI matches its Crossref record.
Ongoing Face Recognition Vendor Test (FRVT) part 6B: face recognition accuracy with face masks using post-COVID-19 algorithms
Abstract
This is the second of a series of reports on the performance of face recognition algorithms on faces occluded by protective face masks commonly worn to reduce inhalation and exhalation of viruses. This is a continuous study is being run under the Ongoing Face Recognition Vendor Test (FRVT) executed by the National Institute of Standards and Technology (NIST). In our first report, we tested "pre-pandemic" algorithms that were already submitted to FRVT 1:1 prior to mid-March 2020. This report augments its predecessor with results for more recent algorithms provided to NIST after the COVID-19 pandemic was declared. While we do not have information on whether or not a particular algorithm was designed with face coverings in mind, the results show evidence that a number of developers have adapted their algorithms to support face recognition on subjects potentially wearing face masks. The algorithms tested were one- to-one algorithms submitted to the FRVT 1:1 Verification track. Future editions of this document will also report accuracy of one-to- many algorithms.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Automated facial identification systems are increasing in performance.
“such as the increasing performance of automated systems [31,75,76]”
From Forensic Facial Comparison: Current Status, Limitations, and Future Directions (2021) · cites this paper as [75]
The deleterious effect in performance seemed to vary based on the color and shape of the masks as well.
“The deleterious effect in performance seemed to vary based on the color and shape of the masks as well [76].”
From Forensic Facial Comparison: Current Status, Limitations, and Future Directions (2021) · cites this paper as [76]
Automated facial identification systems are increasing in performance.
“such as the increasing performance of automated systems [31,75,76]”
From Forensic Facial Comparison: Current Status, Limitations, and Future Directions (2021) · cites this paper as [76]
Face masks have been shown to reduce automated facial recognition performance by 5 to 50% depending on the specific algorithm and extent of the face covered.
“Face masks, which can vary in shape, size, and the resulting proportion of the face covered, have been shown to reduce automated facial recognition performance by 5 to 50% depending on the specific algorithm and extent of the face covered [76].”
From Forensic Facial Comparison: Current Status, Limitations, and Future Directions (2021) · cites this paper as [76]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
No references could be resolved yet.
Research works printed but not linked: 3
- Patrick Grother, George Quinn, and Mei Ngan. Face in video evaluation (five) face recognition of non- cooperative subjects. Interagency Report 8173, National Institute of Standards and Technology, March 2017. https://doi.org/10.6028/NIST.IR.8173. · DOI
- Davis E. King. Dlib-ml: A machine learning toolkit. Journal of Machine Learning Research, 10:1755-1758, 2009. http: //dlib.net.
- Mei Ngan, Patrick Grother, and Kayee Hanaoka. NISTIR 8311 -Ongoing FRVT Part 6A: Face recognition accuracy with face masks using pre-COVID-19 algorithms. Interagency Report 8311, National Institute of Standards and Technology, July 2020. https://doi.org/10.6028/NIST.IR.8311. · DOI
Cited by
Sorted by citations
Towards AI Accountability Infrastructure: Gaps and Opportunities in AI Audit Tooling
Ojewale, V., Steed, R., Vecchione, B., Birhane, A., & Raji, I. D. (2025) · Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3706598.3713301 · Cited by 81
Forensic Facial Comparison: Current Status, Limitations, and Future Directions
Bacci, N., Davimes, J. G., Steyn, M., & Briers, N. (2021) · Biology · doi:10.3390/biology10121269 · Cited by 26
Metadata
- Authors
- Ngan, M. L., Grother, P. J., & Hanaoka, K.
- Published
- 2020-11-29
- Cited by
- 62
- References
- 0
- Funders
- National Institute of Standards and Technology
- Information Technology Laboratory
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
