Face recognition accuracy of forensic examiners, superrecognizers, and face recognition algorithms
Open accessAbstract
Significance: This study measures face identification accuracy for an international group of professional forensic facial examiners working under circumstances that apply in real world casework. Examiners and other human face “specialists,” including forensically trained facial reviewers and untrained superrecognizers, were more accurate than the control groups on a challenging test of face identification. Therefore, specialists are the best available human solution to the problem of face identification. We present data comparing state-of-the-art face recognition technology with the best human face identifiers. The best machine performed in the range of the best humans: professional facial examiners. However, optimal face identification was achieved only when humans and machines worked in collaboration.
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
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?
Face recognition algorithms often outperform forensic examiners at identifying whether pairs of face images show the same or different people.
“face recognition algorithms often outperform forensic examiners (who are significantly better than ordinary people) at identifying whether pairs of face images show the same or different people (13).”
From Deepfake detection by human crowds, machines, and machine-informed crowds (2022) · cites this paper as [13]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
DeepFace: Closing the Gap to Human-Level Performance in Face Verification
Taigman, Y., Yang, M., Ranzato, M., & Wolf, L. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2014.220 · Cited by 6,542 · see citing works
Parkhi, O. M., Vedaldi, A., & Zisserman, A. (2015) · Procedings of the British Machine Vision Conference 2015 · doi:10.5244/c.29.41 · Cited by 5,479 · see citing works
Kittler, J., Hatef, M., Duin, R. P. W., & Matas, J. (1998) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/34.667881 · Cited by 5,312 · see citing works
Strengthening Forensic Science in the United States: A Path Forward
Council, N. R. (2009) · CrimRxiv · doi:10.21428/cb6ab371.b2d683d2 · Cited by 1,461 · see citing works
Strengthening forensic science in the United States : a path forward
Policy, L. (2009) · National Academies Press eBooks · Cited by 961 · see citing works
Super-recognizers: People with extraordinary face recognition ability
Russell, R. P., Duchaine, B., & Nakayama, K. (2009) · Psychonomic Bulletin & Review · doi:10.3758/pbr.16.2.252 · Cited by 554 · see citing works
An All-In-One Convolutional Neural Network for Face Analysis
Ranjan, R., Sankaranarayanan, S., Castillo, C. D., & Chellappa, R. (2017) · International Conference on Automatic Face and Gesture Recognition · doi:10.1109/fg.2017.137 · Cited by 417 · see citing works
Self-citationFRVT 2006 and ICE 2006 Large-Scale Experimental Results
Phillips, P. J., Scruggs, W., O’Toole, A. J., Flynn, P. J., Bowyer, K. W., Schott, C. L., & Sharpe, M. J. (2009) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2009.59 · Cited by 406 · see citing works
Self-citationUnconstrained face verification using deep CNN features
Chen, J.-C., Patel, V. M., & Chellappa, R. (2016) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv.2016.7477557 · Cited by 271 · see citing works
Self-citationFace Recognition Algorithms Surpass Humans Matching Faces Over Changes in Illumination
O’Toole, A. J., Phillips, P. J., Jiang, F., et al. (2007) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2007.1107 · Cited by 215 · see citing works
Self-citationPerceptual expertise in forensic facial image comparison
White, D., Phillips, P. J., Hahn, C. A., Hill, M. Q., & O’Toole, A. J. (2015) · Proceedings of the Royal Society B Biological Sciences · doi:10.1098/rspb.2015.1292 · Cited by 191 · see citing works
Self-citationFace Recognition by Metropolitan Police Super-Recognisers
Robertson, D. J., Noyes, E., Dowsett, A. J., Jenkins, R., & Burton, A. M. (2016) · PLoS ONE · doi:10.1371/journal.pone.0150036 · Cited by 181 · see citing works
Self-citationComparison of human and computer performance across face recognition experiments
Phillips, P. J., & O’Toole, A. J. (2013) · Image and Vision Computing · doi:10.1016/j.imavis.2013.12.002 · Cited by 168 · see citing works
Self-citationAn introduction to the good, the bad, & the ugly face recognition challenge problem
Phillips, P. J., Beveridge, J. R., Draper, B. A., Givens, G. H., O’Toole, A. J., Bolme, D. S., Dunlop, J. P., Lui, Y. M., Sahibzada, H., & Weimer, S. (2011) · International Conference on Automatic Face and Gesture Recognition · doi:10.1109/fg.2011.5771424 · Cited by 135 · see citing works
Self-citationError Rates in Users of Automatic Face Recognition Software
White, D., Dunn, J. D., Schmid, A. C., & Kemp, R. I. (2015) · PLoS ONE · doi:10.1371/journal.pone.0139827 · Cited by 127 · see citing works
Self-citationInvestigating Predictors of Superior Face Recognition Ability in Police Super‐recognisers
Davis, J. P., Lander, K., Evans, R., & Jansari, A. S. (2016) · Applied Cognitive Psychology · doi:10.1002/acp.3260 · Cited by 111 · see citing works
Unfamiliar face matching: Pairs out‐perform individuals and provide a route to training
Dowsett, A. J., & Burton, A. M. (2014) · British Journal of Psychology · doi:10.1111/bjop.12103 · Cited by 108 · see citing works
Unaware Person Recognition From the Body When Face Identification Fails
Rice, A., Phillips, P. J., Natu, V. S., An, X., & O’Toole, A. J. (2013) · Psychological Science · doi:10.1177/0956797613492986 · Cited by 107 · see citing works
Self-citationFace processing: Systems, Disorders and Cultural Differences
Bindemann, M. (2017) · Kent Academic Repository (University of Kent) · Cited by 71 · see citing works
Comparing face recognition algorithms to humans on challenging tasks
O’Toole, A. J., An, X., Dunlop, J. P., Natu, V. S., & Phillips, P. J. (2012) · ACM Transactions on Applied Perception · doi:10.1145/2355598.2355599 · Cited by 70 · see citing works
Self-citationFusing Face-Verification Algorithms and Humans
O’Toole, A. J., Abdi, H., Jiang, F., et al. (2007) · IEEE Transactions on Systems Man and Cybernetics Part B (Cybernetics) · doi:10.1109/tsmcb.2007.907034 · Cited by 67 · see citing works
Self-citationCrowd Effects in Unfamiliar Face Matching
White, D., Burton, A. M., Kemp, R. I., & Jenkins, R. (2013) · Applied Cognitive Psychology · doi:10.1002/acp.2971 · Cited by 57 · see citing works
Self-citationThe Effect of Image Quality and Forensic Expertise in Facial Image Comparisons
Norell, K., Läthén, K. B., Bergström, P., et al. (2014) · Journal of Forensic Sciences · doi:10.1111/1556-4029.12660 · Cited by 45 · see citing works
Noyes, E., Phillips, P. J., & O’Toole, A. J. (2017) · Huddersfield Research Portal (University of Huddersfield) · Cited by 30 · see citing works
Self-citationWisdom of the social versus non‐social crowd in face identification
Jeckeln, G., Hahn, C. A., Noyes, E., Cavazos, J. G., & O’Toole, A. J. (2018) · British Journal of Psychology · doi:10.1111/bjop.12291 · Cited by 29 · see citing works
Self-citationA Cross Benchmark Assessment of a Deep Convolutional Neural Network for Face Recognition
Phillips, P. J. (2017) · International Conference on Automatic Face and Gesture Recognition · doi:10.1109/fg.2017.89 · Cited by 24 · see citing works
Self-citationHu, Y., Jackson, K., Yates, A. N., White, D., Phillips, P. J., & O’Toole, A. J. (2017) · Visual Cognition · doi:10.1080/13506285.2017.1297339 · Cited by 16 · see citing works
Self-citationTHE WINSTON CHURCHILL MEMORIAL TRUST OF AUSTRALIA
Fellowship, C., & Perso, T. (2002) · Cited by 14 · see citing works
Human Factors in Forensic Face Identification
White, D., Norell, K., Phillips, P. J., & O’Toole, A. J. (2017) · Advances in computer vision and pattern recognition · doi:10.1007/978-3-319-50673-9_9 · Cited by 13 · see citing works
Self-citation
Flagged first, then by citations
Research works printed but not linked: 1
- [17]Ranjan R Castillo CD Chellappa R 2017 L2-constrained softmax loss for discriminative face verification. arXiv:170309507
Cited by
Sorted by citations
Four principles of explainable artificial intelligence
Phillips, P. J., Hahn, C. A., Fontana, P. C., Yates, A. N., Greene, K., Broniatowski, D. A., & Przybocki, M. A. (2021) · doi:10.6028/nist.ir.8312 · Cited by 262 · see citing works
Self-citationDeepfake detection by human crowds, machines, and machine-informed crowds
Groh, M., Epstein, Z., Firestone, C., & Picard, R. (2022) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2110013119 · Cited by 248 · see citing works
Four Principles of Explainable Artificial Intelligence
Phillips, P. J., Hahn, C. A., Fontana, P. C., Broniatowski, D. A., & Przybocki, M. A. (2020) · doi:10.6028/nist.ir.8312-draft · Cited by 136 · see citing works
Self-citationDeepfakes and scientific knowledge dissemination
Doss, C., Mondschein, J., Shu, D., Wolfson, T., Kopecky, D., Fitton-Kane, V. A., Bush, L., & Tucker, C. (2023) · Scientific Reports · doi:10.1038/s41598-023-39944-3 · Cited by 90 · see citing works
McGrath, M. J., Duenser, A., Lacey, J., & Paris, C. (2025) · Computers in Human Behavior: Artificial Humans · doi:10.1016/j.chbah.2025.100200 · Cited by 51 · see citing works
Wu (2021) · doi:10.1523/eneuro.0267-20.2020 · Cited by 6 · see citing works
- Authors
- Phillips, P. J., Yates, A. N., Hu, Y., Hahn, C. A., Noyes, E., Jackson, K., Cavazos, J. G., Jeckeln, G., Ranjan, R., Sankaranarayanan, S., Chen, J.-C., Castillo, C. D., Chellappa, R., White, D., & O’Toole, A. J.
- Affiliations
- National Institute of Standards and Technology (US)
- The University of Texas at Dallas (US)
- University of Maryland, College Park (US)
- UNSW Sydney (AU)
- Corresponding author
- Phillips, P. Jonathon
- ORCID
- Venue
- Proceedings of the National Academy of Sciences
- Publisher
- National Academy of Sciences
- Published
- 2018-05-29
- Volume
- 115
- Issue
- 24
- Pages
- 6171-6176
- ISSN
- 0027-8424 (ISSN-L)
- 1091-6490
- PMC
- PMC6004481
- OpenAlex
- W2805754451
- Cited by
- 461
- References
- 29
- Citations by year
- 2026: 27 · 2025: 30 · 2024: 85 · 2023: 99 · 2022: 47 · 2021: 66 · 2020: 53 · 2019: 43 · 2018: 10 · 2017: 1
- FWCI
- 25.61
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- Proceedings of the National Academy of Sciences (publishedVersion, cc-by-nc-nd)
- Europe PMC (PubMed Central) (publishedVersion, cc-by-nc-nd)
- PubMed Central (publishedVersion, cc-by-nc-nd)
- PubMed (publishedVersion)
- Funders
- DOJ | National Institute of Justice (2015-IJ-CX-K014)
- ODNI | Intelligence Advanced Research Projects Activity (2014-14071600012)
- Topics
- Life Sciences › Neuroscience › Cognitive Neuroscience › Face Recognition and Perception
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Face recognition and analysis
- Physical Sciences › Computer Science › Signal Processing › Biometric Identification and Security
- Keywords
- face recognition, human-computer collaboration, deep convolutional neural networks, human performance variability
- MeSH terms
- Machine Learning, Algorithms, Face, Humans, Reproducibility of Results, Forensic Sciences, Biometric Identification
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
