FaceNet: A unified embedding for face recognition and clustering
Open accessThe DOI matches its Crossref record; 23 of the 23 references it cites were found in a registry or an index.
Abstract
Despite significant recent advances in the field of face recognition [10, 14, 15, 17], implementing face verification and recognition efficiently at scale presents serious challenges to current approaches. In this paper we present a system, called FaceNet, that directly learns a mapping from face images to a compact Euclidean space where distances directly correspond to a measure offace similarity. Once this space has been produced, tasks such as face recognition, verification and clustering can be easily implemented using standard techniques with FaceNet embeddings asfeature vectors. Our method uses a deep convolutional network trained to directly optimize the embedding itself, rather than an intermediate bottleneck layer as in previous deep learning approaches. To train, we use triplets of roughly aligned matching / non-matching face patches generated using a novel online triplet mining method. The benefit of our approach is much greater representational efficiency: we achieve state-of-the-artface recognition performance using only 128-bytes perface. On the widely used Labeled Faces in the Wild (LFW) dataset, our system achieves a new record accuracy of 99.63%. On YouTube Faces DB it achieves 95.12%. Our system cuts the error rate in comparison to the best published result [15] by 30% on both datasets.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
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?
The cited source says what the claim attributes to it.
In the citing paper
“Current face recognition systems [8,24,43], already reach very high levels of accuracy in public non-forensic benchmarks, and it is expected that in the coming years they will keep improving.”
From Calibration of score based likelihood ratio estimation in automated forensic facial image comparison (2022) · cites this paper as [43]
The check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“An increasing number of UDA techniques have been applied in HAR; however, most of them work between domains that share the same feature space; for example, transferring the learn models between users [44] or between sensor wearing positions [32].”
From ContrasGAN: Unsupervised domain adaptation in Human Activity Recognition via adversarial and contrastive learning (2021) · cites this paper as [32]
Experimental — accuracy not yet measuredIn the citing paper
“To address this issue, a representation learning method using triplet loss is proposed to directly learn discriminative and compact face embedding [99].”
From Empowering things with intelligence: a survey of the progress, challenges, and opportunities in artificial intelligence of things (2021) · cites this paper as [99]
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
Going deeper with convolutions
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298594 · Cited by 48,562 · cited 8×
“We would like to thank Johannes Steffens for his discussions and great insights on face recognition and Christian Szegedy for providing new network architectures like [16] and discussing network design choices.”
MethodResultInfluentialLearning representations by back-propagating errors
Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986) · Nature · doi:10.1038/323533a0 · Cited by 31,892 · cited 2×
“In all our experiments we train the CNN using Stochastic Gradient Descent (SGD) with standard backprop [8, 11] and AdaGrad [5].”
MethodVisualizing and understanding convolutional networks
Zeiler, M. D., & Fergus, R. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10590-1_53 · Cited by 15,839 · cited 5×
“This table show the structure of our Zeiler&Fergus [22] based model with 1×1 convolutions inspired by [9].”
BackgroundMethodInfluentialBackpropagation Applied to Handwritten Zip Code Recognition
LeCun, Y., Boser, B., Denker, J. S., Henderson, D., Howard, R. E., Hubbard, W., & Jackel, L. D. (1989) · Neural Computation · doi:10.1162/neco.1989.1.4.541 · Cited by 11,891 · cited 2×
“In all our experiments we train the CNN using Stochastic Gradient Descent (SGD) with standard backprop [8, 11] and AdaGrad [5].”
Method(2010) · Cited by 8,261
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 · cited 9×
“[17], for example, performs a complex 3D alignment.”
BackgroundMethodInfluentialBengio, Y., Louradour, J., Collobert, R., & Weston, J. (2009) · Proceedings of the 26th Annual International Conference on Machine Learning · doi:10.1145/1553374.1553380 · Cited by 5,366 · cited 1×
“Choosing which triplets to use turns out to be very important for achieving good performance and, inspired by curriculum learning [1], we present a novel online negative exemplar mining strategy which ensures consistently increasing difficulty of triplets as the network trains.”
BackgroundLarge Scale Distributed Deep Networks
Dean, J. B., Corrado, G. S., Monga, R., Chen, K., Devin, M., Mao, M., Ranzato, M., Senior, A., Tucker, P. A., Yang, K., et al. (2012) · neural information processing systems · Cited by 4,489
Distance Metric Learning for Large Margin Nearest Neighbor Classification
Weinberger, K. Q., & Saul, L. K. (2009) · Journal of Machine Learning Research · doi:10.5555/1577069.1577078 · Cited by 4,044 · cited 4×
“D embedding using a tripletbased loss function based on LMNN [19].”
MethodInfluentialDeep Learning Face Representation by Joint Identification-Verification
Sun, Y., Wang, X., & Tang, X. (2014) · arXiv · doi:10.48550/arxiv.1406.4773 · Cited by 1,906 · cited 4×
“the one using pairs of positives and negatives, as used in [14] Eq.”
BackgroundMethodInfluentialGoing Deeper with Convolutions
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2014) · arXiv · doi:10.48550/arxiv.1409.4842 · Cited by 1,461 · cited 8×
“We would like to thank Johannes Steffens for his discussions and great insights on face recognition and Christian Szegedy for providing new network architectures like [16] and discussing network design choices.”
MethodResultInfluentialThe CMU Pose, Illumination, and Expression (PIE) database
Sim, T., Baker, S., & Bsat, M. (2003) · Proceedings of Fifth IEEE International Conference on Automatic Face Gesture Recognition · doi:10.1109/afgr.2002.1004130 · Cited by 1,419 · cited 1×
“Shown are image pairs from PIE [13] that previously were considered to be very difficult for face verification systems.”
BackgroundFace recognition in unconstrained videos with matched background similarity
Wolf, L., Hassner, T., & Maoz, I. (2011) · CVPR 2011 · doi:10.1109/cvpr.2011.5995566 · Cited by 1,382 · cited 1×
“Youtube Faces DB [21] is a new dataset that has gained popularity in the face recognition community [17, 15].”
BackgroundLearning Fine-Grained Image Similarity with Deep Ranking
Wang, J., Song, Y., Leung, T., Rosenberg, C., Wang, J., Philbin, J., Chen, B., & Wu, Y. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2014.180 · Cited by 1,269 · cited 1×
“[18] for ranking images by semantic and visual similarity.”
BackgroundSelf-citationDeeply learned face representations are sparse, selective, and robust
Sun, Y., Wang, X., & Tang, X. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298907 · Cited by 939 · cited 10×
“Previous face recognition approaches based on deep networks use a classification layer [15, 17] trained over a set of known face identities and then take an intermediate bottleneck layer as a representation used to generalize recognition beyond the set of identities used in training.”
BackgroundMethodResultInfluentialGoodfellow, I. J., Warde-Farley, D., Mirza, M., Courville, A., & Bengio, Y. (2013) · arXiv · doi:10.48550/arxiv.1302.4389 · Cited by 935 · cited 1×
“The kernel is specified as rows× cols, stride and the maxout [6] pooling size as p = 2.”
MethodLearning a Distance Metric from Relative Comparisons
Schultz, M., & Joachims, T. (2003) · Cited by 609
Visualizing and Understanding Convolutional Networks
Zeiler, M. D., & Fergus, R. (2013) · arXiv · doi:10.48550/arxiv.1311.2901 · Cited by 522 · cited 5×
“This table show the structure of our Zeiler&Fergus [22] based model with 1×1 convolutions inspired by [9].”
BackgroundMethodInfluentialBayesian Face Revisited: A Joint Formulation
Chen, D., Cao, X., Wang, L., Wen, F., & Sun, J. (2012) · Lecture Notes in Computer Science · doi:10.1007/978-3-642-33712-3_41 · Cited by 427 · cited 1×
“Both PCA and a Joint Bayesian model [2] that effectively correspond to a linear transform in the embedding space are employed.”
MethodThe general inefficiency of batch training for gradient descent learning
Wilson, D. R., & Martinez, T. R. (2003) · Neural Networks · doi:10.1016/s0893-6080(03)00138-2 · Cited by 418 · cited 2×
“On the one hand we would like to use small mini-batches as these tend to improve convergence during Stochastic Gradient Descent (SGD) [20].”
MethodJoint Cascade Face Detection and Alignment
Chen, D., Ren, S., Wei, Y., Cao, X., & Sun, J. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10599-4_8 · Cited by 405 · cited 1×
“A proprietary face detector (similar to Picasa [3]) is run on the provided LFW thumbnails.”
MethodSurpassing Human-Level Face Verification Performance on LFW with GaussianFace
Lu, C., & Tang, X. (2015) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v29i1.9797 · Cited by 303 · cited 1×
“Despite significant recent advances in the field of face recognition [10, 14, 15, 17], implementing face verification and recognition efficiently at scale presents serious challenges to current approaches.”
BackgroundRecover Canonical-View Faces in the Wild with Deep Neural NetworksPreprint
Zhu, Z., Luo, P., Wang, X., & Tang, X. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1404.3543 · Cited by 129 · cited 2×
“The works of [15, 17, 23] all employ a complex system of multiple stages, that combines the output of a deep convolutional network with PCA for dimensionality reduction and an SVM for classification.”
Method
Research works printed but not linked: 2
- huang, 2007, Labeled faces in the wild: A database for studying face recognition in unconstrained environments, Technical Report, 7
- lin, 2013, Network in network, CaRR abs/1312 4400
Cited by
Rethinking the Inception Architecture for Computer Vision
Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., & Wojna, Z. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.308 · Cited by 32,138
MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
Howard, A. G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., & Adam, H. (2017) · arXiv · doi:10.48550/arxiv.1704.04861 · Cited by 19,015
“The FaceNet model is a state of the art face recognition model [25].”
MethodSelf-citationA survey on Image Data Augmentation for Deep Learning
Shorten, C., & Khoshgoftaar, T. M. (2019) · Journal of Big Data · doi:10.1186/s40537-019-0197-0 · Cited by 13,203
Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks
Reimers, N., & Gurevych, I. (2019) · Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) · doi:10.18653/v1/d19-1410 · Cited by 12,778
“In order to fine-tune BERT / RoBERTa, we create siamese and triplet networks (Schroff et al., 2015) to update the weights such that the produced sentence embeddings are semantically meaningful and can be compared with cosine-similarity.”
MethodLearning Transferable Architectures for Scalable Image Recognition
Zoph, B., Vasudevan, V., Shlens, J., & Le, Q. V. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00907 · Cited by 7,831
Metadata
- Authors
- Schroff, F., Kalenichenko, D., & Philbin, J.
- Affiliations
- Google (United States) (US)
- Venue
- 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2015-06
- Pages
- 815-823
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W2096733369
- Semantic Scholar
- 5aa26299435bdf7db874ef1640a6c3b5a4a2c394
- Cited by
- 15,267
- References
- 22
- Citations by year
- 2026: 846 · 2025: 1556 · 2024: 1590 · 2023: 1823 · 2022: 1872 · 2021: 2113 · 2020: 1902 · 2019: 1574 · 2018: 1062 · 2017: 593 · 2016: 281 · 2015: 51
- FWCI
- 271.65
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings (publishedVersion)
- arXiv (Cornell University) (publishedVersion)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Face recognition and analysis
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Face and Expression Recognition
- Life Sciences › Neuroscience › Cognitive Neuroscience › Face Recognition and Perception
- Keywords
- FaceNet, face recognition, face verification, face clustering, embedding space, triplet loss, deep convolutional neural networks, face similarity, Labeled Faces in the Wild
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.