--- title: "FaceNet: A Unified Embedding for Face — Schroff 2015" description: "Most deep face recognition systems train a network to classify known identities." authors: ["Florian Schroff", "Dmitry Kalenichenko", "James Philbin"] year: 2015 venue: "arXiv:1503.03832" source: "https://arxiv.org/abs/1503.03832" doi: "10.1109/cvpr.2015.7298682" arxiv: "1503.03832" licence: "arXiv — non-exclusive licence to distribute" licence_url: "http://arxiv.org/licenses/nonexclusive-distrib/1.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/en/papers/schroff-2015-facenet-unified-embedding-face-recognition/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/schroff-2015-facenet-unified-embedding-face-recognition/) [home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2015](https://md.research.elamur.ai/free/en/papers/2015/) / FaceNet: A Unified Embedding for Face Recognition and Clustering # FaceNet: A Unified Embedding for Face Recognition and Clustering **Authors:** Florian Schroff, [Dmitry Kalenichenko](https://md.research.elamur.ai/free/en/authors/dmitry-kalenichenko/), James Philbin · **Year:** 2015 · **Venue:** arXiv:1503.03832 · **Licence:** [arXiv — non-exclusive licence to distribute](http://arxiv.org/licenses/nonexclusive-distrib/1.0/) **Links:** [Publisher's page](https://doi.org/10.1109/cvpr.2015.7298682) · [PDF](https://arxiv.org/pdf/1503.03832) · [DOI 10.1109/cvpr.2015.7298682](https://doi.org/10.1109/cvpr.2015.7298682) · [arXiv 1503.03832](https://arxiv.org/abs/1503.03832) · [OpenAlex](https://openalex.org/W2096733369) **Citations:** 11681 citations (openalex-snapshot · 2026-10-14) · [257](#cited-by) in our library, of the 140673 papers whose bibliographies we read ## Summary Most deep face recognition systems train a network to classify known identities. They then reuse an intermediate bottleneck layer as the face descriptor, often adding PCA, SVMs or a Joint Bayesian model. This is indirect and yields large representations. FaceNet, from Google, trains a deep convolutional network end to end so that its L2-normalised output is a compact 128-D embedding on a hypersphere, where squared Euclidean distance measures face similarity. Training uses a triplet loss related to LMNN: an anchor face must be closer to a positive of the same person than to a negative of another person, by a fixed margin. Most triplets are trivially satisfied, so the authors select hard and semi-hard examples online inside large mini-batches. They test two backbones, Zeiler&Fergus-style and Inception-style networks. Inputs are just tight face crops, with no 2D or 3D alignment. The system sets new records on LFW and YouTube Faces DB. With such embeddings, verification, k-NN recognition and clustering become simple. The paper also introduces harmonic embeddings, which keep different model versions directly comparable. ## Key points - Accuracy of 99.63% on Labeled Faces in the Wild (LFW) and 95.12% on YouTube Faces DB, about 30% lower error than the previous best, DeepId2+. - Each face is represented by a 128-D embedding that fits in only 128 bytes. Verification is a distance threshold, recognition is k-NN, and clustering uses k-means or agglomerative methods. - Online triplet mining uses mini-batches of a few thousand faces, with roughly 40 images per identity, following a curriculum-like schedule of increasingly difficult triplets. - Inception-based models need up to 20× fewer parameters than Zeiler&Fergus-style networks for comparable performance. - A harmonic triplet loss lets an improved model (NN2) produce embeddings compatible with an older one (NN1). Mixed NN2–NN1 comparisons on the internal 3G dataset still beat NN1 alone. *Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.* Fields: [Other](https://md.research.elamur.ai/free/en/shelves/other/) ## Cited by Cited by 257 papers of our library cite it (of the 140673 whose bibliographies we read); 10 of them have a page here, 247 not yet. - [A Simple Framework for Contrastive Learning of Visual Representations](https://md.research.elamur.ai/free/en/papers/chen-2020-simple-framework-contrastive-learning-visual/) - [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://md.research.elamur.ai/free/en/papers/howard-2017-mobilenets-efficient-convolutional-neural-networks/) - [POLITICS: Pretraining with Same-story Article Comparison for Ideology Prediction and Stance Detection](https://md.research.elamur.ai/free/en/papers/liu-2022-politics-pretraining-same-story-article/) - [Facilitating Contrastive Learning of Discourse Relational Senses by Exploiting the Hierarchy of Sense Relations](https://md.research.elamur.ai/free/en/papers/long-2022-facilitating-contrastive-learning-discourse-relational/) - [AStitchInLanguageModels: Dataset and Methods for the Exploration of Idiomaticity in Pre-Trained Language Models](https://md.research.elamur.ai/free/en/papers/madabushi-2021-astitchinlanguagemodels-dataset-methods-exploration-idiomaticity/) - [Text and Code Embeddings by Contrastive Pre-Training](https://md.research.elamur.ai/free/en/papers/neelakantan-2022-text-code-embeddings-contrastive-pre/) - [Representation Learning with Contrastive Predictive Coding](https://md.research.elamur.ai/free/en/papers/oord-2018-representation-learning-contrastive-predictive-coding/) - [COMET: A Neural Framework for MT Evaluation](https://md.research.elamur.ai/free/en/papers/rei-2020-comet-neural-framework-mt-evaluation/) - [Multimodal Language Analysis in the Wild: CMU-MOSEI Dataset and Interpretable Dynamic Fusion Graph](https://md.research.elamur.ai/free/en/papers/zadeh-2018-multimodal-language-analysis-wild-cmu/) - [Learning Transferable Architectures for Scalable Image Recognition](https://md.research.elamur.ai/free/en/papers/zoph-2017-learning-transferable-architectures-scalable-image/) ## Cite The paper itself, not this page. BibTeX from the library's record of the work. ```bibtex @misc{schroff2015facenet, title = {{FaceNet: A Unified Embedding for Face Recognition and Clustering}}, author = {Schroff, Florian and Kalenichenko, Dmitry and Philbin, James}, year = {2015}, doi = {10.1109/cvpr.2015.7298682}, eprint = {1503.03832}, archivePrefix = {arXiv}, url = {https://doi.org/10.1109/cvpr.2015.7298682} } ``` - [← previous Exploring the Space of Topic Coherence Measures](https://md.research.elamur.ai/free/en/papers/roder-2015-exploring-space-topic-coherence-measures/) - [next → Factors Influencing the Implicitation of Discourse Relations across Languages](https://md.research.elamur.ai/free/en/papers/hoek-2015-factors-influencing-implicitation-discourse-relations/) - [↑ back to papers of 2015](https://md.research.elamur.ai/free/en/papers/2015/) ## References 23 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. Y. Bengio, J. Louradour, R. Collobert, and J. Weston. Curriculum learning. In Proc. of ICML, New York, NY, USA, 2009. 2 [find](https://scholar.google.com/scholar?q=Y%2E+Bengio%2C+J%2E+Louradour%2C+R%2E+Collobert%2C+and+J%2E+Weston%2E+Curriculum+learning%2E+In+Proc%2E+of+ICML%2C+New+York%2C+NY%2C+USA%2C+2009%2E+2) [our card](https://md.research.elamur.ai/free/en/papers/bengio-2009-curriculum-learning/) 2. D. Chen, X. Cao, L. Wang, F. Wen, and J. Sun. Bayesian face revisited: A joint formulation. In Proc. ECCV, 2012. 2 [find](https://scholar.google.com/scholar?q=D%2E+Chen%2C+X%2E+Cao%2C+L%2E+Wang%2C+F%2E+Wen%2C+and+J%2E+Sun%2E+Bayesian+face+revisited%3A+A+joint+formulation%2E+In+Proc%2E+ECCV%2C+2012%2E+2) 3. D. Chen, S. Ren, Y. Wei, X. Cao, and J. Sun. Joint cascade face detection and alignment. In Proc. ECCV, 2014. 7 [find](https://scholar.google.com/scholar?q=D%2E+Chen%2C+S%2E+Ren%2C+Y%2E+Wei%2C+X%2E+Cao%2C+and+J%2E+Sun%2E+Joint+cascade+face+detection+and+alignment%2E+In+Proc%2E+ECCV%2C+2014%2E+7) 4. J. Dean, G. Corrado, R. Monga, K. Chen, M. Devin, M. Mao, M. Ranzato, A. Senior, P. Tucker, K. Yang, Q. V. Le, and A. Y. Ng. Large scale distributed deep networks. In P. Bartlett, F. Pereira, C. Burges, L. Bottou, and K. Weinberger, editors, NIPS, pages 1232–1240. 2012. 10 [find](https://scholar.google.com/scholar?q=J%2E+Dean%2C+G%2E+Corrado%2C+R%2E+Monga%2C+K%2E+Chen%2C+M%2E+Devin%2C+M%2E+Mao%2C+M%2E+Ranzato%2C+A%2E+Senior%2C+P%2E+Tucker%2C+K%2E+Yang%2C+Q%2E+V%2E+Le%2C+and+A%2E+Y%2E+Ng%2E+Large+scale+distributed+deep+networks%2E+In+P%2E+Bartlett%2C+F%2E+Pereira%2C+C%2E+Burge) 5. J. Duchi, E. Hazan, and Y. Singer. Adaptive subgradient methods for online learning and stochastic optimization. J. Mach. Learn. Res., 12:2121–2159, July 2011. 4 [find](https://scholar.google.com/scholar?q=J%2E+Duchi%2C+E%2E+Hazan%2C+and+Y%2E+Singer%2E+Adaptive+subgradient+methods+for+online+learning+and+stochastic+optimization%2E+J%2E+Mach%2E+Learn%2E+Res%2E%2C+12%3A2121%E2%80%932159%2C+July+2011%2E+4) 6. I. J. Goodfellow, D. Warde-farley, M. Mirza, A. Courville, and Y. Bengio. Maxout networks. In In ICML, 2013. 4 [find](https://scholar.google.com/scholar?q=I%2E+J%2E+Goodfellow%2C+D%2E+Warde%2Dfarley%2C+M%2E+Mirza%2C+A%2E+Courville%2C+and+Y%2E+Bengio%2E+Maxout+networks%2E+In+In+ICML%2C+2013%2E+4) 7. G. B. Huang, M. Ramesh, T. Berg, and E. Learned-Miller. Labeled faces in the wild: A database for studying face recognition in unconstrained environments. Technical Report 07-49, University of Massachusetts, Amherst, October 2007. 5 [find](https://scholar.google.com/scholar?q=G%2E+B%2E+Huang%2C+M%2E+Ramesh%2C+T%2E+Berg%2C+and+E%2E+Learned%2DMiller%2E+Labeled+faces+in+the+wild%3A+A+database+for+studying+face+recognition+in+unconstrained+environments%2E+Technical+Report+07%2D49%2C+University+of+Massach) 8. Y. LeCun, B. Boser, J. S. Denker, D. Henderson, R. E. Howard, W. Hubbard, and L. D. Jackel. Backpropagation applied to handwritten zip code recognition. Neural Computation, 1(4):541–551, Dec. 1989. 2, 4 [find](https://scholar.google.com/scholar?q=Y%2E+LeCun%2C+B%2E+Boser%2C+J%2E+S%2E+Denker%2C+D%2E+Henderson%2C+R%2E+E%2E+Howard%2C+W%2E+Hubbard%2C+and+L%2E+D%2E+Jackel%2E+Backpropagation+applied+to+handwritten+zip+code+recognition%2E+Neural+Computation%2C+1%284%29%3A541%E2%80%93551%2C+Dec%2E+1989%2E+2%2C) 9. M. Lin, Q. Chen, and S. Yan. Network in network. CoRR, abs/1312.4400, 2013. 2, 4, 6 — [arXiv:1312.4400](https://arxiv.org/abs/1312.4400) [find](https://scholar.google.com/scholar?q=M%2E+Lin%2C+Q%2E+Chen%2C+and+S%2E+Yan%2E+Network+in+network%2E+CoRR%2C+abs%2F1312%2E4400%2C+2013%2E+2%2C+4%2C+6+%E2%80%94+%5BarXiv%3A1312%2E4400%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F1312%2E4400%29) 10. C. Lu and X. Tang. Surpassing human-level face verification performance on LFW with gaussianface. CoRR, abs/1404.3840, 2014. 1 — [arXiv:1404.3840](https://arxiv.org/abs/1404.3840) [find](https://scholar.google.com/scholar?q=C%2E+Lu+and+X%2E+Tang%2E+Surpassing+human%2Dlevel+face+verification+performance+on+LFW+with+gaussianface%2E+CoRR%2C+abs%2F1404%2E3840%2C+2014%2E+1+%E2%80%94+%5BarXiv%3A1404%2E3840%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F1404%2E3840%29) 11. D. E. Rumelhart, G. E. Hinton, and R. J. Williams. Learning representations by back-propagating errors. Nature, 1986. 2, 4 [find](https://scholar.google.com/scholar?q=D%2E+E%2E+Rumelhart%2C+G%2E+E%2E+Hinton%2C+and+R%2E+J%2E+Williams%2E+Learning+representations+by+back%2Dpropagating+errors%2E+Nature%2C+1986%2E+2%2C+4) 12. M. Schultz and T. Joachims. Learning a distance metric from relative comparisons. In S. Thrun, L. Saul, and B. Schölkopf, editors, NIPS, pages 41–48. MIT Press, 2004. 2 [find](https://scholar.google.com/scholar?q=M%2E+Schultz+and+T%2E+Joachims%2E+Learning+a+distance+metric+from+relative+comparisons%2E+In+S%2E+Thrun%2C+L%2E+Saul%2C+and+B%2E+Sch%C3%B6lkopf%2C+editors%2C+NIPS%2C+pages+41%E2%80%9348%2E+MIT+Press%2C+2004%2E+2) 13. T. Sim, S. Baker, and M. Bsat. The CMU pose, illumination, and expression (PIE) database. In In Proc. FG, 2002. 2 [find](https://scholar.google.com/scholar?q=T%2E+Sim%2C+S%2E+Baker%2C+and+M%2E+Bsat%2E+The+CMU+pose%2C+illumination%2C+and+expression+%28PIE%29+database%2E+In+In+Proc%2E+FG%2C+2002%2E+2) 14. Y. Sun, X. Wang, and X. Tang. Deep learning face representation by joint identification-verification. CoRR, abs/1406.4773, 2014. 1, 2, 3 — [arXiv:1406.4773](https://arxiv.org/abs/1406.4773) [find](https://scholar.google.com/scholar?q=Y%2E+Sun%2C+X%2E+Wang%2C+and+X%2E+Tang%2E+Deep+learning+face+representation+by+joint+identification%2Dverification%2E+CoRR%2C+abs%2F1406%2E4773%2C+2014%2E+1%2C+2%2C+3+%E2%80%94+%5BarXiv%3A1406%2E4773%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F1406%2E4773%29) 15. Y. Sun, X. Wang, and X. Tang. Deeply learned face representations are sparse, selective, and robust. CoRR, abs/1412.1265, 2014. 1, 2, 5, 8 — [arXiv:1412.1265](https://arxiv.org/abs/1412.1265) [find](https://scholar.google.com/scholar?q=Y%2E+Sun%2C+X%2E+Wang%2C+and+X%2E+Tang%2E+Deeply+learned+face+representations+are+sparse%2C+selective%2C+and+robust%2E+CoRR%2C+abs%2F1412%2E1265%2C+2014%2E+1%2C+2%2C+5%2C+8+%E2%80%94+%5BarXiv%3A1412%2E1265%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F1412%2E1265%29) 16. C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, and A. Rabinovich. Going deeper with convolutions. CoRR, abs/1409.4842, 2014. 2, 3, 4, 5, 6, 10 — [arXiv:1409.4842](https://arxiv.org/abs/1409.4842) [find](https://scholar.google.com/scholar?q=C%2E+Szegedy%2C+W%2E+Liu%2C+Y%2E+Jia%2C+P%2E+Sermanet%2C+S%2E+Reed%2C+D%2E+Anguelov%2C+D%2E+Erhan%2C+V%2E+Vanhoucke%2C+and+A%2E+Rabinovich%2E+Going+deeper+with+convolutions%2E+CoRR%2C+abs%2F1409%2E4842%2C+2014%2E+2%2C+3%2C+4%2C+5%2C+6%2C+10+%E2%80%94+%5BarXiv%3A1409%2E484) 17. Y. Taigman, M. Yang, M. Ranzato, and L. Wolf. Deepface: Closing the gap to human-level performance in face verification. In IEEE Conf. on CVPR, 2014. 1, 2, 5, 7, 8, 9 [find](https://scholar.google.com/scholar?q=Y%2E+Taigman%2C+M%2E+Yang%2C+M%2E+Ranzato%2C+and+L%2E+Wolf%2E+Deepface%3A+Closing+the+gap+to+human%2Dlevel+performance+in+face+verification%2E+In+IEEE+Conf%2E+on+CVPR%2C+2014%2E+1%2C+2%2C+5%2C+7%2C+8%2C+9) 18. J. Wang, Y. Song, T. Leung, C. Rosenberg, J. Wang, J. Philbin, B. Chen, and Y. Wu. Learning fine-grained image similarity with deep ranking. CoRR, abs/1404.4661, 2014. 2 — [arXiv:1404.4661](https://arxiv.org/abs/1404.4661) [find](https://scholar.google.com/scholar?q=J%2E+Wang%2C+Y%2E+Song%2C+T%2E+Leung%2C+C%2E+Rosenberg%2C+J%2E+Wang%2C+J%2E+Philbin%2C+B%2E+Chen%2C+and+Y%2E+Wu%2E+Learning+fine%2Dgrained+image+similarity+with+deep+ranking%2E+CoRR%2C+abs%2F1404%2E4661%2C+2014%2E+2+%E2%80%94+%5BarXiv%3A1404%2E4661%5D%28https%3A%2F%2Far) 19. K. Q. Weinberger, J. Blitzer, and L. K. Saul. Distance metric learning for large margin nearest neighbor classification. In NIPS. MIT Press, 2006. 2, 3 [find](https://scholar.google.com/scholar?q=K%2E+Q%2E+Weinberger%2C+J%2E+Blitzer%2C+and+L%2E+K%2E+Saul%2E+Distance+metric+learning+for+large+margin+nearest+neighbor+classification%2E+In+NIPS%2E+MIT+Press%2C+2006%2E+2%2C+3) 20. D. R. Wilson and T. R. Martinez. The general inefficiency of batch training for gradient descent learning. Neural Networks, 16(10):1429–1451, 2003. 4 [find](https://scholar.google.com/scholar?q=D%2E+R%2E+Wilson+and+T%2E+R%2E+Martinez%2E+The+general+inefficiency+of+batch+training+for+gradient+descent+learning%2E+Neural+Networks%2C+16%2810%29%3A1429%E2%80%931451%2C+2003%2E+4) 21. L. Wolf, T. Hassner, and I. Maoz. Face recognition in unconstrained videos with matched background similarity. In IEEE Conf. on CVPR, 2011. 5 [find](https://scholar.google.com/scholar?q=L%2E+Wolf%2C+T%2E+Hassner%2C+and+I%2E+Maoz%2E+Face+recognition+in+unconstrained+videos+with+matched+background+similarity%2E+In+IEEE+Conf%2E+on+CVPR%2C+2011%2E+5) 22. M. D. Zeiler and R. Fergus. Visualizing and understanding convolutional networks. CoRR, abs/1311.2901, 2013. 2, 3, 4, 6 [find](https://scholar.google.com/scholar?q=M%2E+D%2E+Zeiler+and+R%2E+Fergus%2E+Visualizing+and+understanding+convolutional+networks%2E+CoRR%2C+abs%2F1311%2E2901%2C+2013%2E+2%2C+3%2C+4%2C+6) [our card](https://md.research.elamur.ai/free/en/papers/zeiler-2013-visualizing-understanding-convolutional-networks/) 23. Z. Zhu, P. Luo, X. Wang, and X. Tang. Recover canonical-view faces in the wild with deep neural networks. CoRR, abs/1404.3543, 2014. 2 — [arXiv:1404.3543](https://arxiv.org/abs/1404.3543) [find](https://scholar.google.com/scholar?q=Z%2E+Zhu%2C+P%2E+Luo%2C+X%2E+Wang%2C+and+X%2E+Tang%2E+Recover+canonical%2Dview+faces+in+the+wild+with+deep+neural+networks%2E+CoRR%2C+abs%2F1404%2E3543%2C+2014%2E+2+%E2%80%94+%5BarXiv%3A1404%2E3543%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F1404%2E3543%29)