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.”
BackgroundMethodInfluential
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
Items 1,001–1,094 of 1,094
Attribute-Enhanced Face Recognition with Neural Tensor Fusion Networks
Hu, G., Hua, Y., Yuan, Y., Zhang, Z., Lu, Z., Mukherjee, S. S., Hospedales, T. M., Robertson, N. M., & Yang, Y. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.404 · Cited by 86
“Deep learning methodologies [41, 40, 42, 32] have proven particularly effective recently, thanks to end-to-end repreFigure 1: A sample attribute list is given (col.”
BackgroundMethodDesigning efficient accelerator of depthwise separable convolutional neural network on FPGA
Ding, W., Huang, Z., Huang, Z., Tian, L., Wang, H., & Feng, S. (2019) · Journal of Systems Architecture · doi:10.1016/j.sysarc.2018.12.008 · Cited by 85
Human activity recognition based on smartphone and wearable sensors using multiscale DCNN ensemble
Sena, J., Barreto, J., Caetano, C., Cramer, G., & Schwartz, W. R. (2021) · Neurocomputing · doi:10.1016/j.neucom.2020.04.151 · Cited by 81
An efficient deepfake video detection using robust deep learning
Qadir, A., Mahum, R., El-Meligy, M. A., Ragab, A. E., AlSalman, A., & Awais, M. (2024) · Heliyon · doi:10.1016/j.heliyon.2024.e25757 · Cited by 78
Williford, J. R., May, B. B., & Byrne, J. (2020) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-58621-8_15 · Cited by 76
Longitudinal Study of Child Face Recognition
Deb, D., Nain, N., & Jain, A. K. (2018) · 2018 International Conference on Biometrics (ICB) · doi:10.1109/icb2018.2018.00042 · Cited by 73
Uncertainty-Aware Mesh Decoder for High Fidelity 3D Face Reconstruction
Lee, G.-H., & Lee, S.-W. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00614 · Cited by 70
“Reconstructing a high quality personalized 3D face can be used for many applications, including face recognition [11, 21, 16, 14, 36], facial motion capture [9] or virtual and augmented reality [34, 33].”
BackgroundSurvey: Leakage and Privacy at Inference Time
Jegorova, M., Kaul, C., Mayor, C., O'Neil, A. Q., Weir, A., Murray-Smith, R., & Tsaftaris, S. A. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3229593 · Cited by 77
Cloud Versus Edge Deployment Strategies of Real-Time Face Recognition Inference
Koubaa, A., Ammar, A., Kanhouch, A., & AlHabashi, Y. (2022) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2021.3055835 · Cited by 69
MultiGrain: a unified image embedding for classes and instances
Berman, M., Jégou, H., Vedaldi, A., Kokkinos, I., & Douze, M. (2019) · arXiv · doi:10.48550/arxiv.1902.05509 · Cited by 69
“The triplet loss instead requires an image to be closer to a positive sibling than to a negative sibling [38], which is relative property of image triplets.”
BackgroundProperty Inference from Poisoning
Mahloujifar, S., Ghosh, E., & Chase, M. (2022) · 2022 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp46214.2022.9833623 · Cited by 67
Gao, Z., Guo, L., Guan, W., Liu, A.-A., Ren, T., & Chen, S. (2021) · IEEE Transactions on Image Processing · doi:10.1109/tip.2020.3038372 · Cited by 63
Solving Inefficiency of Self-supervised Representation Learning
Wang, G., Wang, K., Wang, G., Torr, P. H. S., & Lin, L. (2021) · 2021 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00937 · Cited by 65
“Recognizing the hardest negatives can lead to bad lo-cal minima in practice, [37] proposes a semi-hard negative sampling strategy, sharing the merit of our method in avoiding over-trusting the hardest negative sample and benefiting representation learning.”
BackgroundMethodInfluentialA survey on online learning for visual tracking
Abbass, M. Y., Kwon, K.-C., Kim, N., Abdelwahab, S. A., El-Samie, F. E. A., & Khalaf, A. A. M. (2021) · The Visual Computer · doi:10.1007/s00371-020-01848-y · Cited by 64
“These include computer vision [82, 83], speech recognition [84, 85], and natural language processing [86, 87].”
BackgroundA Review of Image Recognition with Deep Convolutional Neural Network
Liu, Q., Zhang, N., Yang, W., Wang, S., Cui, Z., Chen, X., & Chen, L. (2017) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-63309-1_7 · Cited by 64
Audio-Visual Deep Neural Network for Robust Person Verification
Qian, Y., Chen, Z., & Wang, S. (2021) · IEEE/ACM Transactions on Audio, Speech, and Language Processing · doi:10.1109/taslp.2021.3057230 · Cited by 63
“Intherecentyears,researchershaveexplored otherDCNNbasedneuralnetworktoimproveFRsystem,such asDeepIDseries[36]–[39],VGGFace[40]andFaceNet[41].”
BackgroundOn the Intrinsic Dimensionality of Image Representations
Gong, S., Boddeti, V. N., & Jain, A. K. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00411 · Cited by 59
Learning Discriminative Features from Spectrograms Using Center Loss for Speech Emotion Recognition
Dai, D., Wu, Z., Li, R., Wu, X., Jia, J., & Meng, H. (2019) · ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2019.8683765 · Cited by 58
“In [11], a triplet framework was proposed to extract discriminative features by using triplet loss[12], whose input was triplets including two utterances from the same emotion class and one utterance from other classes.”
MethodNeighborhood Contrastive Learning for Scientific Document Representations with Citation Embeddings
Ostendorff, M., Rethmeier, N., Augenstein, I., Gipp, B., & Rehm, G. (2022) · Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2022.emnlp-main.802 · Cited by 56
Domain Generalization via Aggregation and Separation for Audio Deepfake Detection
Xie, Y., Cheng, H., Wang, Y., & Ye, L. (2024) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2023.3324724 · Cited by 59
Modeling Uncertainty with Hedged Instance Embedding
Oh, S. J., Murphy, K., Pan, J., Roth, J., Schroff, F., & Gallagher, A. (2018) · arXiv · doi:10.48550/arxiv.1810.00319 · Cited by 55
Self-citationSpoC: Spoofing Camera Fingerprints
Cozzolino, D., Thies, J., Rossler, A., Niesner, M., & Verdoliva, L. (2021) · 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw53098.2021.00110 · Cited by 54
“However, from the literature on triplet loss, it is well known that better results are obtained by comparing distances [48].”
MethodCatastrophic Forgetting in Deep Learning: A Comprehensive Taxonomy
Aleixo, E. L., Colonna, J. G., Cristo, M., & Fernandes, E. (2023) · arXiv · doi:10.48550/arxiv.2312.10549 · Cited by 54
“This technique proved to be resilient to CF in toy data sets, like MNIST, where the dimension of the input data, (28, 28) pixels, is not as big as it could be in real applications, such as facial recognition [Schroff et al ., 2015].”
MethodLi, H.-C., Deng, Z.-Y., & Chiang, H.-H. (2020) · Sensors · doi:10.3390/s20216114 · Cited by 54
“Additionally, FN13 modifies the online mining method used in FaceNet [2] to generate center loss”
MethodDemystifying TensorRT: Characterizing Neural Network Inference Engine on Nvidia Edge Devices
Shafi, O., Rai, C., Sen, R., & Ananthanarayanan, G. (2021) · 2021 IEEE International Symposium on Workload Characterization (IISWC) · doi:10.1109/iiswc53511.2021.00030 · Cited by 51
Individual identification in acoustic recordings
Knight, E., Rhinehart, T., de Zwaan, D. R., Weldy, M. J., Cartwright, M., Hawley, S. H., Larkin, J. L., Lesmeister, D., Bayne, E., & Kitzes, J. (2024) · Trends in Ecology & Evolution · doi:10.1016/j.tree.2024.05.007 · Cited by 54
Contrastive Learning Models for Sentence Representations
Xu, L., Xie, H., Li, Z., Wang, F. L., Wang, W., & Li, Q. (2023) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3593590 · Cited by 50
“Although positive pairs are important for the success of contrastive learning, negative pairs, particularly hard negatives, are equally significant [ 50 , 77 , 91 , 106 , 122 , 141 ].”
BackgroundCatastrophic Forgetting in Deep Learning: A Comprehensive Taxonomy
Aleixo, E. L., Colonna, J. G., Cristo, M., & Fernandes, E. (2024) · Journal of the Brazilian Computer Society · doi:10.5753/jbcs.2024.3966 · Cited by 54
Parameter Efficient Multimodal Transformers for Video Representation Learning
Lee, S., Yu, Y., Kim, G., Breuel, T., Kautz, J., & Song, Y. (2020) · arXiv · doi:10.48550/arxiv.2012.04124 · Cited by 49
“Hard negative mining has been shown to be crucial for contrastive learning (Arandjelovic & Zisserman, 2017; Owens & Efros, 2018; Korbar et al., 2018; Schroff et al., 2015; Zhuang et al., 2019; Morgado et al., 2020; Wu et al., 2020).”
BackgroundPeng, T., Boxberg, M., Weichert, W., Navab, N., & Marr, C. (2019) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-32239-7_75 · Cited by 48
Compressive Privacy Generative Adversarial Network
Tseng, B.-W., & Wu, P.-Y. (2020) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2020.2968188 · Cited by 48
A Review on Fact Extraction and Verification
Bekoulis, G., Papagiannopoulou, C., & Deligiannis, N. (2023) · ACM Computing Surveys · doi:10.1145/3485127 · Cited by 48
Fan, X., Jiang, W., Luo, H., & Mao, W. (2022) · The Visual Computer · doi:10.1007/s00371-020-02015-z · Cited by 45
“Motivated by the success of triplet loss [37] in the face recognition area [27], some variant of triplet losses have been introduced into person ReID [1,3,6,20], which can push negative pairs away and pull positive pairs together.”
BackgroundModeling multimodal cues in a deep learning-based framework for emotion recognition in the wild
Pini, S., Ahmed, O. B., Cornia, M., Baraldi, L., Cucchiara, R., & Huet, B. (2017) · Proceedings of the 19th ACM International Conference on Multimodal Interaction · doi:10.1145/3136755.3143006 · Cited by 44
“Recently, with the rapid growth of Convolutional Neural Networks (CNNs), extracting visual features from video frames has been investigated in many emotion recognition tasks and there are various face pretrained models made available [34, 36, 40].”
BackgroundMethodSanabria, A. R., Zambonelli, F., Dobson, S., & Ye, J. (2021) · Pervasive and Mobile Computing · doi:10.1016/j.pmcj.2021.101477 · Cited by 43
“[40] introduce a triplet loss in feature extraction for face images.”
BackgroundFritz, B. A., Chen, Y., Murray-Torres, T. M., Gregory, S., Ben Abdallah, A., Kronzer, A., McKinnon, S. L., Budelier, T., Helsten, D. L., Wildes, T. S., et al. (2018) · BMJ Open · doi:10.1136/bmjopen-2017-020124 · Cited by 42
“Deep learning methods, especially those using convolutional neural networks (CNNs),(37) have achieved great success in learning useful representations (features) from images.(38) 39 However, its uses in time-series classification are very limited.”
MethodInvestigating bias in deep face analysis: The KANFace dataset and empirical study
Georgopoulos, M., Panagakis, Y., & Pantic, M. (2020) · Image and Vision Computing · doi:10.1016/j.imavis.2020.103954 · Cited by 42
Classification in the Wild: The Science and Art of Transparent Decision Making
Katsikopoulos, K. V., Şimşek, Ö., Buckmann, M., & Gigerenzer, G. (2021) · doi:10.7551/mitpress/11790.001.0001 · Cited by 42
Heimberg, G., Kuo, T., DePianto, D., Heigl, T., Diamant, N., Salem, O., Scalia, G., Biancalani, T., Turley, S., Rock, J., et al. (2023) · doi:10.1101/2023.07.18.549537 · Cited by 40
Visual Deepfake Detection: Review of Techniques, Tools, Limitations, and Future Prospects
Ur Rehman Ahmed, N., Badshah, A., Adeel, H., Tajammul, A., Daud, A., & Alsahfi, T. (2025) · IEEE Access · doi:10.1109/access.2024.3523288 · Cited by 45
Deep Insights of Deepfake Technology : A Review
Mahmud, B. U., & Sharmin, A. (2021) · arXiv · doi:10.48550/arxiv.2105.00192 · Cited by 38
“As example deep learning based face recognition technique VGG [68] and Facenet [69] are unable to detect Deepfakes properly.”
MethodAccessory pathway analysis using a multimodal deep learning model
Nishimori, M., Kiuchi, K., Nishimura, K., Kusano, K., Yoshida, A., Adachi, K., Hirayama, Y., Miyazaki, Y., Fujiwara, R., Sommer, P., et al. (2021) · Scientific Reports · doi:10.1038/s41598-021-87631-y · Cited by 36
“It is relatively simple to collect data in authenticating ubiquitous objects, such as cars and human faces 10 , but the amount of medical data that we can collect is limited because of the limited number of patients.”
BackgroundKaur, S., Kumar, P., & Kumaraguru, P. (2020) · Journal of Electronic Imaging · doi:10.1117/1.jei.29.3.033013 · Cited by 35
Mix-and-Match Tuning for Self-Supervised Semantic Segmentation
Zhan, X., Liu, Z., Luo, P., Tang, X., & Loy, C. (2018) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v32i1.12331 · Cited by 34
MINTIME: Multi-Identity Size-Invariant Video Deepfake Detection
Coccomini, D. A., Zilos, G. K., Amato, G., Caldelli, R., Falchi, F., Papadopoulos, S., & Gennaro, C. (2024) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2024.3409054 · Cited by 33
“Firstly, the features of each face are extracted via an InceptionResnetV1 pretrained on FaceNet [41].”
MethodMultimodal Metric Learning for Tag-Based Music Retrieval
Won, M., Oramas, S., Nieto, O., Gouyon, F., & Serra, X. (2021) · ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp39728.2021.9413514 · Cited by 33
Advani, S., Zientara, P., Shukla, N., Okafor, I., Irick, K., Sampson, J., Datta, S., & Narayanan, V. (2017) · IEEE Consumer Electronics Magazine · doi:10.1109/mce.2016.2614422 · Cited by 32
“Another important insight, recently unearthed by [5], is that the accuracies of CNNs can saturate after a few million images of training data.”
BackgroundAppearance Matters, So Does Audio: Revealing the Hidden Face via Cross-Modality Transfer
Kong, C., Chen, B., Yang, W., Li, H., Chen, P., & Wang, S. (2022) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2021.3057457 · Cited by 32
Mingote, V., Miguel, A., Ortega, A., & Lleida, E. (2020) · Computer Speech & Language · doi:10.1016/j.csl.2020.101078 · Cited by 32
Moving in the Right Direction: A Regularization for Deep Metric Learning
Mohan, D. D., Sankaran, N., Fedorishin, D., Setlur, S., & Govindaraju, V. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.01460 · Cited by 31
“Semi-hard negative mining [17] for face recognition looks at specific triplets that violate the triplet margin constraint.”
BackgroundInfluentialCNN-Based Multimodal Human Recognition in Surveillance Environments
Koo, J. H., Cho, S. W., Baek, N. R., Kim, M. C., & Park, K. R. (2018) · Sensors · doi:10.3390/s18093040 · Cited by 30
AdapterSoup: Weight Averaging to Improve Generalization of Pretrained Language Models
Chronopoulou, A., Peters, M., Fraser, A., & Dodge, J. (2023) · Findings of the Association for Computational Linguistics: EACL 2023 · doi:10.18653/v1/2023.findings-eacl.153 · Cited by 30
Diverse types of expertise in facial recognition
Towler, A., Dunn, J. D., Castro Martínez, S., Moreton, R., Eklöf, F., Ruifrok, A., Kemp, R. I., & White, D. (2023) · Scientific Reports · doi:10.1038/s41598-023-28632-x · Cited by 30
Kaarmukilan, S. P., Poddar, S., & Thomas K., A. (2020) · 2020 Fourth International Conference on Computing Methodologies and Communication (ICCMC) · doi:10.1109/iccmc48092.2020.iccmc-00088 · Cited by 29
Learning Visual Robotic Control Efficiently with Contrastive Pre-training and Data Augmentation
Zhan, A., Zhao, R., Pinto, L., Abbeel, P., & Laskin, M. (2020) · arXiv · doi:10.48550/arxiv.2012.07975 · Cited by 28
Learning Universal Authorship Representations
Rivera-Soto, R. A., Miano, O. E., Ordonez, J., Chen, B. Y., Khan, A., Bishop, M., & Andrews, N. (2021) · Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2021.emnlp-main.70 · Cited by 25
“This formulation includes several common metric learning losses as special cases, such as triplet loss (Schroff et al., 2015).”
BackgroundTowards Automatic Boundary Detection for Human-AI Collaborative Hybrid Essay in Education
Zeng, Z., Sha, L., Li, Y., Yang, K., Gašević, D., & Chen, G. (2023) · arXiv · doi:10.48550/arxiv.2307.12267 · Cited by 25
“We then fine-tune the initial encoder with the triplet Bert-networks (Schroff, Kalenichenko, and Philbin 2015) architecture, which is described as follows: for a sentence triplet ( a, x + , x − ) , where a is the anchor sentence whose label is the same as that of the positive sentence x + , but…”
MethodMissing the Forest for the Trees: End-to-End AI Application Performance in Edge Data Centers
Richins, D., Doshi, D., Blackmore, M., Thulaseedharan Nair, A., Pathapati, N., Patel, A., Daguman, B., Dobrijalowski, D., Illikkal, R., Long, K., et al. (2020) · 2020 IEEE International Symposium on High Performance Computer Architecture (HPCA) · doi:10.1109/hpca47549.2020.00049 · Cited by 23
“The Face Recognition application consists of four primary processing stages (Figure 1a) and is built from MT-CNNs (multi-task cascaded convolutional networks) [25] along with Google’s FaceNet [13, 26].”
MethodTie your embeddings down: Cross-modal latent spaces for end-to-end spoken language understanding
Agrawal, B., Muller, M., Choudhary, S., Radfar, M., Mouchtaris, A., McGowan, R., Susanj, N., & Kunzmann, S. (2022) · ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp43922.2022.9747759 · Cited by 23
Likelihood ratios for deep neural networks in face comparison
Macarulla Rodriguez, A., Geradts, Z., & Worring, M. (2020) · Journal of Forensic Sciences · doi:10.1111/1556-4029.14324 · Cited by 21
“(21), but OpenFace has faster running time than FaceNet because of its lower number of dimensions.”
Deep Neural Network Model and FPGA Accelerator Co-Design: Opportunities and Challenges
Hao, C., & Chen, D. (2018) · 2018 14th IEEE International Conference on Solid-State and Integrated Circuit Technology (ICSICT) · doi:10.1109/icsict.2018.8564956 · Cited by 19
“Based on the quantitative analysis, we mapped the FaceNet [16] on FPGA, where the IPs are configured to obtain best performance for different network components.”
MethodAttention-Based Deep Metric Learning for Near-Duplicate Video Retrieval
Wang, K.-H., Cheng, C.-C., Chen, Y.-L., Song, Y., & Lai, S.-H. (2021) · 2020 25th International Conference on Pattern Recognition (ICPR) · doi:10.1109/icpr48806.2021.9412710 · Cited by 19
“Here, we apply online hard negative mining [36] to sample negatives in every mini-batch.”
MethodCalibration of score based likelihood ratio estimation in automated forensic facial image comparison
Rodriguez, A. M., Geradts, Z., & Worring, M. (2022) · Forensic Science International · doi:10.1016/j.forsciint.2022.111239 · Cited by 19
Face Recognition for Newborns, Toddlers, and Pre-School Children: A Deep Learning Approach
Siddiqui, S., Vatsa, M., & Singh, R. (2018) · 2018 24th International Conference on Pattern Recognition (ICPR) · doi:10.1109/icpr.2018.8545742 · Cited by 19
“We have implemented our own version of the deep convolutional network as described in [19] where cosine distance is used to calculate face similarity while computing the triplet loss.”
MethodText Transformations in Contrastive Self-Supervised Learning: A Review
Bhattacharjee, A., Karami, M., & Liu, H. (2022) · Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2022/757 · Cited by 20
Training Socially Aligned Language Models on Simulated Social Interactions
Liu, R., Yang, R., Jia, C., Zhang, G., Zhou, D., Dai, A. M., Yang, D., & Vosoughi, S. (2023) · arXiv · doi:10.48550/arxiv.2305.16960 · Cited by 18
“Second, rather than using a fixed margin as commonly found in hinge loss (Rennie & Srebro, 2005) or triplet loss (Schroff et al., 2015), we introduce a dynamic modulation of the margin for each mini-batch based on the differences in ratings.”
BackgroundMethodRobust Cross-View Gait Recognition with Evidence: A Discriminant Gait GAN (DiGGAN) Approach
Hu, B., Guan, Y., Gao, Y., Long, Y., Lane, N., & Ploetz, T. (2018) · arXiv · doi:10.48550/arxiv.1811.10493 · Cited by 16
“In the past few decades, many kinds of biometric traits such as voice [7], face [30], fingerprint [24], iris [35] and gait [8], have been utilised for recognition and authentication.”
BackgroundI Find Your Lack of Uncertainty in Computer Vision Disturbing
Valdenegro-Toro, M. (2021) · 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw53098.2021.00139 · Cited by 15
“Feature Learning Facial Feature Learning [45] FaceNet None E F G Embedding distances could be an (uncalibrated) uncertainty measure.”
BackgroundFace Recognition in Children: A Longitudinal Study
Bahmani, K., & Schuckers, S. (2022) · 2022 International Workshop on Biometrics and Forensics (IWBF) · doi:10.1109/iwbf55382.2022.9794551 · Cited by 15
VIMES: A Wearable Memory Assistance System for Automatic Information Retrieval
Bermejo, C., Braud, T., Yang, J., Mirjafari, S., Shi, B., Xiao, Y., & Hui, P. (2020) · Proceedings of the 28th ACM International Conference on Multimedia · doi:10.1145/3394171.3413663 · Cited by 15
“The face processing pipeline is as follows: (i) face detection using feature points [49].”
MethodJiao, J., Sun, H., Huang, Y., Xia, M., Qiao, M., Ren, Y., Wang, Y., & Guo, Y. (2023) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2023.3278259 · Cited by 15
“Equation (6) is modified from the triplet loss [27].”
BackgroundFine-tuning protein language models unlocks the potential of underrepresented viral proteomes
Sawhney, R., Ferrell, B. D., Dejean, T., Schreiber, Z., Harrigan, W., Polson, S. W., Wommack, K. E., & Belcaid, M. (2025) · PeerJ · doi:10.7717/peerj.19919 · Cited by 14
“The Siamese network was trained on triplets of protein sequences: an anchor, a positive (a sequence from the same orthologous group), and a negative (a sequence from a different orthologous group) according to the method described in Schroff, Kalenichenko & Philbin (2015) .”
GANprintR: Improved Fakes and Evaluation of the State of the Art in Face Manipulation Detection
Neves, J. C., Tolosana, R., Vera-Rodriguez, R., Lopes, V., Proenca, H., & Fierrez, J. (2020) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2020.3007250 · Cited by 12
“Their proposed approach, named FakeSpoter, extracted as features neuron coverage behaviors of real and fake faces from deep face recognition systems (i.e., VGG-Face [20], OpenFace [21], and FaceNet [22]), and then trained a SVM for the final classification.”
MethodAn Overview of Deepfake: The Sword of Damocles in AI
Tong, X., Wang, L., Pan, X., & Wang, J. G. (2020) · 2020 International Conference on Computer Vision, Image and Deep Learning (CVIDL) · doi:10.1109/cvidl51233.2020.00-88 · Cited by 12
DEEPFAKER: A Unified Evaluation Platform for Facial Deepfake and Detection Models
Wang, L., Meng, X., Li, D., Zhang, X., Ji, S., & Guo, S. (2024) · ACM Transactions on Privacy and Security · doi:10.1145/3634914 · Cited by 11
“To further demonstrate the functionality of DEEPFAKER as a comprehensive and practical evaluation platform, we present three case studies in this section.”
MethodLearning Interpretable Style Embeddings via Prompting LLMs
Patel, A., Rao, D., Kothary, A., McKeown, K., & Callison-Burch, C. (2023) · arXiv · doi:10.48550/arxiv.2305.12696 · Cited by 10
“We attach these on top of the L ISA model and train just the layer us-ing a contrastive learning objective and triplet loss (Khosla et al., 2020; Schroff et al., 2015).”
MethodMultimodal Fusion Refiner Networks
Sankaran, S., Yang, D., & Lim, S.-N. (2021) · arXiv · doi:10.48550/arxiv.2104.03435 · Cited by 10
“Con-trastive loss [16, 18] and triplet losses [43, 11] are being widely used in several applications.”
Person Re-ID by Fusion of Video Silhouettes and Wearable Signals for Home Monitoring Applications
Masullo, A., Burghardt, T., Damen, D., Perrett, T., & Mirmehdi, M. (2020) · Sensors · doi:10.3390/s20092576 · Cited by 10
“In the original paper on the triplet loss [44], hard mining of triplets was considered as a crucial step to deal with this problem.”
BackgroundDeepfake Detection for Facial Images with Facemasks
Ko, D., Lee, S., Park, J., Shin, S., Hong, D., & Woo, S. S. (2022) · arXiv · doi:10.48550/arxiv.2202.11359 · Cited by 10
“Previous deepfake detectionmethods mainly utilized facial recognition systems with deep neural network-based models [29, 31] such as FaceNet [28] and VGGFace2 [3].”
MethodGrondin, F., Lomanowska, A. M., Poiré, V., & Jackson, P. L. (2022) · Journal of Clinical Medicine · doi:10.3390/jcm11123461 · Cited by 9
“To extract the changing coordinates of the eyes and face across time, recordings of each session were first analyzed using a facial behavior analysis toolkit, OpenFace [21].”
MethodLatent Pattern Sensing: Deepfake Video Detection via Predictive Representation Learning
Ge, S., Lin, F., Li, C., Zhang, D., Tan, J., Wang, W., & Zeng, D. (2021) · ACM Multimedia Asia · doi:10.1145/3469877.3490586 · Cited by 9
“1, the spatiotemporal variances of feature maps (named latent patterns) extracted with a pretrained face descriptor FaceNet [31] from three continuous frames are very different between fake and real videos.”
BackgroundMethodInfluentialYao, W., Farooq, M. A., Lemley, J., & Corcoran, P. (2023) · IEEE Access · doi:10.1109/access.2023.3312612 · Cited by 8
The changing fortunes of pattern recognition and computer vision
Chellappa, R. (2016) · Image and Vision Computing · doi:10.1016/j.imavis.2016.04.005 · Cited by 8
Smaug: Fixing Failure Modes of Preference Optimisation with DPO-Positive
Pal, A., Karkhanis, D., Dooley, S., Roberts, M., Naidu, S., & White, C. (2024) · arXiv · doi:10.48550/arxiv.2402.13228 · Cited by 6
“We use the concept of anchor points or embeddings for both positive and negative points as in triplet loss Schroff et al. [2015].”
MethodLearning Environment-Aware Affordance for 3D Articulated Object Manipulation under Occlusions
Wu, R., Cheng, K., Shen, Y., Ning, C., Zhan, G., & Dong, H. (2023) · arXiv · doi:10.48550/arxiv.2309.07510 · Cited by 11
“The original task and its positive and negative paired tasks constitute triplets, so we use triplet loss [32] to learn the scene representations of them contrastively ( α is the boundary constant):”
Longitudinal Evaluation of Child Face Recognition and the Impact of Underlying Age
Singh, S., Bahmani, K., & Schuckers, S. (2024) · 2024 IEEE International Joint Conference on Biometrics (IJCB) · doi:10.1109/ijcb62174.2024.10744505 · Cited by 5
Prakash, A., Thejaswin, S., Nambiar, A., & Bernardino, A. (2023) · 2023 IEEE International Joint Conference on Biometrics (IJCB) · doi:10.1109/ijcb57857.2023.10448765 · Cited by 5
“…t and w k as the target vector and learnable vector respectively, the loss is computed as: An additional metric for evaluation is triplet loss [28], which is used to minimize the distance between the anchor and positive frames while maximizing the distance between the anchor and negative…”
MethodWISE: Rethinking the Knowledge Memory for Lifelong Model Editing of Large Language Models
Wang, P., Li, Z., Zhang, N., Xu, Z., Yao, Y., Jiang, Y., Xie, P., Huang, F., & Chen, H. (2024) · arXiv · doi:10.48550/arxiv.2405.14768 · Cited by 19
“To achieve the above objective, we design a margin-based loss function during editing training, similar to contrastive [41] or triplet loss [42].”
Child Face Age-Progression via Deep Feature Aging
Deb, D., Aggarwal, D., & Jain, A. K. (2020) · arXiv · doi:10.48550/arxiv.2003.08788 · Cited by 4
“20 FaceNet [37] 38.”
BackgroundMethodInfluentialA simple framework for contrastive learning of visual representations
Chen, T., Kornblith, S., Norouzi, M., & Hinton, G. (2020) · arXiv · doi:10.48550/arxiv.2002.05709 · Cited by 10,857
“Normalized cross entropy loss with adjustable temperature works better than alternatives We compare the NT-Xent loss against other commonly used contrastive loss functions, such as logistic loss (Mikolov et al., 2013), and margin loss (Schroff et al., 2015).”
Jaderberg, M., Simonyan, K., Zisserman, A., & Kavukcuoglu, K. (2015) · arXiv · doi:10.48550/arxiv.1506.02025 · Cited by 5,137
“Though not a recent invention, we now see a cornucopia of CNN-based models achieving state-of-the-art results in classification [19, 27, 34], localisation [30, 36], semantic segmentation [24], and action recognition [12, 31] tasks, amongst others.”
Background(2016) · neural information processing systems · Cited by 1,572
(2016) · Cited by 844
Unsupervised Scalable Representation Learning for Multivariate Time Series
Franceschi, J.-Y., Dieuleveut, A., & Jaggi, M. (2019) · arXiv · doi:10.48550/arxiv.1901.10738 · Cited by 305
“Triplet losses help to achieve the former (Schroff et al., 2015), but require to provide pairs of similar inputs, thus challenging the latter.”
Background
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.