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
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
ArcFace: Additive Angular Margin Loss for Deep Face Recognition
Deng, J., Guo, J., Xue, N., & Zafeiriou, S. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00482 · Cited by 7,781
“Under large protocol, ArcFace surpasses FaceNet [27] by a clear margin and obtains comparable results on identification and better results on verification compared to CosFace [35].”
MethodResultInfluentialRecent advances in convolutional neural networks
Gu, J., Wang, Z., Kuen, J., Ma, L., Shahroudy, A., Shuai, B., Liu, T., Wang, X., Wang, G., Cai, J., et al. (2018) · Pattern Recognition · doi:10.1016/j.patcog.2017.10.013 · Cited by 6,576
Representation Learning with Contrastive Predictive Coding
Oord, A. V. D., Li, Y., & Vinyals, O. (2018) · arXiv · doi:10.48550/arxiv.1807.03748 · Cited by 6,242
“For example, the techniques proposed by [21, 22, 23] were based on triplet losses using a max-margin approach to separate positive from negative examples.”
MethodA Survey of Convolutional Neural Networks: Analysis, Applications, and Prospects
Li, Z., Liu, F., Yang, W., Peng, S., & Zhou, J. (2022) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2021.3084827 · Cited by 5,273
“FaceNet [71] uses Adagrad as the optimizer in training.”
MethodInfluentialParkhi, 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
Fully-Convolutional Siamese Networks for Object Tracking
Bertinetto, L., Valmadre, J., Henriques, J. F., Vedaldi, A., & Torr, P. H. S. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-48881-3_56 · Cited by 4,546
“Deep Siamese conv-nets have previously been applied to tasks such as face verification [14,18,20], keypoint descriptor learning [19,21] and one-shot character recognition [22].”
MethodA Discriminative Feature Learning Approach for Deep Face Recognition
Wen, Y., Zhang, K., Li, Z., & Qiao, Y. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46478-7_31 · Cited by 4,077
“Compared to [27] (200M), [34] (4M) and [25] (2M), it is a small scale training set.”
BackgroundInfluentialUnsupervised Feature Learning via Non-parametric Instance Discrimination
Wu, Z., Xiong, Y., Yu, S. X., & Lin, D. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00393 · Cited by 3,664
Supervised Contrastive Learning
Khosla, P., Teterwak, P., Wang, C., Sarna, A., Tian, Y., Isola, P., Maschinot, A., Liu, C., & Krishnan, D. (2020) · arXiv · doi:10.48550/arxiv.2004.11362 · Cited by 3,569
“This implicit property allows the contrastive loss to sidestep the need for explicit hard mining, which is a delicate but critical part of many losses, such as triplet loss [42].”
InfluentialDeep Convolutional Neural Networks for Image Classification: A Comprehensive Review
Rawat, W., & Wang, Z. (2017) · Neural Computation · doi:10.1162/neco_a_00990 · Cited by 3,503
“, 2006) and the triplet loss (Schroff et al., 2015), were suggested.”
BackgroundMethodInfluentialDeep learning for computer vision: A brief review
Voulodimos, A., Doulamis, N., Doulamis, A., & Protopapadakis, E. (2018) · Computational Intelligence and Neuroscience · doi:10.1155/2018/7068349 · Cited by 3,403
“Moreover, Google’s FaceNet [83] and Facebook’s DeepFace [84] are both based on CNNs. DeepFace [84] models a face in 3D and aligns it to appear as a frontal face.”
BackgroundSphereFace: Deep Hypersphere Embedding for Face Recognition
Liu, W., Wen, Y., Yu, Z., Li, M., Raj, B., & Song, L. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.713 · Cited by 2,280
“Most widely used loss functions for deep metric learning are contrastive loss [1, 3] and triplet loss [32, 22, 6], and both impose Euclidean margin to features.”
BackgroundMethodResultInfluentialCosFace: Large Margin Cosine Loss for Deep Face Recognition
Wang, H., Wang, Y., Zhou, Z., Ji, X., Gong, D., Zhou, J., Li, Z., & Liu, W. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00552 · Cited by 3,034
Survey on deep learning with class imbalance
Johnson, J. M., & Khoshgoftaar, T. M. (2019) · Journal of Big Data · doi:10.1186/s40537-019-0192-5 · Cited by 3,032
VGGFace2: A Dataset for Recognising Faces across Pose and Age
Cao, Q., Shen, L., Xie, W., Parkhi, O. M., & Zisserman, A. (2018) · 2018 13th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2018) · doi:10.1109/fg.2018.00020 · Cited by 3,032
Convolutional Two-Stream Network Fusion for Video Action Recognition
Feichtenhofer, C., Pinz, A., & Zisserman, A. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.213 · Cited by 2,684
“However, action recognition has not yet seen the substantial gains in performance that have been achieved in other areas by ConvNets, e.g. image classification [12, 23, 27], human face recognition [21], and human pose estimation [29].”
BackgroundNetVLAD: CNN Architecture for Weakly Supervised Place Recognition
Arandjelovic, R., Gronat, P., Torii, A., Pajdla, T., & Sivic, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.572 · Cited by 2,584
“Note that the above loss is related to the commonly used triplet loss [66, 67, 84, 85], but adapted to our weakly supervised scenario using a formulation (given by equation (5)) similar to multiple instance learning [21, 36, 83].”
MethodVoxCeleb2: Deep Speaker Recognition
Chung, J. S., Nagrani, A., & Zisserman, A. (2018) · Interspeech 2018 · doi:10.21437/interspeech.2018-1929 · Cited by 2,256
MS-Celeb-1M: A Dataset and Benchmark for Large-Scale Face Recognition
Guo, Y., Zhang, L., Hu, Y., He, X., & Gao, J. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46487-9_6 · Cited by 2,122
“The publicly available datasets are much smaller than that being used privately in industry, such as Facebook [2,3] and Google [4], as summarized in Table 1.”
BackgroundMethodDeep Facial Expression Recognition: A Survey
Li, S., & Deng, W. (2022) · IEEE Transactions on Affective Computing · doi:10.1109/taffc.2020.2981446 · Cited by 1,753
“Based on the triplet loss [131], which requires one positive example to be closer to the anchor than one negative example”
BackgroundDeepFashion: Powering Robust Clothes Recognition and Retrieval with Rich Annotations
Liu, Z., Luo, P., Qiu, S., Wang, X., & Tang, X. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.124 · Cited by 1,803
“Fourth, to learn the metric described by clothes pairs, we employ triplet loss introduced in [22], which enforces distance constraints among positive and negative samples”
MethodConditional Image Generation with PixelCNN Decoders
van den Oord, A., Kalchbrenner, N., Vinyals, O., Espeholt, L., Graves, A., & Kavukcuoglu, K. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1606.05328 · Cited by 1,833
“The network was trained with a triplet loss function [23] that ensured that the embedding h produced for an image x of a specific person was closer to the embeddings for all other images of the same person than it was to any embedding of another person.”
MethodA survey on deep learning and its applications
Dong, S., Wang, P., & Abbas, K. (2021) · Computer Science Review · doi:10.1016/j.cosrev.2021.100379 · Cited by 1,800
NetVLAD: CNN architecture for weakly supervised place recognitionPreprint
Arandjelović, R., Gronát, P., Torii, A., Pajdla, T., & Šivic, J. (2015) · arXiv (Cornell University) · Cited by 1,790
Relational Knowledge Distillation
Park, W., Kim, D., Lu, Y., & Cho, M. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00409 · Cited by 1,681
“Baseline (Triplet [31]) FitNet [27] Attention [47] DarkRank [7] Ours RKD-D RKD-A RKD-DA”
MethodDeep Learning in Mobile and Wireless Networking: A Survey
Zhang, C., Patras, P., & Haddadi, H. (2019) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2019.2904897 · Cited by 1,768
Deep Metric Learning via Lifted Structured Feature Embedding
Song, H. O., Xiang, Y., Jegelka, S., & Savarese, S. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.434 · Cited by 1,630
“However, the approach is not scalable with the size of the training data because it’s very costly to obtain ranking annotations in contrast to multiclass labels (i.e., product name) and because the approach is limited to ranking the data in triplet form.”
BackgroundMethodResultInfluentialThe iNaturalist Species Classification and Detection Dataset
Van Horn, G., Mac Aodha, O., Song, Y., Cui, Y., Sun, C., Shepard, A., Adam, H., Perona, P., & Belongie, S. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00914 · Cited by 1,451
Fine-Tuning CNN Image Retrieval with No Human Annotation
Radenovic, F., Tolias, G., & Chum, O. (2019) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2018.2846566 · Cited by 1,472
Deep face recognition: A survey
Wang, M., & Deng, W. (2021) · Neurocomputing · doi:10.1016/j.neucom.2020.10.081 · Cited by 1,466
Multimodal Language Analysis in the Wild: CMU-MOSEI Dataset and Interpretable Dynamic Fusion Graph
Bagher Zadeh, A., Liang, P. P., Poria, S., Cambria, E., & Morency, L.-P. (2018) · Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p18-1208 · Cited by 1,433
“Finally, we extract face embeddings from commonly used facial recognition models such as DeepFace (Taig-man et al., 2014), FaceNet (Schroff et al., 2015) and SphereFace (Liu et al., 2017).”
MethodAdditive Margin Softmax for Face Verification
Wang, F., Cheng, J., Liu, W., & Liu, H. (2018) · IEEE Signal Processing Letters · doi:10.1109/lsp.2018.2822810 · Cited by 1,106
“Metric learning loss functions such as contrastive loss [17] or triplet loss [16] usually require carefully designed sample mining strategies and the final performance is very sensitive to these strategies, so increasingly more researchers shift their attentions to building deep face verification models based on impro…”
BackgroundMethodPerson Re-identification by Multi-Channel Parts-Based CNN with Improved Triplet Loss Function
Cheng, D., Gong, Y., Zhou, S., Wang, J., & Zheng, N. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.149 · Cited by 1,247
“As illustrated in figure 2, similar to the works in [39, 34], the proposed person re-id method uses triplet examples to train the network.”
MethodLarge-Margin Softmax Loss for Convolutional Neural Networks
Liu, W., Wen, Y., Yu, Z., & Yang, M. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1612.02295 · Cited by 408
“Inspired by such idea, the contrastive loss (Hadsell et al., 2006) and triplet loss (Schroff et al., 2015) were proposed to enforce extra intra-class compactness and inter-class separa-† Figure 2.”
BackgroundMethodTrojaning Attack on Neural Networks
Liu, Y., Ma, S., Aafer, Y., Lee, W.-C., Zhai, J., Wang, W., & Zhang, X. (2018) · Proceedings 2018 Network and Distributed System Security Symposium · doi:10.14722/ndss.2018.23291 · Cited by 1,270
Beyond Triplet Loss: A Deep Quadruplet Network for Person Re-identification
Chen, W., Chen, X., Zhang, J., & Huang, K. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.145 · Cited by 1,224
“(a) the triplet loss with Euclidean distance [26]; (b) the triplet loss with learned”
BackgroundMethodInfluentialDeepfakes and beyond: A Survey of face manipulation and fake detection
Tolosana, R., Vera-Rodriguez, R., Fierrez, J., Morales, A., & Ortega-Garcia, J. (2020) · Information Fusion · doi:10.1016/j.inffus.2020.06.014 · Cited by 1,237
Exploring the Limits of Weakly Supervised Pretraining
Mahajan, D., Girshick, R., Ramanathan, V., He, K., Paluri, M., Li, Y., Bharambe, A., & van der Maaten, L. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-01216-8_12 · Cited by 1,224
Towards Real-Time Multi-Object Tracking
Wang, Z., Zheng, L., Liu, Y., Li, Y., & Wang, S. (2020) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-58621-8_7 · Cited by 1,195
A Light CNN for Deep Face Representation With Noisy Labels
Wu, X., He, R., Sun, Z., & Tan, T. (2018) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2018.2833032 · Cited by 1,186
“They train the CNN model on 2622 identities of 2.6M images collected from Internet and then fine-tune the model via triplet-based metric learning method like FaceNet.”
BackgroundMethodInfluentialGraph Contrastive Learning with Adaptive Augmentation
Zhu, Y., Xu, Y., Yu, F., Liu, Q., Wu, S., & Wang, L. (2021) · Proceedings of the Web Conference 2021 · doi:10.1145/3442381.3449802 · Cited by 1,066
“In contrast, the triplet loss is known to be computationally expensive [38].”
BackgroundAccessorize to a Crime: Real and Stealthy Attacks on State-of-the-Art Face Recognition
Sharif, M., Bhagavatula, S., Bauer, L., & Reiter, M. K. (2016) · Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/2976749.2978392 · Cited by 1,530
“, Google used 200M [38]), their DNN still achieves 98.”
MethodTADAM: Task dependent adaptive metric for improved few-shot learning
Oreshkin, B. N., Rodríguez, P., & Lacoste, A. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1805.10123 · Cited by 191
“In the same vein, in [5, 26, 30], triplet loss is used for learning a representation for few-shot learning.”
MethodTargeted Backdoor Attacks on Deep Learning Systems Using Data Poisoning
Chen, X., Liu, C., Li, B., Lu, K., & Song, D. (2017) · arXiv · doi:10.48550/arxiv.1712.05526 · Cited by 1,039
“YouTube Faces dataset contains 3,425 YouTube videos of 1,595 different people, and it has been a popular benchmark for face recognition and face verification tasks [58], [63], [55], [62].”
BackgroundCircle Loss: A Unified Perspective of Pair Similarity Optimization
Sun, Y., Cheng, C., Zhang, Y., Zhang, C., Zheng, L., Wang, Z., & Wei, Y. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00643 · Cited by 987
“, triplet loss [9, 22]) to optimize the similarity between samples.”
BackgroundInfluentialDisentangled Representation Learning GAN for Pose-Invariant Face Recognition
Tran, L., Yin, X., & Liu, X. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.141 · Cited by 1,005
Learning Deep Representation for Imbalanced Classification
Huang, C., Li, Y., Loy, C. C., & Tang, X. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.580 · Cited by 988
“Compared with the closely related triplet sampling method [34], our quintuplet sampling bett r preserves the embedding discrimination on imbalanced data.”
BackgroundMethodInfluentialContrastive Adaptation Network for Unsupervised Domain Adaptation
Kang, G., Jiang, L., Yang, Y., & Hauptmann, A. G. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00503 · Cited by 941
“the contrastive loss [12] and the triplet loss [33].”
BackgroundPerson Re-identification: Past, Present and Future
Zheng, L., Yang, Y. P., & Hauptmann, A. G. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1610.02984 · Cited by 922
“The second is the siamese model using image pairs [65] or triplets [66] as input.”
MethodLearning Deep Feature Representations with Domain Guided Dropout for Person Re-identification
Xiao, T., Li, H., Ouyang, W., & Wang, X. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.140 · Cited by 958
“For the training phase, there are several frameworks that use pairwise [1, 23] or triplet [33] inputs for learning feature embeddings.”
MethodWang, T., & Isola, P. (2020) · arXiv · doi:10.48550/arxiv.2005.10242 · Cited by 921
“Introduction A vast number of recent empirical works learn representations with a unit l2 norm constraint, effectively restricting the output space to the unit hypersphere (Parkhi et al., 2015; Schroff et al., 2015; Liu et al., 2017; Hasnat et al., 2017; Wang et al., 2017; Bojanowski & Joulin, 2017; Mettes et al., 201…”
BackgroundThe MegaFace Benchmark: 1 Million Faces for Recognition at Scale
Kemelmacher-Shlizerman, I., Seitz, S. M., Miller, D., & Brossard, E. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.527 · Cited by 912
“Face recognition has seen major breakthroughs in the last couple of years, with new results by multiple groups [25, 29, 27] surpassing human performance on the leading Labeled Faces in the Wild (LFW) benchmark [15] and achieving near perfect results.”
BackgroundInfluentialMulti-Similarity Loss With General Pair Weighting for Deep Metric Learning
Wang, X., Han, X., Huang, W., Dong, D., & Scott, M. R. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00516 · Cited by 854
Joint Unsupervised Learning of Deep Representations and Image Clusters
Yang, J., Parikh, D., & Batra, D. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.556 · Cited by 879
Sampling Matters in Deep Embedding Learning
Wu, C.-Y., Manmatha, R., Smola, A. J., & Krahenbuhl, P. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.309 · Cited by 836
Contrastive Representation Learning: A Framework and Review
Le-Khac, P. H., Healy, G., & Smeaton, A. F. (2020) · IEEE Access · doi:10.1109/access.2020.3031549 · Cited by 862
Generalized End-to-End Loss for Speaker Verification
Wan, L., Wang, Q., Papir, A., & Moreno, I. L. (2018) · 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2018.8462665 · Cited by 820
Frontal to profile face verification in the wild
Sengupta, S., Chen, J.-C., Castillo, C., Patel, V. M., Chellappa, R., & Jacobs, D. W. (2016) · 2016 IEEE Winter Conference on Applications of Computer Vision (WACV) · doi:10.1109/wacv.2016.7477558 · Cited by 805
“Many of the current state-of-the-art deep learning implementations [33], [32], [27], are not publicly available.”
BackgroundPose-Driven Deep Convolutional Model for Person Re-identification
Su, C., Li, J., Zhang, S., Xing, J., Gao, W., & Tian, Q. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.427 · Cited by 832
“In recent years, deep learning has demonstrated strong model capabilities and obtains very promising performances in many computer vision tasks [24, 14, 31, 38, 8].”
BackgroundDeeply-Learned Part-Aligned Representations for Person Re-identification
Zhao, L., Li, X., Zhuang, Y., & Wang, J. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.349 · Cited by 813
AttGAN: Facial attribute editing by only changing what you want
He, Z., Zuo, W., Kan, M., Shan, S., & Chen, X. (2019) · IEEE Transactions on Image Processing · doi:10.1109/tip.2019.2916751 · Cited by 833
“For conventional face recognition [1], [2] and facial attribute prediction [3], [4] tasks, significant advances have been made along with the development of deep convolutional neural networks (CNNs) and large scale labeled datasets.”
BackgroundDeep Learning Based Vulnerability Detection: Are We There Yet?
Chakraborty, S., Krishna, R., Ding, Y., & Ray, B. (2022) · IEEE Transactions on Software Engineering · doi:10.1109/tse.2021.3087402 · Cited by 790
“It has been observed that, over several iterations, the latent representation hðxgÞ of the input xg tend to increase in magnitude arbitrarily [45].”
BackgroundFully Convolutional Geometric Features
Choy, C., Park, J., & Koltun, V. (2019) · 2019 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00905 · Cited by 759
Low-Shot Visual Recognition by Shrinking and Hallucinating Features
Hariharan, B., & Girshick, R. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.328 · Cited by 792
“Amongst representation learning approaches, metric learning, such as the triplet loss [41, 38, 14] or siamese networks [22, 17], has been used to automatically learn feature representations where objects of the same class are closer together.”
MethodDeep Relative Distance Learning: Tell the Difference between Similar Vehicles
Liu, H., Tian, Y., Wang, Y., Pang, L., & Huang, T. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.238 · Cited by 782
MobileFaceNets: Efficient CNNs for Accurate Real-Time Face Verification on Mobile Devices
Chen, S., Liu, Y., Gao, X., & Han, Z. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-97909-0_46 · Cited by 771
Deep Image Retrieval: Learning Global Representations for Image Search
Gordo, A., Almazán, J., Revaud, J., & Larlus, D. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46466-4_15 · Cited by 787
“three stream Siamese networks) have been considered for metric learning [43,44] and face identification [45].”
BackgroundLearning Deep Structure-Preserving Image-Text Embeddings
Wang, L., Li, Y., & Lazebnik, S. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.541 · Cited by 760
“, [39]) can be considered as special cases of our framework where the two views come from the same modality and the two branches share weights.”
BackgroundAccurate 3D Face Reconstruction With Weakly-Supervised Learning: From Single Image to Image Set
Deng, Y., Yang, J., Xu, S., Chen, D., Jia, Y., & Tong, X. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2019.00038 · Cited by 763
“In this work, we train a FaceNet [44] structure using an in-house face recognition dataset with 3M face images of 50K identities crawled from the Internet, and use it as our deep feature extractor.”
MethodZhang, J., & Tao, D. (2021) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.3039359 · Cited by 774
“To address this issue, a representation learning method using triplet loss is proposed to directly learn discriminative and compact face embedding [99].”
MethodLow-Shot Learning from Imaginary Data
Wang, Y.-X., Girshick, R., Hebert, M., & Hariharan, B. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00760 · Cited by 744
Towards K-means-friendly Spaces: Simultaneous Deep Learning and Clustering
Yang, B., Fu, X., Sidiropoulos, N. D., & Hong, M. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1610.04794 · Cited by 565
“…DR approaches such as those used in spectral clustering (Ng et al., 2002; Von Luxburg, 2007) and DNN-based DR (Hinton & Salakhutdinov, 2006; Schroff et al., 2015; Her-shey et al., 2016) are also widely used as pre-processing before K-means or other clustering algorithms, see also (Vin-cent…”
MethodIdentifying facial phenotypes of genetic disorders using deep learning
Gurovich, Y., Hanani, Y., Bar, O., Nadav, G., Fleischer, N., Gelbman, D., Basel-Salmon, L., Krawitz, P. M., Kamphausen, S. B., Zenker, M., et al. (2019) · Nature Medicine · doi:10.1038/s41591-018-0279-0 · Cited by 739
Face detection using deep learning: An improved faster RCNN approach
Sun, X., Wu, P., & Hoi, S. C. H. (2018) · Neurocomputing · doi:10.1016/j.neucom.2018.03.030 · Cited by 716
A Comprehensive Survey of Loss Functions in Machine Learning
Wang, Q., Ma, Y., Zhao, K., & Tian, Y. (2022) · Annals of Data Science · doi:10.1007/s40745-020-00253-5 · Cited by 711
“Triplet loss function The input of triple loss [62] is a set of triples {( x , , where x a is a randomly selected face image, x p and x a belong to the same class (the same person), x n and x a belong to different classes (different people).”
BackgroundInfluentialVSE++: Improving Visual-Semantic Embeddings with Hard Negatives
Faghri, F., Fleet, D. J., Kiros, J., & Fidler, S. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1707.05612 · Cited by 543
Imbalance Problems in Object Detection: A Review
Oksuz, K., Cam, B. C., Kalkan, S., & Akbas, E. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2020.2981890 · Cited by 704
Deep Metric Learning: A Survey
KAYA, M., & BİLGE, H. Ş. (2019) · Symmetry · doi:10.3390/sym11091066 · Cited by 703
“Semi-hard negative mining used for the first time in [32] aims to find negative samples within the margin.”
BackgroundMethodInfluentialHuang, R., Zhang, S., Li, T., & He, R. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.267 · Cited by 687
“One category tries to adopt hand-crafted or learned pose-invariant features [4,28], while the other resorts to synthesis techniques to recover a frontal view image from a large pose face image and then use the recovered face images for face recognition [45,46].”
BackgroundTwo-Stream Neural Networks for Tampered Face Detection
Zhou, P., Han, X., Morariu, V. I., & Davis, L. S. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2017.229 · Cited by 700
“Instead of generating hard negatives in an online fashion [22], we randomly sample 15000 patch triplets from authentic images.”
BackgroundMethodIn Silico Labeling: Predicting Fluorescent Labels in Unlabeled Images
Christiansen, E. M., Yang, S. J., Ando, D. M., Javaherian, A., Skibinski, G., Lipnick, S., Mount, E., O’Neil, A., Shah, K., Lee, A. K., et al. (2018) · Cell · doi:10.1016/j.cell.2018.03.040 · Cited by 635
A Strong Baseline and Batch Normalization Neck for Deep Person Re-Identification
Luo, H., Jiang, W., Gu, Y., Liu, F., Liao, X., Lai, S., & Gu, J. (2020) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2019.2958756 · Cited by 593
“Inspired by FaceNet [15], Hermans et al. proposed an online hard example mining for triplet loss (TriHard loss).”
MethodPointNetVLAD: Deep Point Cloud Based Retrieval for Large-Scale Place Recognition
Uy, M. A., & Lee, G. H. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00470 · Cited by 670
Revisiting Self-Supervised Visual Representation Learning
Kolesnikov, A., Zhai, X., & Beyer, L. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00202 · Cited by 675
“[8] proposes to use the triplet loss [40, 18] in order to scale this pretext task to a large number of images (hence, classes) present in the ImageNet dataset.”
MethodLi, D., Opazo, C. R., Yu, X., & Li, H. (2020) · 2020 IEEE Winter Conference on Applications of Computer Vision (WACV) · doi:10.1109/wacv45572.2020.9093512 · Cited by 641
“To this end, we first employ the face detector and the face embedding provided by FaceNet [53] to encode faces of the dataset, and then compare the Euclidean distances among the face embeddings.”
MethodTriplet Loss in Siamese Network for Object Tracking
Dong, X., & Shen, J. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-01261-8_28 · Cited by 648
“Triplet loss in computer vision: Triplet loss has been widely applied for numerous applications in computer vision, such as face recognition [23], image retrieval [14], [37], [26], and person re-identification [3], [30], [13].”
MethodIARPA Janus Benchmark - C: Face Dataset and Protocol
Maze, B., Adams, J., Duncan, J. A., Kalka, N., Miller, T., Otto, C., Jain, A. K., Niggel, W. T., Anderson, J., Cheney, J., et al. (2018) · 2018 International Conference on Biometrics (ICB) · doi:10.1109/icb2018.2018.00033 · Cited by 639
No Fuss Distance Metric Learning Using Proxies
Movshovitz-Attias, Y., Toshev, A., Leung, T. K., Ioffe, S., & Singh, S. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.47 · Cited by 626
“This improved training stability but required large mini-batches - 1800 images in the case of [12], and training was still slow.”
BackgroundMethodInfluentialS^3FD: Single Shot Scale-Invariant Face Detector
Zhang, S., Zhu, X., Lei, Z., Shi, H., Wang, X., & Li, S. Z. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.30 · Cited by 653
Li, Y., Hu, P., Liu, Z., Peng, D., Zhou, J. T., & Peng, X. (2021) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v35i10.17037 · Cited by 714
Deep learning ensembles for melanoma recognition in dermoscopy images
Codella, N. C. F., Nguyen, Q.-B., Pankanti, S., Gutman, D. A., Helba, B., Halpern, A. C., & Smith, J. R. (2017) · IBM Journal of Research and Development · doi:10.1147/jrd.2017.2708299 · Cited by 644
“ructure presented here contains over 543,888,390 parameters, and yet was trained from a dataset of only 900 examples (as a reference, VGG (Visual Geometry Group) Face [49], DeepFace [50], and FaceNet [51] used 2.6 million, 4.4 million, and 200 million faces, respectively, for training face recognition networks). When…”
MethodInfluentialMagFace: A Universal Representation for Face Recognition and Quality Assessment
Meng, Q., Zhao, S., Huang, Z., & Zhou, F. (2021) · 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr46437.2021.01400 · Cited by 625
“Most of early works rely on metric-learning based loss, including contrastive loss [8], triplet loss [27], npair loss [30], angular loss [41], etc.”
BackgroundMethodCOMET: A Neural Framework for MT Evaluation
Rei, R., Stewart, C., Farinha, A. C., & Lavie, A. (2020) · Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2020.emnlp-main.213 · Cited by 617
Generate to Adapt: Aligning Domains Using Generative Adversarial Networks
Sankaranarayanan, S., Balaji, Y., Castillo, C. D., & Chellappa, R. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00887 · Cited by 619
Unsupervised Person Re-identification: Clustering and Fine-tuning
Fan, H., Zheng, L., Yan, C., & Yang, Y. (2018) · ACM Transactions on Multimedia Computing, Communications, and Applications · doi:10.1145/3243316 · Cited by 636
“3 can be replaced with the contrastive loss [26] or triplet loss [27].”
BackgroundMethod
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.