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 501–600 of 1,094
A Survey on an Emerging Area: Deep Learning for Smart City Data
Chen, Q., Wang, W., Wu, F., De, S., Wang, R., Zhang, B., & Huang, X. (2019) · IEEE Transactions on Emerging Topics in Computational Intelligence · doi:10.1109/tetci.2019.2907718 · Cited by 167
Attention-Based Models for Text-Dependent Speaker Verification
Rezaur rahman Chowdhury, F. A., Wang, Q., Moreno, I. L., & Wan, L. (2018) · 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2018.8461587 · Cited by 164
Learning a Fixed-Length Fingerprint Representation
Engelsma, J. J., Cao, K., & Jain, A. K. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2019.2961349 · Cited by 163
“face recognition systems which have almost entirely abandoned traditional handcrafted features in favor of deep features extracted by deep networks with remarkable success [11], [12], [13].”
BackgroundAutoencoder-based deep metric learning for network intrusion detection
Andresini, G., Appice, A., & Malerba, D. (2021) · Information Sciences · doi:10.1016/j.ins.2021.05.016 · Cited by 166
Cross-View Gait Recognition by Discriminative Feature Learning
Zhang, Y., Huang, Y., Yu, S., & Wang, L. (2020) · IEEE Transactions on Image Processing · doi:10.1109/tip.2019.2926208 · Cited by 166
Advances in Deep Learning-Based Medical Image Analysis
Liu, X., Gao, K., Liu, B., Pan, C., Liang, K., Yan, L., Ma, J., He, F., Zhang, S., Pan, S., et al. (2021) · Health Data Science · doi:10.34133/2021/8786793 · Cited by 165
“A number of researches on novel loss function design, such as focal loss [120], grading loss [121], contrastive loss [122], and triplet loss [123], have been proposed to tackle this problem.”
BackgroundExploring Disentangled Feature Representation Beyond Face Identification
Liu, Y., Wei, F., Shao, J., Sheng, L., Yan, J., & Wang, X. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00222 · Cited by 163
Exploit the Connectivity: Multi-Object Tracking with TrackletNet
Wang, G., Wang, Y., Zhang, H., Gu, R., & Hwang, J.-N. (2019) · Proceedings of the 27th ACM International Conference on Multimedia · doi:10.1145/3343031.3350853 · Cited by 165
“For each frame t , a vector consisting of the bounding box parameters, i.e., ( x t , y t , w t , h t ) , concatenated by an embedded appearance feature extracted from the FaceNet [30], is used to represent an individual detection from a tracklet.”
MethodSpectral Feature Transformation for Person Re-Identification
Luo, C., Chen, Y., Wang, N., & Zhang, Z.-X. (2019) · 2019 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00508 · Cited by 158
FastComposer: Tuning-Free Multi-subject Image Generation with Localized Attention
Xiao, G., Yin, T., Freeman, W. T., Durand, F., & Han, S. (2024) · International Journal of Computer Vision · doi:10.1007/s11263-024-02227-z · Cited by 430
“Identity preservation is determined by detecting faces in the reference and generated images using MTCNN [41], and then calculating a pairwise identity similarity using FaceNet [34].”
Classification is a Strong Baseline for Deep Metric LearningPreprint
Zhai, A., & Wu, H. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1811.12649 · Cited by 263
“Standard deep neural network metric learning approaches learn image representations through the local relationships between images in the form of pairs [3] [1] or triplets [8] [19].”
BackgroundMethodInfluentialDeep Metric Learning with BIER: Boosting Independent Embeddings Robustly
Opitz, M., Waltner, G., Possegger, H., & Bischof, H. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2018.2848925 · Cited by 159
Large-Scale Visual Speech Recognition
Shillingford, B., Assael, Y., Hoffman, M. W., Paine, T., Hughes, C., Prabhu, U., Liao, H., Sak, H., Rao, K., Bennett, L., et al. (2019) · Interspeech 2019 · doi:10.21437/interspeech.2019-1669 · Cited by 154
Preserving Semantic Relations for Zero-Shot Learning
Biswas, S., & Annadani, Y. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00793 · Cited by 163
Learning Meta Face Recognition in Unseen Domains
Guo, J., Zhu, X., Zhao, C., Cao, D., Lei, Z., & Li, S. Z. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00620 · Cited by 158
Universal Guidance for Diffusion Models
Bansal, A., Chu, H.-M., Schwarzschild, A., Sengupta, S., Goldblum, M., Geiping, J., & Goldstein, T. (2023) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw59228.2023.00091 · Cited by 537
Learning Invariant Deep Representation for NIR-VIS Face Recognition
He, R., Wu, X., Sun, Z., & Tan, T. N. (2017) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v31i1.10786 · Cited by 168
“Despite recent advances in the field of deep learning based VIS face recognition (Sun et al. 2014; Taigman et al. 2014; Parkhi, Vedaldi, and Zisserman 2015; Schroff, Kalenichenko, and Philbin 2015), implementing NIR-VIS face recognition efficiently presents serious challenges to current approaches.”
BackgroundChowdhury, A., & Ross, A. (2020) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2019.2941773 · Cited by 163
“5) Cosine Triplet Embedding Loss: The main aim of triplet based CNNs, as introduced in [46], is to learn an embedding f (x) ∈ d .”
BackgroundClustering Millions of Faces by Identity
Otto, C., Wang, D., & Jain, A. K. (2018) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2017.2679100 · Cited by 161
Latent Factor Guided Convolutional Neural Networks for Age-Invariant Face Recognition
Wen, Y., Li, Z., & Qiao, Y. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.529 · Cited by 158
Super-Identity Convolutional Neural Network for Face Hallucination
Zhang, K., Zhang, Z., Cheng, C.-W., Hsu, W. H., Qiao, Y., Liu, W., & Zhang, T. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-01252-6_12 · Cited by 169
Micro-attention for micro-expression recognition
Wang, C., Peng, M., Bi, T., & Chen, T. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.06.005 · Cited by 161
Vulnerability assessment and detection of Deepfake videos
Korshunov, P., & Marcel, S. (2019) · 2019 International Conference on Biometrics (ICB) · doi:10.1109/icb45273.2019.8987375 · Cited by 162
“In this section, we evaluate the vulnerability of face VGG [16] and Facenet6 [19] based recognition systems to videos with swapped faces and test several baseline detection systems.”
BackgroundMethodUnsupervised Adversarial Domain Adaptation for Cross-Domain Face Presentation Attack Detection
Wang, G., Han, H., Shan, S., & Chen, X. (2021) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2020.3002390 · Cited by 161
“1) Joint Loss: Instead of using the common cross-entropy loss like previous face PAD methods [21], [52], we propose a novel joint loss consisting of triplet loss [66] and center loss [67] for metric learning to handle the big withinclass diversity issue, particularly for the spoof face class, which consists of various…”
MethodAnomaly detection with convolutional neural networks for industrial surface inspection
Staar, B., Lütjen, M., & Freitag, M. (2019) · Procedia CIRP · doi:10.1016/j.procir.2019.02.123 · Cited by 157
AdvFaces: Adversarial Face Synthesis
Deb, D., Zhang, J., & Jain, A. K. (2020) · IEEE International Joint Conference on Biometrics (IJCB) · doi:10.1109/ijcb48548.2020.9304898 · Cited by 208
“Szegedy et al. first showed the dangers of adversarial examples in the image classification domain, where perturbing the pixels in the input image can cause 1Adversarial perturbations refer to altering an input image instance with small, human imperceptible changes in a manner that can evade CNN models.”
BackgroundMethodInfluentialZhang, J., Zhao, C., Ni, B., Xu, M., & Yang, X. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00177 · Cited by 192
“In addition, contrastive loss [17, 25] and triplet loss [40, 44] are used for strengthen learned metrics by fully exploring pair/triplet relationships within the dataset.”
MethodVisualPhishNet: Zero-Day Phishing Website Detection by Visual Similarity
Abdelnabi, S., Krombholz, K., & Fritz, M. (2020) · ACM SIGSAC Conference on Computer and Communications Security (CCS) · doi:10.1145/3372297.3417233 · Cited by 241
Driver Drowsiness Detection by Applying Deep Learning Techniques to Sequences of Images
Magán, E., Sesmero, M. P., Alonso-Weber, J. M., & Sanchis, A. (2022) · Applied Sciences · doi:10.3390/app12031145 · Cited by 160
“In [37], the authors use the FaceNet CNN [44] to extract facial features from the drivers’ images, and then use either a multiclass SVM or a K-NN to classify those features into one of the three categories of the UTA-RLDD dataset.”
MethodLong range iris recognition: A survey
Nguyen, K., Fookes, C., Jillela, R., Sridharan, S., & Ross, A. (2017) · Pattern Recognition · doi:10.1016/j.patcog.2017.05.021 · Cited by 158
Representation Learning by Rotating Your Faces
Tran, L., Yin, X., & Liu, X. (2019) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2018.2868350 · Cited by 158
A deep learning-based method for detecting non-certified work on construction sites
Qi, F., Li, H., Luo, X., Ding, L., Rose, T. M., An, W., & Yu, Y. (2018) · Advanced Engineering Informatics · doi:10.1016/j.aei.2018.01.001 · Cited by 183
Intelligent ICU for Autonomous Patient Monitoring Using Pervasive Sensing and Deep Learning
Davoudi, A., Malhotra, K. R., Shickel, B., Siegel, S., Williams, S., Ruppert, M., Bihorac, E., Ozrazgat-Baslanti, T., Tighe, P. J., Bihorac, A., et al. (2019) · Scientific Reports · doi:10.1038/s41598-019-44004-w · Cited by 158
Deep Learning for Micro-Expression Recognition: A Survey
Li, Y., Wei, J., Liu, Y. M., Kauttonen, J., & Zhao, G. (2022) · IEEE Transactions on Affective Computing · doi:10.1109/taffc.2022.3205170 · Cited by 218
“The metric learning techniques, such as contrastive loss [119] and triplet loss [120], was developed to ensure intra-class”
MethodDeep Metric Learning for Visual Understanding: An Overview of Recent Advances
Lu, J., Hu, J., & Zhou, J. (2017) · IEEE Signal Processing Magazine · doi:10.1109/msp.2017.2732900 · Cited by 179
“[24] proposed a FaceNet method by learning a projection to map facial images to a compact Euclidean space.”
BackgroundMethod3D CNN-Based Speech Emotion Recognition Using K-Means Clustering and Spectrograms
Hajarolasvadi, N., & Demirel, H. (2019) · Entropy · doi:10.3390/e21050479 · Cited by 172
Traceability Transformed: Generating More Accurate Links with Pre-Trained BERT Models
Lin, J.-F., Liu, Y., Zeng, Q., Jiang, M., & Cleland-Huang, J. (2021) · International Conference on Software Engineering · doi:10.1109/icse43902.2021.00040 · Cited by 182
Deep Learning for Free-Hand Sketch: A Survey
Xu, P., Hospedales, T. M., Yin, Q., Song, Y.-Z., Xiang, T., & Wang, L. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3148853 · Cited by 221
“Similarly, the challenges of these tasks are also mainly due to the large domain gap between free-hand sketches and other modalities.”
BackgroundMethodAd-Corre: Adaptive Correlation-Based Loss for Facial Expression Recognition in the Wild
Fard, A. P., & Mahoor, M. H. (2022) · IEEE Access · doi:10.1109/access.2022.3156598 · Cited by 157
“The most challenging issue of triplet loss [4] is its sensitivity to the process of choosing the anchor, positive and negative samples.”
BackgroundMethodInfluentialAnalyzing and Reducing the Damage of Dataset Bias to Face Recognition With Synthetic Data
Kortylewski, A., Egger, B., Schneider, A., Gerig, T., Morel-Forster, A., & Vetter, T. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2019.00279 · Cited by 154
Peng, X., Feng, J., Zhou, J. T., Lei, Y., & Yan, S. (2020) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2020.2968848 · Cited by 174
STEm-Seg: Spatio-Temporal Embeddings for Instance Segmentation in Videos
Athar, A., Mahadevan, S., Ošep, A., Leal-Taixé, L., & Leibe, B. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58621-8_10 · Cited by 196
(2020) · doi:10.1007/978-3-030-58586-0_27 · Cited by 156
Self-supervised learning for videos: A survey
Schiappa, M. C., Rawat, Y. S., & Shah, M. (2023) · ACM Computing Surveys · doi:10.1145/3577925 · Cited by 155
“The objective is formulated as a Soft Max Margin Ranking loss [152, 185] in which positive and negative pairs are from the same video/caption and different caption/video, respectively, which are then weighted by the similarity weights.”
MethodEdge Video Analytics for Public Safety: A Review
Zhang, Q., Sun, H., Wu, X., & Zhong, H. (2019) · Proceedings of the IEEE · doi:10.1109/jproc.2019.2925910 · Cited by 156
“[95] F.”
MethodEverything You Wanted to Know about Deep Learning for Computer Vision but Were Afraid to Ask
Ponti, M. A., Ribeiro, L. S. F., de Nazaré, T. S., Bui, T., & Collomosse, J. P. (2017) · SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) · doi:10.1109/sibgrapi-t.2017.12 · Cited by 187
“such as face recognition [79] and tracking [80].”
BackgroundSemanticAdv: Generating Adversarial Examples via Attribute-Conditioned Image Editing
Qiu, H., Xiao, C., Yang, L., Yan, X., Lee, H., & Li, B. H. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58568-6_2 · Cited by 214
“Deep neural networks (DNNs) have demonstrated great successes in various vision tasks [38,60,63,26,59,43,83,13,80].”
Discrimination-aware Network Pruning for Deep Model Compression
Liu, J., Zhuang, B., Zhuang, Z., Guo, Y. F., Huang, J., Zhu, J., & Tan, M. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2021.3066410 · Cited by 195
Hardness-Aware Deep Metric Learning
Zheng, W., Chen, Z., Lu, J., & Zhou, J. (2019) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2019.00016 · Cited by 189
“We demonstrate our framework on two losses with different tuple formations: triplet loss [25] and N-pair loss [28].”
BackgroundMethodInfluentialReconstruction-Based Disentanglement for Pose-Invariant Face Recognition
Peng, X., Yu, X., Sohn, K., Metaxas, D., & Chandraker, M. K. (2017) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.180 · Cited by 179
2D3D-Matchnet: Learning To Match Keypoints Across 2D Image And 3D Point Cloud
Feng, M., Hu, S., Ang, M. H., & Lee, G. H. (2019) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2019.8794415 · Cited by 213
“In contrast to the basic Triplet loss [26], [30], it also avoids the need of selecting an optimal margin.”
BackgroundLightweight Face Recognition Challenge
Deng, J., Guo, J., Zhang, D., Deng, Y., Lu, X., & Shi, S. X. (2019) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2019.00322 · Cited by 200
“Recently, great progress has been achieved in face recognition with deep learning-based methods [31, 32, 28, 26, 8, 21, 37, 35, 5] DCNNs map the face image, typically after InsightFace is a nonprofit Github project for 2D and 3D face analysis.”
BackgroundSpot the Conversation: Speaker Diarisation in the Wild
Chung, J. S., Huh, J., Nagrani, A., Afouras, T., & Zisserman, A. (2020) · Interspeech (USB) · doi:10.21437/interspeech.2020-2337 · Cited by 186
“Our method relies heavily on the recent successes of active speaker detection [12] and face and speaker verification [13, 14, 15].”
MethodRelational Reflection Entity Alignment
Mao, X., Wang, W., Xu, H., Wu, Y., & Lan, M. (2020) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3340531.3412001 · Cited by 203
Live Face De-Identification in Video
Gafni, O., Wolf, L., & Taigman, Y. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00947 · Cited by 193
Improved Embeddings with Easy Positive Triplet Mining
Xuan, H., Stylianou, A., & Pless, R. B. (2020) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv45572.2020.9093432 · Cited by 191
“In [15], the authors additionally propose the concept of Semi-Hard Negative mining, which chooses a anchor-negative pair that is farther than the anchor-positive pair, but within the margin, and so still contributes a positive loss.”
Masked Face Recognition for Secure AuthenticationPreprint
Anwar, A., & Raychowdhury, A. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2008.11104 · Cited by 277
“We select the state-of-the-art deep network-based face recognition system Facenet [13].”
MethodInfluentialSparsifying Neural Network Connections for Face Recognition
Sun, Y., Wang, X., & Tang, X. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.525 · Cited by 151
“The extremely large train data required makes it impossible to be reproduced by us, and we did not choose FaceNet as the baseline.”
MethodInfluentialFashion Style in 128 Floats: Joint Ranking and Classification Using Weak Data for Feature Extraction
Simo‐Serra, E., & Ishikawa, H. (2016) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2016.39 · Cited by 167
“In contrast with [25], we normalize the distances instead of normalizing the feature pairs to have unitary norm.”
BackgroundLearnable PINs: cross-modal embeddings for person identity
Nagrani, A., Albanie, S., & Zisserman, A. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-01261-8_5 · Cited by 151
CR-GAN: Learning Complete Representations for Multi-view Generation
Tian, Y., Peng, X., Zhao, L., Zhang, S., & Metaxas, D. N. (2018) · Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2018/131 · Cited by 149
“We evaluate the identity similarities between the real and generated images by feeding them to FaceNet.”
MethodInfluentialThe Elements of End-to-end Deep Face Recognition: A Survey of Recent Advances
Du, H., Shi, H., Zeng, D., Zhang, X.-P. S., & Mei, T. (2022) · ACM Computing Surveys · doi:10.1145/3507902 · Cited by 200
“DeepID2 [209], FaceNet [189], VGG Face [172], Lifted structured [203], N-pair loss [200], Multibatch[218],TPE[187],Smartmining[152],ContrastiveCNN[74] Hybrid Applyingclassificationandfeatureembed-dingtogetherasthesupervisorysignals.”
BackgroundMethodInfluentialClassical and modern face recognition approaches: a complete review
Ali, W., Tian, W., Din, S. U., Iradukunda, D., & Khan, A. A. (2020) · Multimedia Tools and Applications · doi:10.1007/s11042-020-09850-1 · Cited by 193
A state-of-the-art review on mobile robotics tasks using artificial intelligence and visual data
Cebollada, S., Payá, L., Flores, M., Peidró, A., & Reinoso, O. (2021) · Expert Systems with Applications · doi:10.1016/j.eswa.2020.114195 · Cited by 151
(2019) · doi:10.1109/wacv.2019.00178 · Cited by 151
Yan, K., Wang, X., Lu, L., Zhang, L., Harrison, A. P., Bagheri, M., & Summers, R. M. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00965 · Cited by 150
Protecting Facial Privacy: Generating Adversarial Identity Masks via Style-robust Makeup Transfer
Hu, S., Liu, X., Zhang, Y., Li, M., Zhang, L. Y., Jin, H., & Wu, L. (2022) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52688.2022.01459 · Cited by 221
Template Adaptation for Face Verification and Identification
Crosswhite, N., Byrne, J., Stauffer, C., Parkhi, O., Cao, Q., & Zisserman, A. (2017) · Image and Vision Computing · doi:10.1016/j.imavis.2018.09.002 · Cited by 173
Coloring With Limited Data: Few-Shot Colorization via Memory Augmented Networks
Yoo, S., Bahng, H., Chung, S., Lee, J., Chang, J., & Choo, J. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.01154 · Cited by 141
“Previously proposed triplet losses [26, 12] aim to make images of the same classes (positive neighbors) closer to each other while making images of different classes (negative neighbors) further away.”
BackgroundGround-to-Aerial Image Geo-Localization With a Hard Exemplar Reweighting Triplet Loss
Cai, S., Guo, Y., Khan, S. H., Hu, J., & Wen, G. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00848 · Cited by 190
“Inspired by face recognition [28, 31], Lin et al.”
BackgroundMethodInfluentialLezama, J., Qiu, Q., & Sapiro, G. (2017) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2017.720 · Cited by 160
“This is due in part to the excellent performance of deep neural networks [29, 33, 36, 40], which benefit from the availability of very large datasets of face photos, typically mined from the Internet [16, 40].”
BackgroundMethodHard-Aware Point-to-Set Deep Metric for Person Re-identification
Yu, R., Dou, Z., Bai, S., Zhang, Z., Xu, Y., & Bai, X. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-01270-0_12 · Cited by 147
Dual Cross-Entropy Loss for Small-Sample Fine-Grained Vehicle Classification
Li, X., Yu, L., Chang, D., Ma, Z., & Cao, J. (2019) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2019.2895651 · Cited by 155
“Triplet loss focuses on enforcing the margin between each pair of faces, from one person’s face to all other faces [39].”
BackgroundChen, B., Deng, W., & Du, J. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.428 · Cited by 146
“Many promising techniques have been developed, such as novel network structures [29, 13, 41], non-linear activation functions [12, 7, 6, 38], pooling strategies [11, 8, 53] and objective loss functions [43, 36], etc.”
BackgroundPreserving Semantic Relations for Zero-Shot Learning
Biswas, S., & Annadani, Y. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.48550/arxiv.1803.03049 · Cited by 246
“method is similar to the hard negative mining approach for triplet based learning algorithms [6, 33, 35].”
MethodA Recurrent Encoder-Decoder Network for Sequential Face Alignment
Peng, X., Feris, R. S., Wang, X., & Metaxas, D. N. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46448-0_3 · Cited by 146
Combining DC-GAN with ResNet for blood cell image classification
Ma, L., Shuai, R., Ran, X., Liu, W., & Ye, C. (2020) · Medical & Biological Engineering & Computing · doi:10.1007/s11517-020-02163-3 · Cited by 174
“Inspired by the triplet loss [38] and center loss functions [52], we consider a new loss function whose specific method is as follows.”
MethodDriver Fatigue Detection Based on Convolutional Neural Networks Using EM-CNN
Zhao, Z., Zhou, N., Zhang, L., Yan, H., Xu, Y., & Zhang, Z. (2020) · Computational Intelligence and Neuroscience · doi:10.1155/2020/7251280 · Cited by 185
Auxiliary Information-Guided Industrial Data Augmentation for Any-Shot Fault Learning and Diagnosis
Zhuo, Y., & Ge, Z. (2021) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2021.3053106 · Cited by 157
PhotoMaker: Customizing Realistic Human Photos via Stacked ID Embedding
Li, Z., Cao, M., Wang, X., Qi, Z., Cheng, M., & Shan, Y. (2024) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52733.2024.00825 · Cited by 464
“To evaluate the quality of the generation, we employ the FID metric [22, 47].”
SynFace: Face Recognition with Synthetic Data
Qiu, H., Yu, B., Gong, D., Li, Z., Liu, W., & Tao, D. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01070 · Cited by 219
PCL: Proxy-based Contrastive Learning for Domain Generalization
Yao, X., Bai, Y., Zhang, X., Zhang, Y., Sun, Q. L., Chen, R., Li, R., & Yu, B. (2022) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52688.2022.00696 · Cited by 232
“Pair-based methods, such as [17, 43, 48, 61], focus on sample-to-sample relations.”
MethodA Comprehensive Study on Face Recognition Biases Beyond Demographics
Terhörst, P., Kolf, J. N., Huber, M., Kirchbuchner, F., Damer, N., Morales, A., Fiérrez, J., & Kuijper, A. (2021) · IEEE Transactions on Technology and Society · doi:10.1109/tts.2021.3111823 · Cited by 193
Heterogeneous Dataflow Accelerators for Multi-DNN Workloads
Kwon, H., Lai, L., Pellauer, M., Krishna, T., Chen, Y.-H., & Chandra, V. (2021) · 2021 IEEE International Symposium on High-Performance Computer Architecture (HPCA) · doi:10.1109/hpca51647.2021.00016 · Cited by 144
“) results, many applications now employ DNNs as their backbone to perform tasks like face recognition [25], image segmentation [11], [26], depth estimation [17], and so on.”
BackgroundSynthesizing Normalized Faces from Facial Identity Features
Cole, F. H., Belanger, D., Krishnan, D., Sarna, A., Mosseri, I., & Freeman, W. T. (2017) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2017.361 · Cited by 166
Learning Cross-Modal Embeddings With Adversarial Networks for Cooking Recipes and Food Images
Wang, H., Sahoo, D., Liu, C., Lim, E., & Hoi, S. C. H. (2019) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2019.01184 · Cited by 171
AnalyticDB-V: a hybrid analytical engine towards query fusion for structured and unstructured data
Wei, C., Wu, B., Wang, S., Lou, R., Zhan, C., Li, F., & Cai, Y. (2020) · Proceedings of the VLDB Endowment · doi:10.14778/3415478.3415541 · Cited by 251
“Such vector retrievals are widespread in various domains, such as face recognition [47, 18], person/vehicle re-identification [56, 32], recommendation [49], and voiceprint recognition [42].”
BackgroundThe World is Binary: Contrastive Learning for Denoising Next Basket Recommendation
Qin, Y., Wang, P., & Li, C. (2021) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3404835.3462836 · Cited by 150
Xie, J., Xiang, J., Chen, J., Hou, X., Zhao, X., & Shen, L. (2022) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52688.2022.00106 · Cited by 227
“The idea of contrastive learning is to pull close the samples from the positive pair and to push apart the samples from the negative pair [4, 7, 10, 11, 29].”
Deep Learning for Iris Recognition: A Survey
Nguyen, K., Proença, H. P. M. C., & Alonso‐Fernandez, F. (2024) · ACM Computing Surveys · doi:10.1145/3651306 · Cited by 212
“Triplet loss: Since the pairwise network is trained with separate genuine and impostor pairs, it may not converge well, which has been proven in face recognition [145].”
BackgroundMethodYovel, G., & O’Toole, A. J. (2016) · Trends in Cognitive Sciences · doi:10.1016/j.tics.2016.02.005 · Cited by 158
Faceless Person Recognition: Privacy Implications in Social Media
Oh, S. J., Benenson, R., Fritz, M., & Schiele, B. (2016) · Lecture notes in computer science · doi:10.1007/978-3-319-46487-9_2 · Cited by 169
Learning Coarse-to-Fine Structured Feature Embedding for Vehicle Re-Identification
Guo, H., Zhao, C., Liu, Z., Wang, J., & Lu, H. (2018) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v32i1.12237 · Cited by 157
“Later schroff et al. (2015) adopted triplet loss to learn effective deep face representation.”
BackgroundPIRM Challenge on Perceptual Image Enhancement on Smartphones: Report
Ignatov, A., Timofte, R., Vu, T., Luu, T. M., Pham, T. X., Nguyen, C. V., Kim, Y., Choi, J.-S., Kim, M., Huang, J., et al. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-11021-5_20 · Cited by 169
A survey of semi- and weakly supervised semantic segmentation of images
Zhang, M., Zhou, Y. Z., Zhao, J., Man, Y., Liu, B., & Yao, R. (2019) · Artificial Intelligence Review · doi:10.1007/s10462-019-09792-7 · Cited by 171
“The tuning loss used to fine-tune the parameters is formulated by transforming from graph optimization to triplet ranking (Schroff et al. 2015), shown as Eq.”
MethodLi, M., Chang, K., Bearce, B., Chang, C. Y., Huang, A. J., Campbell, J. P., Brown, J. M., Singh, P., Hoebel, K., Erdogmus, D., et al. (2020) · npj Digital Medicine · doi:10.1038/s41746-020-0255-1 · Cited by 169
Whitening for Self-Supervised Representation LearningPreprint
Ermolov, A., Siarohin, A., Sangineto, E., & Sebe, N. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2007.06346 · Cited by 233
Learning to cluster faces on an affinity graph
Yang, L., Zhan, X., Chen, D., Yan, J., Loy, C. C., & Lin, D. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00240 · Cited by 142
“Thanks to the advances in deep learning techniques, the performance of face recognition has been remarkably boosted [25, 22, 27, 3].”
BackgroundPalmprint Recognition in Uncontrolled and Uncooperative Environment
Matkowski, W. M., Chai, T., & Kong, A. W. (2019) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2019.2945183 · Cited by 156
“Many face recognition methods are already close to human-level performance [2], [3].”
BackgroundMethodMasked face recognition with convolutional neural networks and local binary patterns
Vu, H. N., Nguyen, M. H., & Pham, C. (2021) · Applied Intelligence · doi:10.1007/s10489-021-02728-1 · Cited by 174
“Herein, we choose Insightface, which is an implement of ArcFace [27], for the face recognition task with an anticipate that the system could overcome the problem of large-scale identities by giving a 512 dimensional output vector (512d-vector) instead of 128d as originally proposed FaceNet [24] or Dlib [63].”
Method
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.