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 701–800 of 1,094
Marginal loss and exclusion loss for partially supervised multi-organ segmentation
Shi, G., Xiao, L., Chen, Y., & Zhou, S. (2021) · Medical Image Analysis · doi:10.1016/j.media.2021.101979 · Cited by 157
Tang, W., Huang, S., Zhang, X., Zhou, F., Zhang, Y., & Liu, B. (2023) · 2023 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00377 · Cited by 116
“Hard sample mining is a popular technique to speed up convergence and enhance the discriminative power of the model in many deep learning areas, such as face recognition [25], object detection [29, 36], person re-identification [1, 28, 33, 34], and deep metric learning [30, 32].”
BackgroundCNN-based anti-spoofing two-tier multi-factor authentication system
Sajjad, M., Khan, S., Hussain, T., Muhammad, K., Sangaiah, A., Castiglione, A., Esposito, C., & Baik, S. (2019) · Pattern Recognition Letters · doi:10.1016/j.patrec.2018.02.015 · Cited by 119
Unequal-Training for Deep Face Recognition With Long-Tailed Noisy Data
Zhong, Y., Deng, W., Wang, M., Hu, J., Peng, J., Tao, X., & Huang, Y. (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.2019.00800 · Cited by 149
Liu, H., Cheng, J., Wang, W., Su, Y., & Bai, H. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.01.089 · Cited by 130
Label Independent Memory for Semi-Supervised Few-shot Video Classification
Zhu, L., & Yang, Y. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2020.3007511 · Cited by 113
“Kaiser et al. [5] used metric learning [44] to optimize the distances between the positive samples and the negative samples.”
BackgroundReinforced, Incremental and Cross-Lingual Event Detection From Social Messages
Peng, H. M., Zhang, R., Li, S., Cao, Y., Pan, S., & Yu, P. S. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3144993 · Cited by 148
“Inspired by [33], [34] and [35], [36], we introduce triplet losses in BasCL to balance a large number of negative samples of other event classes and a small number of positive samples of the same event class, and add global-local pair loss on this basis to preserve the graph structure information…”
MethodHCFNN: High-order coverage function neural network for image classification
Ning, X., Tian, W., Yu, Z., Li, W., Bai, X. P., & Wang, Y. (2022) · Pattern Recognition · doi:10.1016/j.patcog.2022.108873 · Cited by 133
Contrast-Phys: Unsupervised Video-Based Remote Physiological Measurement via Spatiotemporal Contrast
Sun, Z., & Li, X. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19775-8_29 · Cited by 176
A GAN-Based Model of Deepfake Detection in Social Media
Preeti, Kumar, M., & Sharma, H. K. (2023) · Procedia Computer Science · doi:10.1016/j.procs.2023.01.191 · Cited by 125
A Benchmark of Facial Recognition Pipelines and Co-Usability Performances of Modules
Serengil, S. I., & Özpınar, A. (2024) · Bilişim Teknolojileri Dergisi · doi:10.17671/gazibtd.1399077 · Cited by 243
“Presently, state-of-the-art models for the representation module predominantly leverage convolutional neural networks, exemplified by models like FaceNet [2], VGG-Face [3], ArcFace [4], Dlib [5], Sface [6], OpenFace [7], DeepFace [1], and DeepId [8].”
BackgroundMethodSegSort: Segmentation by Discriminative Sorting of Segments
Hwang, J.-J., Yu, S., Shi, J., Collins, M., Yang, T.-J., Zhang, X., & Chen, L.-C. (2019) · 2019 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00743 · Cited by 124
“An important technical point on metric learning is normalization [68, 60] so that features lie on a hypersphere, which is why the vMF distribution is of particular interest.”
BackgroundA Comprehensive Performance Evaluation of Deformable Face Tracking “In-the-Wild”
Chrysos, G. G., Antonakos, E., Snape, P., Asthana, A., & Zafeiriou, S. P. (2017) · International Journal of Computer Vision · doi:10.1007/s11263-017-0999-5 · Cited by 142
Improving triplet-wise training of convolutional neural network for vehicle re-identification
Zhang, Y., Liu, D., & Zha, Z.-J. (2017) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2017.8019491 · Cited by 133
“face recognition [13] and person reid [14].”
BackgroundDeep Domain Generalization With Structured Low-Rank Constraint
Ding, Z., & Fu, Y. (2018) · IEEE Transactions on Image Processing · doi:10.1109/tip.2017.2758199 · Cited by 124
“face recognition [32], image super-resolution [25], action classification [33], [34], feature selection [35], image clustering [36] and multi-view learning [37].”
BackgroundRegularFace: Deep Face Recognition via Exclusive Regularization
Zhao, K., Xu, J., & Cheng, M. (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.2019.00123 · Cited by 147
“Face recognition is one of the most widely studied topics in computer vision, and recently significant improvement has been made with the convolutional neural networks (CNNs) becoming the workhorse.”
BackgroundMethodInfluentialChristlein, V., Spranger, L., Seuret, M., Nicolaou, A., Král, P., & Maier, A. (2019) · IAPR International Conference on Document Analysis and Recognition (ICDAR) · doi:10.1109/icdar.2019.00177 · Cited by 167
“The networks are trained by means of the triplet loss [32]:”
MethodFace Recognition: Too Bias, or Not Too Bias?
Robinson, J. P., Livitz, G., Henon, Y., Qin, C., Fu, Y., & Timoner, S. J. (2020) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw50498.2020.00008 · Cited by 180
Contrastive Adaptation Network for Single- and Multi-Source Domain Adaptation
Kang, G., Jiang, L., Wei, Y., Yang, Y., & Hauptmann, A. G. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2020.3029948 · Cited by 145
“This paper is also related to the work that explicitly models the intra-class compactness and the inter-class separability, e.g. the contrastive loss [34], and the triplet loss [35].”
BackgroundGraph-based High-Order Relation Discovery for Fine-grained Recognition
Zhao, Y., Yan, K., Huang, F., & Li, J. (2021) · 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/cvpr46437.2021.01483 · Cited by 151
A Scalable and Privacy-Aware IoT Service for Live Video Analytics
Wang, J., Amos, B., Das, A., Pillai, P. S., Sadeh, N., & Satyanarayanan, M. (2017) · ACM Multimedia Systems Conference (MMSys) · doi:10.1145/3083187.3083192 · Cited by 123
Detailed 2D-3D Joint Representation for Human-Object Interaction
Li, Y., Liu, X., Han, L., Wang, S., Liu, J., Li, J., & Lu, C. (2020) · 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/cvpr42600.2020.01018 · Cited by 181
E-Key: An EEG-Based Biometric Authentication and Driving Fatigue Detection System
Xu, T., Wang, H., Lu, G., Wan, F., Deng, M., Qi, P., Bezerianos, A., Guan, C., & Sun, Y. (2021) · IEEE Transactions on Affective Computing · doi:10.1109/taffc.2021.3133443 · Cited by 132
“Convolutional neural network (CNN) is a useful tool that has been widely used in pattern recognition such as image recognition [36], classification of handwriting [37], natural language processing [38] and face recognition [39].”
BackgroundAutomatic assessment of pain based on deep learning methods: A systematic review
Gkikas, S., & Tsiknakis, M. (2023) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2023.107365 · Cited by 165
The Future of False Information Detection on Social Media: New Perspectives and Trends
Guo, B., Ding, Y., Yao, L., Liang, Y., & Yu, Z. (2021) · ACM Computing Surveys · doi:10.1145/3393880 · Cited by 122
EmoReact: a multimodal approach and dataset for recognizing emotional responses in children
Nojavanasghari, B., Baltrušaitis, T., Hughes, C. E., & Morency, L.-P. (2016) · Proceedings of the 18th ACM International Conference on Multimodal Interaction · doi:10.1145/2993148.2993168 · Cited by 119
“We used the approach proposed by Florian et al [47], followed by a manual inspection and correction.”
MethodLearning Metrics From Teachers: Compact Networks for Image Embedding
Yu, L., Yazici, V. O., Liu, X., van de Weijer, J., Cheng, Y., & Ramisa, A. (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.2019.00302 · Cited by 149
“Unfortunately, this led to severe overfitting in many cases, and semi-hard negative mining was introduced as a solution [25].”
BackgroundRevisiting IM2GPS in the Deep Learning Era
Vo, N., Jacobs, N., & Hays, J. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.286 · Cited by 122
“Image retrieval using learned, deep representations is useful to a wide range of tasks such as product ranking [4, 12], sketch based image retrieval [25], face recognition [31, 26], cross-view localization [18, 37, 34] and scene retrieval [33, 2, 23, 8].”
BackgroundFLASH: Fast and Robust Framework for Privacy-preserving Machine Learning
Byali, M., Chaudhari, H., Patra, A., & Suresh, A. (2020) · Proceedings on Privacy Enhancing Technologies · doi:10.2478/popets-2020-0036 · Cited by 120
“The influence of ML has found its way in a broad range of areas such as facial recognition [52], banking, medicine [25], recommendation systems and so on.”
BackgroundTowards Face Encryption by Generating Adversarial Identity Masks
Yang, X. H., Dong, Y., Pang, T., Su, H., Zhu, J., Chen, Y., & Xue, H. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00387 · Cited by 184
Talk-to-Edit: Fine-Grained Facial Editing via Dialog
Jiang, Y., Huang, Z., Pan, X., Loy, C. C., & Liu, Z. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01354 · Cited by 182
Hajamohideen, F., Shaffi, N., Mahmud, M., Karthikeyan, S., Sariri, A. A., Viswan, V., Abdesselam, A., & Initiative, A. D. N. (2023) · Brain Informatics · doi:10.1186/s40708-023-00184-w · Cited by 140
“The SCNN is proven to be an effective model for classifying data points whenever the samples in the dataset are limited or imbalanced [9, 10].”
BackgroundMethodDeep Learning in Mobile and Wireless Networking: A Survey
Zhang, C., Patras, P., & Haddadi, H. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1803.04311 · Cited by 122
Unsupervised Learning of Semantic Audio Representations
Jansen, A., Plakal, M., Pandya, R., Ellis, D. P. W., Hershey, S., Liu, J., Moore, R. C., & Saurous, R. A. (2018) · 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2018.8461684 · Cited by 117
Detecting and Mitigating Adversarial Perturbations for Robust Face Recognition
Goswami, G., Agarwal, A., Ratha, N., Singh, R., & Vatsa, M. (2019) · International Journal of Computer Vision · doi:10.1007/s11263-019-01160-w · Cited by 127
Detection of Face Morphing Attacks Based on PRNU Analysis
Scherhag, U., Debiasi, L., Rathgeb, C., Busch, C., & Uhl, A. (2019) · IEEE Transactions on Biometrics, Behavior, and Identity Science · doi:10.1109/tbiom.2019.2942395 · Cited by 121
“, VGG-Face [56] or FaceNet [57], to detect morphing attacks.”
BackgroundMethodInfluentialRevisiting metric learning for few-shot image classification
Li, X., Yu, L., Fu, C., Fang, M., & Heng, P. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.04.040 · Cited by 134
Deep Learning for Deepfakes Creation and Detection: A Survey
Nguyen, T. T., Nguyen, M. C., Nguyen, D. T., Nguyen, D. T., & Nahavandi, S. (2020) · Figshare · doi:10.6084/m9.figshare.12731039.v1 · Cited by 121
ATRW: A Benchmark for Amur Tiger Re-identification in the Wild
Li, S., Li, J., Tang, H., Qian, R., & Lin, W. (2020) · Proceedings of the 28th ACM International Conference on Multimedia · doi:10.1145/3394171.3413569 · Cited by 121
“Second , recent re-ID methods [14, 33] typically use pedestrians and cars as target objects, which usually have limited pose variations in a relatively constrained environment.”
BackgroundMethodRecent Advances in Deep Learning Techniques for Face Recognition
Fuad, M. T. H., Fime, A. A., Sikder, D., Iftee, M. A. R., Rabbi, J., Al-Rakhami, M. S., Gumaei, A., Sen, O., Fuad, M., & Islam, M. N. (2021) · IEEE Access · doi:10.1109/access.2021.3096136 · Cited by 152
“The authors [68] claimed that the FaceNet model achieved 95.12% accuracy on the YTF dataset and 99.63% accuracy on the LFW dataset.”
BackgroundMethodInfluentialGhostFaceNets: Lightweight Face Recognition Model From Cheap Operations
Alansari, M., Hay, O. A., Javed, S., Shoufan, A., Zweiri, Y., & Werghi, N. (2023) · IEEE Access · doi:10.1109/access.2023.3266068 · Cited by 178
“FaceNet [32], SphereFace [53], CosFace [54], Prodpoly [68],”
BackgroundMethodResultDAiSEE: Towards User Engagement Recognition in the WildPreprint
Gupta, A., D'Cunha, A., Awasthi, K., & Balasubramanian, V. N. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1609.01885 · Cited by 333
Simultaneous Feature and Dictionary Learning for Image Set Based Face Recognition
Lu, J., Wang, G., & Zhou, J. (2017) · IEEE Transactions on Image Processing · doi:10.1109/tip.2017.2713940 · Cited by 131
Deep Spectral Clustering Learning.
Law, M. T., Urtasun, R., & Zemel, R. S. (2017) · International Conference on Machine Learning · Cited by 120
Soft-Margin Softmax for Deep Classification
Liang, X., Wang, X., Lei, Z., Liao, S., & Li, S. Z. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-70096-0_43 · Cited by 146
XGAN: Unsupervised Image-to-Image Translation for Many-to-Many Mappings
Royer, A., Bousmalis, K., Gouws, S., Bertsch, F., Mosseri, I., Cole, F. H., & Murphy, K. P. (2020) · Springer eBooks · doi:10.1007/978-3-030-30671-7_3 · Cited by 156
Identity-aware CycleGAN for face photo-sketch synthesis and recognition
Fang, Y., Deng, W., Du, J., & Hu, J. (2020) · Pattern Recognition · doi:10.1016/j.patcog.2020.107249 · Cited by 151
“Therefore, we use a triplet loss [30] to regularize the inter-class and intra-class distances, which aims to make the anchor image close to the positive sample and far from the negative samples.”
MethodFederated Meta-Learning for Fraudulent Credit Card Detection
Zheng, W., Yan, L., Gou, C., & Wang, F.-Y. (2020) · Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2020/642 · Cited by 115
Understanding and Mitigating Annotation Bias in Facial Expression Recognition
Chen, Y., & Joo, J. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01471 · Cited by 157
“In particular, we use the triplet loss function [73] as a regularizer to penalize unfairness.”
MethodDeep learning-based detection of aluminum casting defects and their types
Parlak, İ. E., & Emel, E. (2022) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2022.105636 · Cited by 141
Deep Learning in Visual Computing and Signal Processing
Xie, D., Zhang, L., & Li, B. (2017) · Applied Computational Intelligence and Soft Computing · doi:10.1155/2017/1320780 · Cited by 136
“Accuracy Human-level (cropped) [74] 0.9753 FaceNet [67] 0.9964 ± 0.009 DeepFace-ensemble [66] 0.9735 ± 0.0025 OpenFace [70] 0.9292 ± 0.0134 final predictions.”
BackgroundInfluentialIntegrating Deep and Shallow Models for Multi-Modal Depression Analysis—Hybrid Architectures
Yang, L., Jiang, D., & Sahli, H. (2021) · IEEE Transactions on Affective Computing · doi:10.1109/taffc.2018.2870398 · Cited by 118
Learning deep representation of imbalanced SCADA data for fault detection of wind turbines
Chen, L., Xu, G., Zhang, Q., & Zhang, X. (2019) · Measurement · doi:10.1016/j.measurement.2019.03.029 · Cited by 121
TextRS: Deep Bidirectional Triplet Network for Matching Text to Remote Sensing Images
Abdullah, T., Bazi, Y., Rahhal, M. M. A., Mekhalfi, M. L., Rangarajan, L., & Zuair, M. (2020) · Remote Sensing · doi:10.3390/rs12030405 · Cited by 134
“• Describe what exists instead of what does not exist in the scene.”
3D Self-Supervised Methods for Medical Imaging
Taleb, A., Loetzsch, W., Danz, N., Severin, J., Gaertner, T., Bergner, B., & Lippert, C. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.03829 · Cited by 119
“) is a pairwise distance function, for which we use the L 2 distance, following [52]. α is a margin (gap) that is enforced between positive and negative pairs, which we set to 1 .”
MethodAre Convolutional Neural Networks or Transformers more like human vision?
Tuli, S., Dasgupta, I., Grant, E., & Griffiths, T. L. (2021) · arXiv · doi:10.48550/arxiv.2105.07197 · Cited by 119
“…Networks (CNNs) are currently the de-facto standard for many computer vision tasks, including object detection (Ren et al., 2015), image classification (Krizhevsky et al., 2012), segmentation (Girshick et al., 2014), facial recognition (Schroff et al., 2015), and captioning (L. Chen et al., 2017).”
Background3D Convolutional Neural Networks for Cross Audio-Visual Matching Recognition
Torfi, A., Iranmanesh, S. M., Nasrabadi, N. M., & Dawson, J. M. (2017) · IEEE Access · doi:10.1109/access.2017.2761539 · Cited by 134
DeformNet: Free-Form Deformation Network for 3D Shape Reconstruction from a Single Image
Kurenkov, A., Ji, J., Garg, A., Mehta, V., Gwak, J., Choy, C., & Savarese, S. (2018) · 2018 IEEE Winter Conference on Applications of Computer Vision (WACV) · doi:10.1109/wacv.2018.00099 · Cited by 118
A Compact Embedding for Facial Expression Similarity
Vemulapalli, R., & Agarwala, A. (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.2019.00583 · Cited by 138
“We use the NN2 version of pre-trained FaceNet [46] up to the inception (4e) block 3 whose output is a 7 × 7 feature map with 1024 channels.”
BackgroundMethodInfluentialDivide and Conquer the Embedding Space for Metric Learning
Sanakoyeu, A., Tschernezki, V., Büchler, U., & Ommer, B. (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.2019.00056 · Cited by 134
“We demonstrate a consistent performance boost when applying our approach to the widely-used triplet loss [38] and more complex state-of-the-art metric learning losses such as Proxy-NCA [30] and Margin loss [49].”
BackgroundMethodResultInfluentialTV-GAN: Generative Adversarial Network Based Thermal to Visible Face Recognition
Zhang, T., Wiliem, A., Siqi, Y., & Lovell, B. C. (2018) · International Conference on Biometrics · doi:10.1109/icb2018.2018.00035 · Cited by 124
Kim, D., Hernandez, M., Choi, J., & Medioni, G. (2017) · IEEE International Joint Conference on Biometrics (IJCB) · doi:10.1109/btas.2017.8272691 · Cited by 134
“2D face recognitionMost of the recent work on 2D face recognition [24, 29, 34] relies on deep learning approaches using massive datasets.”
BackgroundMethodMao, X., Wan, M., Wu, Y., & Lan, M. (2021) · International World Wide Web Conference (WWW) · doi:10.1145/3442381.3449897 · Cited by 181
Hyperbolic Vision Transformers: Combining Improvements in Metric Learning
Ermolov, A., Mirvakhabova, L., Khrulkov, V., Sebe, N., & Oseledets, I. V. (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/cvpr52688.2022.00726 · Cited by 230
Computational pathology: A survey review and the way forward
Hosseini, M. S., Bejnordi, B. E., Trinh, V. Q., Chan, L., Hasan, D., Li, X., Yang, S. C., Kim, T., Zhang, H., Wu, T., et al. (2024) · Journal of Pathology Informatics · doi:10.1016/j.jpi.2023.100357 · Cited by 176
Differentiable Rendering: A SurveyPreprint
Kato, H., Beker, D., Morariu, M. A., Ando, T., Matsuoka, T., Kehl, W., & Gaidon, A. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.12057 · Cited by 286
“They leverage the features of a facial recognition network [115] to regress the parameters of a statistical shape model of the face, using both real-world and synthetic face images.”
BackgroundHeterogeneous Face Recognition with CNNs
Saxena, S., & Verbeek, J. J. (2016) · Lecture notes in computer science · doi:10.1007/978-3-319-49409-8_40 · Cited by 117
Mehta, N., & Devarakonda, M. V. (2018) · Journal of Allergy and Clinical Immunology · doi:10.1016/j.jaci.2018.02.025 · Cited by 117
Cao, J., Li, Y., & Zhang, Z. M. (2018) · 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.2018.00451 · Cited by 129
“In fact, it is common for face verification methods [33, 4, 28, 36, 17] to adopt a local constraint loss such as Contrastive Loss [33] or Triple Loss [28] besides the global classification loss.”
MethodNetwork offloading policies for cloud robotics: a learning-based approach
Chinchali, S., Sharma, A., Harrison, J., Elhafsi, A., Kang, D., Pergament, E., Cidon, E., Katti, S., & Pavone, M. (2021) · Autonomous Robots · doi:10.1007/s10514-021-09987-4 · Cited by 117
“frobot, we use a combination of FaceNet [48], a widely-used pre-trained face detection model which embeds faces into embedding vectors, together with an SVM classifier over these embeddings.”
MethodToward improving ECG biometric identification using cascaded convolutional neural networks
Li, Y., Pang, Y., Wang, K., & Li, X. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.01.019 · Cited by 126
Generalizing Person Re-Identification by Camera-Aware Invariance Learning and Cross-Domain Mixup
Luo, C., Song, C., & Zhang, Z. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58555-6_14 · Cited by 132
“In term of the former direction, a series of deep metric learning methods [31,6,16,3,47] been proposed to enhance intra-class compactness and inter-class separability in the manifold.”
BackgroundMasked Face Recognition Challenge: The InsightFace Track Report
Deng, J., Guo, J., An, X., Zhu, Z., & Zafeiriou, S. P. (2021) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw54120.2021.00165 · Cited by 155
“Recently, great progress has been achieved in face recognition with large-scale training data [14, 24, 1, 39], sophisticated network structures [26, 16] and advanced loss designs [28, 29, 26, 25, 8, 20, 31, 30, 4, 3, 18, 6].”
BackgroundVAIN: Attentional Multi-agent Predictive ModelingPreprint
Hoshen, Y. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1706.06122 · Cited by 239
“Deep Neural Networks (DNNs) have had much success in machine perception e.g. Computer Vision [1, 2, 3], Natural Language Processing [4] and Speech Recognition [5, 6].”
Zhang, Y., Zhang, F., Yao, P., & Tang, J. (2018) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3219819.3219859 · Cited by 124
“Instead of projecting to a single point, triplet loss enables documents with the same identity to reside on a manifold [20], and at the same time maintain a distance from other documents.”
BackgroundHandcrafted and Deep Trackers: Recent Visual Object Tracking Approaches and Trends
Fiaz, M., Mahmood, A., Javed, S., & Jung, S. K. (2019) · ACM Computing Surveys · doi:10.1145/3309665 · Cited by 155
“Later, Siamese networks were used in various applications, including face recognition and verification [137], stereo matching [180], optical flow [40], large scale video classification [82], and patch matching [179].”
MethodCross-Modal Relationship Inference for Grounding Referring Expressions
Yang, S., Li, G., & Yu, Y. (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.2019.00427 · Cited by 146
Learning to Combine: Knowledge Aggregation for Multi-source Domain Adaptation
Wang, H., Xu, M., Ni, B., & Zhang, W. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58598-3_43 · Cited by 153
Pavement anomaly detection based on transformer and self-supervised learning
Lin, Z., Wang, H., & Li, S. (2022) · Automation in Construction · doi:10.1016/j.autcon.2022.104544 · Cited by 126
Visual Kinship Recognition of Families in the Wild
Robinson, J. P., Shao, M., Wu, Y., Liu, H., Gillis, T., & Fu, Y. (2018) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2018.2826549 · Cited by 125
“has been facial recognition [35], [36], [37].”
cGAN Based Facial Expression Recognition for Human-Robot Interaction
Deng, J. H., Pang, G., Zhang, Z., Pang, Z., Yang, H., & Yang, G. (2019) · IEEE Access · doi:10.1109/access.2019.2891668 · Cited by 125
“performance on a variety of tasks, such as object recognition [37], scene classification [38], and face recognition [39]–[41].”
BackgroundCentroid-based Deep Metric Learning for Speaker Recognition
Wang, J., Wang, K.-C., Law, M. T., Rudzicz, F., & Brudno, M. (2019) · ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2019.8683393 · Cited by 109
Face recognition in unconstrained environment with CNN
Fredj, H. B., Bouguezzi, S., & Souani, C. (2020) · The Visual Computer · doi:10.1007/s00371-020-01794-9 · Cited by 137
Complementary Relation Contrastive Distillation
Zhu, J., Tang, S., Chen, D., Yu, S., Liu, Y., Rong, M., Yang, A., & Wang, X. (2021) · 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/cvpr46437.2021.00914 · Cited by 146
“We compare our relation contrastive loss LRC with other commonly used contrastive loss, such as triplet loss with margin (LMT ) [36] and contrastive logistic loss (LCL) [26, 11].”
Masked Face Recognition Using Deep Learning: A Review
Alzu’bi, A., Albalas, F. A., Al-Hadhrami, T., Younis, L. B., & Bashayreh, A. (2021) · Electronics · doi:10.3390/electronics10212666 · Cited by 139
Yang, J., Xiao, S., Li, A., Lu, W., Gao, X., & Li, Y. (2022) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2021.3133859 · Cited by 115
“To improve the intra-class compactness and the separability between real and fake samples, we also use multi-scale triplet loss [42] for adjusting the sample distribution according to the triplets’ characteristic.”
MethodSelf Paced Deep Learning for Weakly Supervised Object Detection
Sangineto, E., Nabi, M., Culibrk, D., & Sebe, N. (2019) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2018.2804907 · Cited by 114
“, hard-negative mining) can be useful in a supervised scenario [18], [19].”
BackgroundWang, D., Otto, C., & Jain, A. K. (2016) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2582166 · Cited by 125
“The superior performance of deep learning based algorithms can be attributed to (a) large number of training images of large number of subjects (> 100 K) [22], (b) data augmentation methods, e.”
BackgroundResultInfluentialDeep Extreme Multi-label Learning
Zhang, W., Yan, J., Wang, X., & Zha, H. (2018) · International Conference on Multimedia Retrieval · doi:10.1145/3206025.3206030 · Cited by 136
Critical features for face recognition
Abudarham, N., Shkiller, L., & Yovel, G. (2018) · Cognition · doi:10.1016/j.cognition.2018.09.002 · Cited by 141
Face Recognition based Surveillance System Using FaceNet and MTCNN on Jetson TX2
Jose, E., Greeshma, M., Haridas, M. T. P., & Supriya, M. H. (2019) · International Conference on Advanced Computing and Communication Systems (ICACCS) · doi:10.1109/icaccs.2019.8728466 · Cited by 126
“The FaceNet algorithm transforms the face image into 128-dimensional Euclidean space similar to word embedding[9].”
MethodStochastic Attraction-Repulsion Embedding for Large Scale Image Localization
Liu, L., Li, H., & Dai, Y. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00266 · Cited by 137
““semi-hard” [33] negative with ‖fθ(q)− fθ(p)‖ < ‖fθ(q)− fθ(n)‖ (bottom-right-triangle in gradients figure) will still trigger force to pull q ∼ p pair while pushing q ∼ n pair, however, the force decays with increasing ‖fθ(q)− fθ(n)‖.”
Multi-Stage Feature Constraints Learning for Age Estimation
Xia, M., Zhang, X., Liu, W., Weng, L., & Xu, Y. (2020) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2020.2969552 · Cited by 116
“Aiming at the age common features of different individuals (or different states of the same individual) in each class, this paper carries out a center constraint on the features in the same age range (the idea of constraining feature has been well performed in face recognition tasks [29], [30]), and gradually narrows…”
BackgroundPhysical Adversarial Attack Meets Computer Vision: A Decade Survey
Wei, H., Tang, H., Jia, X., Wang, Z., Yu, H., Li, Z., Satoh, S., Gool, L. V., & Wang, Z. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3430860 · Cited by 207
“And they choose to attack the state-of-the-art FRS, i.e., FaceNet [161] and SphereFace [162].”
BackgroundSelf-Supervised Visual Terrain Classification From Unsupervised Acoustic Feature Learning
Zürn, J., Burgard, W., & Valada, A. (2020) · IEEE Transactions on Robotics · doi:10.1109/tro.2020.3031214 · Cited by 149
“To form such discriminative clusters of embeddings, triplet losses have been proposed [22].”
BackgroundOnline Student Authentication and Proctoring System Based on Multimodal Biometrics Technology
Labayen, M., Vea, R., Florez, J., Aginako, N., & Sierra, B. (2021) · IEEE Access · doi:10.1109/access.2021.3079375 · Cited by 112
“2) Aspect normalization, feature selection and classification: The extraction of biometric features through a deep neural network [41] has been improved training a database with 10M of images of 100K individuals with great variability of appearances and facial shapes, lighting, facial expressions, accessories and pose…”
MethodA Survey on Deep Transfer Learning and Beyond
Yu, F., Xiu, X., & Li, Y. (2022) · Mathematics · doi:10.3390/math10193619 · Cited by 139
Multi-View 3D Object Retrieval With Deep Embedding Network
Guo, H., Wang, J., Gao, Y., Li, J., & Lu, H. (2016) · IEEE Transactions on Image Processing · doi:10.1109/tip.2016.2609814 · Cited by 115
James, C. D., Aimone, J. B., Miner, N. E., Vineyard, C. M., Rothganger, F. H., Carlson, K. D., Mulder, S. A., Draelos, T. J., Faust, A., Marinella, M. J., et al. (2017) · Biologically Inspired Cognitive Architectures · doi:10.1016/j.bica.2016.11.002 · Cited by 153
Unsupervised Domain Adaptation for Face Recognition in Unlabeled Videos
Sohn, K., Liu, S., Zhong, G., Yu, X., Yang, M.-H., & Chandraker, M. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.630 · Cited by 110
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.