SOTAVerified

Face Recognition

Facial Recognition is the task of making a positive identification of a face in a photo or video image against a pre-existing database of faces. It begins with detection - distinguishing human faces from other objects in the image - and then works on identification of those detected faces.

The state of the art tables for this task are contained mainly in the consistent parts of the task : the face verification and face identification tasks.

( Image credit: Face Verification )

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 201–250 of 2329 papers

TitleStatusHype
LARNet: Lie Algebra Residual Network for Face RecognitionCode1
Learnable Multi-level Frequency Decomposition and Hierarchical Attention Mechanism for Generalized Face Presentation Attack DetectionCode1
Learning Meta Pattern for Face Anti-SpoofingCode1
Learning to Cluster Faces on an Affinity GraphCode1
Long-Term Face Tracking for Crowded Video-Surveillance ScenariosCode1
Loss Function Search for Face RecognitionCode1
Representation Learning with Statistical Independence to Mitigate BiasCode1
Masked Face Recognition for Secure AuthenticationCode1
Adversarial Sticker: A Stealthy Attack Method in the Physical WorldCode1
MEBeauty: a multi-ethnic facial beauty dataset in-the-wildCode1
MixFaceNets: Extremely Efficient Face Recognition NetworksCode1
MixFairFace: Towards Ultimate Fairness via MixFair Adapter in Face RecognitionCode1
MOS: A Low Latency and Lightweight Framework for Face Detection, Landmark Localization, and Head Pose EstimationCode1
MS-Celeb-1M: A Dataset and Benchmark for Large-Scale Face RecognitionCode1
Network Pruning using Adaptive Exemplar FiltersCode1
OCFR 2022: Competition on Occluded Face Recognition From Synthetically Generated Structure-Aware OcclusionsCode1
Adversarial Mask: Real-World Universal Adversarial Attack on Face Recognition ModelCode1
One-Class Face Anti-spoofing via Spoof Cue Map-Guided Feature LearningCode1
Open-Set Biometrics: Beyond Good Closed-Set ModelsCode1
Open-Set Face Identification on Few-Shot Gallery by Fine-TuningCode1
Biometric Quality: Review and Application to Face Recognition with FaceQnetCode1
A Unified Framework for Masked and Mask-Free Face Recognition via Feature RectificationCode1
Attention-based Partial Face RecognitionCode1
Perturbation Inactivation Based Adversarial Defense for Face RecognitionCode1
PocketNet: Extreme Lightweight Face Recognition Network using Neural Architecture Search and Multi-Step Knowledge DistillationCode1
Poison Frogs! Targeted Clean-Label Poisoning Attacks on Neural NetworksCode1
Age Estimation Using Expectation of Label Distribution LearningCode1
ArcFace: Additive Angular Margin Loss for Deep Face RecognitionCode1
AdvFaces: Adversarial Face SynthesisCode1
Bias in Automated Speaker RecognitionCode1
A realistic approach to generate masked faces applied on two novel masked face recognition data setsCode1
QuantFace: Towards Lightweight Face Recognition by Synthetic Data Low-bit QuantizationCode1
Racial Faces in-the-Wild: Reducing Racial Bias by Information Maximization Adaptation NetworkCode1
Rank-based verification for long-term face tracking in crowded scenesCode1
Adv-Makeup: A New Imperceptible and Transferable Attack on Face RecognitionCode1
RepMLP: Re-parameterizing Convolutions into Fully-connected Layers for Image RecognitionCode1
Rethinking Bias Mitigation: Fairer Architectures Make for Fairer Face RecognitionCode1
Re-thinking Model Inversion Attacks Against Deep Neural NetworksCode1
SemanticAdv: Generating Adversarial Examples via Attribute-conditional Image EditingCode1
Circle Loss: A Unified Perspective of Pair Similarity OptimizationCode1
Semi-Siamese Training for Shallow Face LearningCode1
SER-FIQ: Unsupervised Estimation of Face Image Quality Based on Stochastic Embedding RobustnessCode1
ShrinkTeaNet: Million-scale Lightweight Face Recognition via Shrinking Teacher-Student NetworksCode1
Simultaneously Optimizing Perturbations and Positions for Black-box Adversarial Patch AttacksCode1
Spherical Confidence Learning for Face RecognitionCode1
Stacked Dense U-Nets with Dual Transformers for Robust Face AlignmentCode1
SwinFace: A Multi-task Transformer for Face Recognition, Expression Recognition, Age Estimation and Attribute EstimationCode1
SynFace: Face Recognition with Synthetic DataCode1
Teaching Where to Look: Attention Similarity Knowledge Distillation for Low Resolution Face RecognitionCode1
CurricularFace: Adaptive Curriculum Learning Loss for Deep Face RecognitionCode1
Show:102550
← PrevPage 5 of 47Next →
All datasetsLFWCFP-FPCASIA-WebFace+masksCelebA+masksMLFWAgeDB-30Color FERETIJB-BCALFWmebeblurfMFW+ (M-M)MFW+ (U-M)

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1PIC - ArcFaceFNMR [%] @ 10-3 FMR4.38—Unverified
2GhostFaceNetV2-1 (MS1MV3)Accuracy1—Unverified
3SymFace + AdaFace + ResNet100 +WebFace (MS1MV2)Accuracy1—Unverified
4ProdpolyAccuracy1—Unverified
5DiscFaceAccuracy1—Unverified
6ArcFace + MS1MV2 + R100Accuracy1—Unverified
7DCQAccuracy1—Unverified
8AdaFace + WebFace4M + R100Accuracy1—Unverified
9EdgeFace - S (g=0.5)Accuracy1—Unverified
10EdgeFace - XS (g=0.6)Accuracy1—Unverified
#ModelMetricClaimedVerifiedStatus
1GhostFaceNetV2-1Accuracy0.99—Unverified
2SymFace + AdaFace + ResNet100 +WebFaceAccuracy0.99—Unverified
3ElasticFace-ArcAccuracy0.99—Unverified
4DiscFaceAccuracy0.99—Unverified
5CircleLoss(ours)Accuracy0.96—Unverified
6EdgeFace - S (g=0.5)Accuracy0.96—Unverified
7DCQAccuracy0.93—Unverified
#ModelMetricClaimedVerifiedStatus
1Fine-tuned ArcFaceAccuracy91.47—Unverified
2Fine-tuned FaceNetAccuracy88.06—Unverified
3ArcFaceAccuracy87.95—Unverified
4Fine-tuned VGG-FaceAccuracy86.85—Unverified
5FaceNetAccuracy84.21—Unverified
6VGG-FaceAccuracy79.65—Unverified
#ModelMetricClaimedVerifiedStatus
1Fine-tuned ArcFaceAccuracy95.43—Unverified
2Fine-tuned FaceNetAccuracy93.58—Unverified
3ArcFaceAccuracy91.78—Unverified
4Fine-tuned VGG-FaceAccuracy91.51—Unverified
5FaceNetAccuracy90.96—Unverified
6VGG-FaceAccuracy84.56—Unverified
#ModelMetricClaimedVerifiedStatus
1MS1MV2, R100, SFaceAccuracy91.57—Unverified
2MS1MV2, R100, ArcfaceAccuracy90.57—Unverified
3MS1MV2, R100, CurricularfaceAccuracy90.43—Unverified
4VGGFace2, R50, ArcFaceAccuracy85.95—Unverified
5CASIA-WebFace, R50, CosFaceAccuracy82.52—Unverified
6Private-Asia, R50, ArcFaceAccuracy77.2—Unverified
#ModelMetricClaimedVerifiedStatus
1ProdpolyAccuracy0.98—Unverified
2ElasticFace-CosAccuracy0.98—Unverified
3Transformer loss+ArcFace ResNet100Accuracy0.98—Unverified
4DCQAccuracy0.98—Unverified
#ModelMetricClaimedVerifiedStatus
1VGG based5-class test accuracy99.2—Unverified
2PIC - ArcFaceFNMR [%] @ 10-3 FMR4.22—Unverified
3PIC - MagFaceFNMR [%] @ 10-3 FMR3.92—Unverified
4PIC - QMagFaceFNMR [%] @ 10-3 FMR3.24—Unverified
#ModelMetricClaimedVerifiedStatus
1ElasticFace-CosTAR @ FAR=0.00010.95—Unverified
2ArcFace+CSFMRank-10.95—Unverified
3ArcFace + MS1MV2 + R100Rank-10.95—Unverified
4AdaFace + MS1MV3 + R100TAR @ FAR=0.00010.94—Unverified
#ModelMetricClaimedVerifiedStatus
1ProdpolyAccuracy0.96—Unverified
2ElasticFace-ArcAccuracy0.96—Unverified
3GhostFaceNetV2-1Accuracy0.96—Unverified
#ModelMetricClaimedVerifiedStatus
1PIC - ArcFaceFNMR [%] @ 10-3 FMR10.1—Unverified
2PIC - QMagFaceFNMR [%] @ 10-3 FMR9.78—Unverified
3PIC - MagFaceFNMR [%] @ 10-3 FMR9.75—Unverified
#ModelMetricClaimedVerifiedStatus
1ArcFace + PPLTAR@FAR=0.000178.8—Unverified
2MaskInv-HGTAR@FAR=0.000178.36—Unverified
3FocusFaceTAR@FAR=0.000177.77—Unverified
#ModelMetricClaimedVerifiedStatus
1AdaFace+PPLTAR@FAR=0.000184.27—Unverified
2MaskInv-HGTAR@FAR=0.000183.82—Unverified
3FocusFaceTAR@FAR=0.000183.4—Unverified
#ModelMetricClaimedVerifiedStatus
1PIC - QMagFaceFNMR [%] @ 10-3 FMR0.96—Unverified
2PIC - MagFaceFNMR [%] @ 10-3 FMR0.96—Unverified
3PIC - ArcFaceFNMR [%] @ 10-3 FMR0.05—Unverified
#ModelMetricClaimedVerifiedStatus
1FaceNet+Adaptive ThresholdAverage Accuracy (10 times)84.3—Unverified
2FaceNet+Fixed Threshold (0.2487)Average Accuracy (10 times)80.6—Unverified
#ModelMetricClaimedVerifiedStatus
1Model with Up Convolution + DoG Filter (Aligned)Rank-185—Unverified
2DPMRank-171—Unverified
#ModelMetricClaimedVerifiedStatus
1FaceNet+Adaptive ThresholdAverage Accuracy (10 times)83.79—Unverified
2FaceNet+Fixed Threshold (0.3968)Average Accuracy (10 times)80.72—Unverified
#ModelMetricClaimedVerifiedStatus
1GhostFaceNetV2-1Accuracy0.95—Unverified
2ElasticFace-ArcAccuracy0.93—Unverified
#ModelMetricClaimedVerifiedStatus
1FaceNet+Adaptive ThresholdAverage Accuracy (10 times)76.46—Unverified
2FaceNet+Fixed Threshold (0.3779)Average Accuracy (10 times)53.97—Unverified
#ModelMetricClaimedVerifiedStatus
1Wang et al. [5]Rank-194.5—Unverified
2Multi-taskRank-184.32—Unverified
#ModelMetricClaimedVerifiedStatus
1Model with Up Convolution + DoG FilterRank-187.2—Unverified
2DPMRank-183.73—Unverified
#ModelMetricClaimedVerifiedStatus
1MCNTAR @ FAR=0.010.54—Unverified
#ModelMetricClaimedVerifiedStatus
1GhostFaceNetV2-1Accuracy99.91—Unverified
#ModelMetricClaimedVerifiedStatus
1Model with Up Convolution + DoG FilterRank-188.33—Unverified
#ModelMetricClaimedVerifiedStatus
1Partial FCMFR-ALL97.85—Unverified
#ModelMetricClaimedVerifiedStatus
1FaceTransformer+OctupletLossAccuracy0.95—Unverified