SOTAVerified

Face Verification

Face Verification is a machine learning task in computer vision that involves determining whether two facial images belong to the same person or not. The task involves extracting features from the facial images, such as the shape and texture of the face, and then using these features to compare and verify the similarity between the images.

( Image credit: Pose-Robust Face Recognition via Deep Residual Equivariant Mapping )

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 201–250 of 360 papers

TitleStatusHype
Saliency Supervision: An Intuitive and Effective Approach for Pain Intensity Regression—0
SELM: Siamese Extreme Learning Machine with Application to Face Biometrics—0
Semantic Adversarial Attacks on Face Recognition through Significant Attributes—0
Set Distribution Networks: a Generative Model for Sets of Images—0
Siamese Capsule Networks—0
Single Morphing Attack Detection using Siamese Network and Few-shot Learning—0
Softmax Dissection: Towards Understanding Intra- and Inter-class Objective for Embedding Learning—0
SqueezeFacePoseNet: Lightweight Face Verification Across Different Poses for Mobile Platforms—0
SqueezerFaceNet: Reducing a Small Face Recognition CNN Even More Via Filter Pruning—0
Striking the Right Balance with Uncertainty—0
Structure Destruction and Content Combination for Face Anti-Spoofing—0
Surpassing Human-Level Face Verification Performance on LFW with GaussianFace—0
SymFace: Additional Facial Symmetry Loss for Deep Face Recognition—0
Synthesis of High-Quality Visible Faces from Polarimetric Thermal Faces using Generative Adversarial Networks—0
Targeting Ultimate Accuracy: Face Recognition via Deep Embedding—0
Template Adaptation for Face Verification and Identification—0
Temporal Non-Volume Preserving Approach to Facial Age-Progression and Age-Invariant Face Recognition—0
The Do's and Don'ts for CNN-based Face Verification—0
The Effect of Wearing a Face Mask on Face Image Quality—0
Thermal to Visible Image Synthesis under Atmospheric Turbulence—0
TinySiamese Network for Biometric Analysis—0
Toward Open-Set Face Recognition—0
Toward Privacy and Utility Preserving Image Representation—0
Towards A Comprehensive Visual Saliency Explanation Framework for AI-based Face Recognition Systems—0
Towards Fair Face Verification: An In-depth Analysis of Demographic Biases—0
Towards Protecting Face Embeddings in Mobile Face Verification Scenarios—0
Towards the Design of an End-to-End Automated System for Image and Video-based Recognition—0
Towards Visual Saliency Explanations of Face Verification—0
Training face verification models from generated face identity data—0
Triplet Similarity Embedding for Face Verification—0
Unconstrained Face Detection and Open-Set Face Recognition Challenge—0
Unconstrained Face Verification using Deep CNN Features—0
Unconstrained Still/Video-Based Face Verification with Deep Convolutional Neural Networks—0
Understanding Fake Faces—0
Unifying Specialist Image Embedding into Universal Image Embedding—0
Universal Adversarial Spoofing Attacks against Face Recognition—0
Unsupervised Compound Domain Adaptation for Face Anti-Spoofing—0
Unsupervised Deep Learning Image Verification Method—0
US-GAN: On the importance of Ultimate Skip Connection for Facial Expression Synthesis—0
Utilizing GANs for Fraud Detection: Model Training with Synthetic Transaction Data—0
VariFace: Fair and Diverse Synthetic Dataset Generation for Face Recognition—0
Verification of Very Low-Resolution Faces Using An Identity-Preserving Deep Face Super-Resolution Network—0
V-MAD: Video-based Morphing Attack Detection in Operational Scenarios—0
von Mises-Fisher Mixture Model-based Deep learning: Application to Face Verification—0
WebFace260M: A Benchmark Unveiling the Power of Million-Scale Deep Face Recognition—0
Global and Local Consistent Wavelet-domain Age Synthesis—0
Harnessing Geometric Constraints from Emotion Labels to improve Face Verification—0
Hierarchical-PEP Model for Real-World Face Recognition—0
Hierarchical Representation Learning for Kinship Verification—0
Human-Machine Comparison for Cross-Race Face Verification: Race Bias at the Upper Limits of Performance?—0
Show:102550
← PrevPage 5 of 8Next →
All datasetsIJB-CIJB-AIJB-BMegaFaceYouTube Faces DBBTS3.1Labeled Faces in the WildTrillion Pairs DatasetAgeDB-30CFP-FPBUAA-VisNirCASIA NIR-VIS 2.0

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1WebFace42M baselineTAR @ FAR=1e-497.7—Unverified
2AdaFace (WebFace4M)TAR @ FAR=1e-497.39—Unverified
3FFCTAR @ FAR=1e-497.31—Unverified
4CAFace+AdaFace (WebFace4M)TAR @ FAR=1e-497.3—Unverified
5Partial FCTAR @ FAR=1e-597.23—Unverified
6AdaFace (MS1MV3)TAR @ FAR=1e-497.09—Unverified
7PartialFCTAR @ FAR=1e-596.93—Unverified
8AdaFace (MS1MV2)TAR @ FAR=1e-496.89—Unverified
9ElasticFace-CosTAR @ FAR=1e-496.57—Unverified
10Arc+UNPGTAR @ FAR=1e-496.33—Unverified
#ModelMetricClaimedVerifiedStatus
1Dual-Agent GANsTAR @ FAR=0.0197.6—Unverified
2PFEfuse + matchTAR @ FAR=0.0197.5—Unverified
3SE-GV-4-g1TAR @ FAR=0.0197.2—Unverified
4L2-constrained softmax lossTAR @ FAR=0.0197—Unverified
5VGGFace2_ftTAR @ FAR=0.0196.8—Unverified
6StyleFNMTAR @ FAR=0.0194.6—Unverified
7Deep Residual Equivariant MappingTAR @ FAR=0.0194.4—Unverified
8NANTAR @ FAR=0.0194.1—Unverified
9Template adaptationTAR @ FAR=0.0193.9—Unverified
10All-in-one CNNTAR @ FAR=0.0192.2—Unverified
#ModelMetricClaimedVerifiedStatus
1QMagFaceTAR @ FAR=0.0197.72—Unverified
2Arc+UNPGTAR @ FAR=0.0197.7—Unverified
3Mag+UNPGTAR @ FAR=0.0197.63—Unverified
4Cos+UNPGTAR @ FAR=0.0197.36—Unverified
5CAFace+AdaFace (WebFace4M)TAR @ FAR=0.00196.91—Unverified
6PartialFC(WebFace42M)TAR@FAR=0.000196.71—Unverified
7FPNTAR @ FAR=0.0196.5—Unverified
8SE-GV-3-g2TAR @ FAR=0.0196.4—Unverified
9AdaFace (WebFace4M)TAR@FAR=0.000196.03—Unverified
10AdaFace (MS1MV3)TAR@FAR=0.000195.84—Unverified
#ModelMetricClaimedVerifiedStatus
1ProdpolyAccuracy98.95—Unverified
2ElasticFace-ArcAccuracy98.81—Unverified
3GhostFaceNetV2-1Accuracy98.72—Unverified
4ArcFace + MS1MV2 + R100 + RAccuracy98.48—Unverified
5DiscFaceAccuracy97.44—Unverified
6Dynamic AdaCosAccuracy97.41—Unverified
7SV-AM-SoftmaxAccuracy97.38—Unverified
8CosFaceAccuracy96.65—Unverified
9PFEfuse + matchAccuracy92.51—Unverified
10SphereFace (3-patch ensemble)Accuracy89.14—Unverified
#ModelMetricClaimedVerifiedStatus
1SeqFace, 1 ResNet-64Accuracy98.12—Unverified
2ArcFace + MS1MV2 + R100,Accuracy98.02—Unverified
3CosFaceAccuracy97.6—Unverified
4VGG-FaceAccuracy97.4—Unverified
5PFEfuse+matchAccuracy97.36—Unverified
6QANAccuracy96.17—Unverified
7Light CNN-29Accuracy95.54—Unverified
8Git LossAccuracy95.3—Unverified
9FaceNetAccuracy95.12—Unverified
10SphereFaceAccuracy95—Unverified
#ModelMetricClaimedVerifiedStatus
1ProxyFusion (Adaface)TAR @ FAR=0.010.69—Unverified
2CoNAN (Adaface)TAR @ FAR=0.010.56—Unverified
3NAN (Adaface)TAR @ FAR=0.010.54—Unverified
4MCN (Adaface)TAR @ FAR=0.010.54—Unverified
5CAFace (Adaface)TAR @ FAR=0.010.51—Unverified
6MCN (Arcface)TAR @ FAR=0.010.39—Unverified
7NAN (Arcface)TAR @ FAR=0.010.39—Unverified
#ModelMetricClaimedVerifiedStatus
1ArcFace + MS1MV2 + R100,Accuracy99.83—Unverified
2FaceNetAccuracy99.63—Unverified
3DlibAccuracy99.38—Unverified
4VGG-FaceAccuracy98.78—Unverified
5DeepFaceAccuracy98.37—Unverified
6DeepIDAccuracy97.05—Unverified
7OpenFaceAccuracy92.92—Unverified
#ModelMetricClaimedVerifiedStatus
1SV-AM-SoftmaxAccuracy72.71—Unverified
2AM-SoftmaxAccuracy61.61—Unverified
3Arc-SoftmaxAccuracy57.45—Unverified
4A-SoftmaxAccuracy43.76—Unverified
5F-SoftmaxAccuracy37.14—Unverified
6HM-SoftmaxAccuracy34.46—Unverified
#ModelMetricClaimedVerifiedStatus
1PartialFC(R200)Accuracy0.99—Unverified
2GhostFaceNetV2-1Accuracy0.99—Unverified
3DiscFaceAccuracy0.98—Unverified
4VarGFaceNetAccuracy0.98—Unverified
5VarGNetAccuracy0.97—Unverified
#ModelMetricClaimedVerifiedStatus
1PartialFC (R200)Accuracy1—Unverified
2QMagFaceAccuracy0.99—Unverified
3VarGFaceNetAccuracy0.99—Unverified
4VarGNetAccuracy0.9—Unverified
5QMagFaceAccuracy0.84—Unverified
#ModelMetricClaimedVerifiedStatus
1LightCNN-29 + DVGTAR @ FAR=0.00197.3—Unverified
2DVR Wu et al. (2019)TAR @ FAR=0.00196.9—Unverified
3W-CNN He et al. (2018)TAR @ FAR=0.00191.9—Unverified
#ModelMetricClaimedVerifiedStatus
1LightCNN-29 + DVGTAR @ FAR=0.00199.8—Unverified
2DVR Wu et al. (2019)TAR @ FAR=0.00199.6—Unverified
3W-CNN He et al. (2018)TAR @ FAR=0.00198.4—Unverified
#ModelMetricClaimedVerifiedStatus
1ArcFaceR50 + EM-FRRBFAR33.65—Unverified
2ArcFaceR50 + EM-CBFAR2.44—Unverified
3ArcFaceR50 + EM-FARBFAR2.11—Unverified
#ModelMetricClaimedVerifiedStatus
1LightCNN-29 + DVGTAR @ FAR=0.00192.9—Unverified
2DVR Wu et al. (2019)TAR @ FAR=0.00184.9—Unverified
3W-CNN He et al. (2018)TAR @ FAR=0.00154.6—Unverified
#ModelMetricClaimedVerifiedStatus
1DiscFaceAccuracy96.15—Unverified
2SFaceAccuracy93.95—Unverified
#ModelMetricClaimedVerifiedStatus
1DiscFaceAccuracy93.37—Unverified
2SFaceAccuracy91.05—Unverified
#ModelMetricClaimedVerifiedStatus
1AdaFace+CSFMRank-1 (Video2Booking)72.65—Unverified
2ArcFace+CSFMRank-1 (Video2Booking)65.95—Unverified
#ModelMetricClaimedVerifiedStatus
1SphereFaceAccuracy93.8—Unverified
#ModelMetricClaimedVerifiedStatus
1LightCNN-29 + DVGTAR @ FAR=0.0197.86—Unverified
#ModelMetricClaimedVerifiedStatus
1DeepId2+Accuracy96.5—Unverified
#ModelMetricClaimedVerifiedStatus
1DiscFaceTAR @ FAR=0.135.9—Unverified