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 Added Most Hyped Most Active Needs Verification Most Verified

Showing 276–300 of 360 papers

Title	Date	Tasks	Status
Toward Open-Set Face Recognition	May 3, 2017	Face IdentificationFace Recognition	—Unverified
SphereFace: Deep Hypersphere Embedding for Face Recognition	Apr 26, 2017	Face IdentificationFace Recognition	CodeCode Available
NormFace: L2 Hypersphere Embedding for Face Verification	Apr 21, 2017	Face VerificationMetric Learning	CodeCode Available
Quality Aware Network for Set to Set Recognition	Apr 11, 2017	Face VerificationPerson Re-Identification	CodeCode Available
L2-constrained Softmax Loss for Discriminative Face Verification	Mar 28, 2017	Face Verification	CodeCode Available
Temporal Non-Volume Preserving Approach to Facial Age-Progression and Age-Invariant Face Recognition	Mar 24, 2017	Age-Invariant Face RecognitionDensity Estimation	—Unverified
Discriminate-and-Rectify Encoders: Learning from Image Transformation Sets	Mar 14, 2017	Face VerificationObject Recognition	—Unverified
A Proximity-Aware Hierarchical Clustering of Faces	Mar 14, 2017	ClusteringFace Clustering	—Unverified
Regularizing Face Verification Nets For Pain Intensity Regression	Feb 22, 2017	Face VerificationPain Intensity Regression	CodeCode Available
Deep Heterogeneous Feature Fusion for Template-Based Face Recognition	Feb 15, 2017	Face RecognitionFace Verification	—Unverified
Regressing Robust and Discriminative 3D Morphable Models with a very Deep Neural Network	Dec 15, 2016	3D Face ReconstructionFace Recognition	CodeCode Available
Joint Bayesian Gaussian discriminant analysis for speaker verification	Dec 13, 2016	Face VerificationSpeaker Verification	—Unverified
Improved Deep Metric Learning with Multi-class N-pair Loss Objective	Dec 1, 2016	ClusteringFace Verification	—Unverified
Cross-Domain Face Verification: Matching ID Document and Self-Portrait Photographs	Nov 17, 2016	Face Verification	—Unverified
DeMeshNet: Blind Face Inpainting for Deep MeshFace Verification	Nov 16, 2016	Face AlignmentFace Verification	—Unverified
An All-In-One Convolutional Neural Network for Face Analysis	Nov 3, 2016	Age EstimationAll	CodeCode Available
Active User Authentication for Smartphones: A Challenge Data Set and Benchmark Results	Oct 25, 2016	Face DetectionFace Verification	—Unverified
Learning Grimaces by Watching TV	Oct 7, 2016	Emotion RecognitionFace Verification	—Unverified
OPML: A One-Pass Closed-Form Solution for Online Metric Learning	Sep 29, 2016	Event DetectionFace Verification	—Unverified
Pose-Selective Max Pooling for Measuring Similarity	Sep 22, 2016	DiversityFace Recognition	CodeCode Available
A Discriminative Feature Learning Approach for Deep Face Recognition	Sep 16, 2016	Face RecognitionFace Verification	CodeCode Available
Online Feature Selection with Group Structure Analysis	Aug 21, 2016	Face Verificationfeature selection	—Unverified
Face Detection with End-to-End Integration of a ConvNet and a 3D Model	Jun 2, 2016	Face DetectionFace Model	CodeCode Available
Recurrent Face Aging	Jun 1, 2016	Face Verification	—Unverified
Cross-Domain Visual Matching via Generalized Similarity Measure and Feature Learning	May 13, 2016	Face VerificationModel Optimization	—Unverified

Show:10 25 50

← PrevPage 12 of 15Next →

All datasets IJB-C IJB-A IJB-B MegaFace YouTube Faces DB BTS3.1 Labeled Faces in the Wild Trillion Pairs Dataset AgeDB-30 CFP-FP BUAA-VisNir CASIA NIR-VIS 2.0

Benchmark Results

#	Model	Metric	Claimed	Verified	Status
1	WebFace42M baseline	TAR @ FAR=1e-4	97.7	—	Unverified
2	AdaFace (WebFace4M)	TAR @ FAR=1e-4	97.39	—	Unverified
3	FFC	TAR @ FAR=1e-4	97.31	—	Unverified
4	CAFace+AdaFace (WebFace4M)	TAR @ FAR=1e-4	97.3	—	Unverified
5	Partial FC	TAR @ FAR=1e-5	97.23	—	Unverified
6	AdaFace (MS1MV3)	TAR @ FAR=1e-4	97.09	—	Unverified
7	PartialFC	TAR @ FAR=1e-5	96.93	—	Unverified
8	AdaFace (MS1MV2)	TAR @ FAR=1e-4	96.89	—	Unverified
9	ElasticFace-Cos	TAR @ FAR=1e-4	96.57	—	Unverified
10	Arc+UNPG	TAR @ FAR=1e-4	96.33	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Dual-Agent GANs	TAR @ FAR=0.01	97.6	—	Unverified
2	PFEfuse + match	TAR @ FAR=0.01	97.5	—	Unverified
3	SE-GV-4-g1	TAR @ FAR=0.01	97.2	—	Unverified
4	L2-constrained softmax loss	TAR @ FAR=0.01	97	—	Unverified
5	VGGFace2_ft	TAR @ FAR=0.01	96.8	—	Unverified
6	StyleFNM	TAR @ FAR=0.01	94.6	—	Unverified
7	Deep Residual Equivariant Mapping	TAR @ FAR=0.01	94.4	—	Unverified
8	NAN	TAR @ FAR=0.01	94.1	—	Unverified
9	Template adaptation	TAR @ FAR=0.01	93.9	—	Unverified
10	All-in-one CNN	TAR @ FAR=0.01	92.2	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	QMagFace	TAR @ FAR=0.01	97.72	—	Unverified
2	Arc+UNPG	TAR @ FAR=0.01	97.7	—	Unverified
3	Mag+UNPG	TAR @ FAR=0.01	97.63	—	Unverified
4	Cos+UNPG	TAR @ FAR=0.01	97.36	—	Unverified
5	CAFace+AdaFace (WebFace4M)	TAR @ FAR=0.001	96.91	—	Unverified
6	PartialFC(WebFace42M)	TAR@FAR=0.0001	96.71	—	Unverified
7	FPN	TAR @ FAR=0.01	96.5	—	Unverified
8	SE-GV-3-g2	TAR @ FAR=0.01	96.4	—	Unverified
9	AdaFace (WebFace4M)	TAR@FAR=0.0001	96.03	—	Unverified
10	AdaFace (MS1MV3)	TAR@FAR=0.0001	95.84	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Prodpoly	Accuracy	98.95	—	Unverified
2	ElasticFace-Arc	Accuracy	98.81	—	Unverified
3	GhostFaceNetV2-1	Accuracy	98.72	—	Unverified
4	ArcFace + MS1MV2 + R100 + R	Accuracy	98.48	—	Unverified
5	DiscFace	Accuracy	97.44	—	Unverified
6	Dynamic AdaCos	Accuracy	97.41	—	Unverified
7	SV-AM-Softmax	Accuracy	97.38	—	Unverified
8	CosFace	Accuracy	96.65	—	Unverified
9	PFEfuse + match	Accuracy	92.51	—	Unverified
10	SphereFace (3-patch ensemble)	Accuracy	89.14	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SeqFace, 1 ResNet-64	Accuracy	98.12	—	Unverified
2	ArcFace + MS1MV2 + R100,	Accuracy	98.02	—	Unverified
3	CosFace	Accuracy	97.6	—	Unverified
4	VGG-Face	Accuracy	97.4	—	Unverified
5	PFEfuse+match	Accuracy	97.36	—	Unverified
6	QAN	Accuracy	96.17	—	Unverified
7	Light CNN-29	Accuracy	95.54	—	Unverified
8	Git Loss	Accuracy	95.3	—	Unverified
9	FaceNet	Accuracy	95.12	—	Unverified
10	SphereFace	Accuracy	95	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	ProxyFusion (Adaface)	TAR @ FAR=0.01	0.69	—	Unverified
2	CoNAN (Adaface)	TAR @ FAR=0.01	0.56	—	Unverified
3	NAN (Adaface)	TAR @ FAR=0.01	0.54	—	Unverified
4	MCN (Adaface)	TAR @ FAR=0.01	0.54	—	Unverified
5	CAFace (Adaface)	TAR @ FAR=0.01	0.51	—	Unverified
6	MCN (Arcface)	TAR @ FAR=0.01	0.39	—	Unverified
7	NAN (Arcface)	TAR @ FAR=0.01	0.39	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	ArcFace + MS1MV2 + R100,	Accuracy	99.83	—	Unverified
2	FaceNet	Accuracy	99.63	—	Unverified
3	Dlib	Accuracy	99.38	—	Unverified
4	VGG-Face	Accuracy	98.78	—	Unverified
5	DeepFace	Accuracy	98.37	—	Unverified
6	DeepID	Accuracy	97.05	—	Unverified
7	OpenFace	Accuracy	92.92	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SV-AM-Softmax	Accuracy	72.71	—	Unverified
2	AM-Softmax	Accuracy	61.61	—	Unverified
3	Arc-Softmax	Accuracy	57.45	—	Unverified
4	A-Softmax	Accuracy	43.76	—	Unverified
5	F-Softmax	Accuracy	37.14	—	Unverified
6	HM-Softmax	Accuracy	34.46	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	PartialFC(R200)	Accuracy	0.99	—	Unverified
2	GhostFaceNetV2-1	Accuracy	0.99	—	Unverified
3	DiscFace	Accuracy	0.98	—	Unverified
4	VarGFaceNet	Accuracy	0.98	—	Unverified
5	VarGNet	Accuracy	0.97	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	PartialFC (R200)	Accuracy	1	—	Unverified
2	QMagFace	Accuracy	0.99	—	Unverified
3	VarGFaceNet	Accuracy	0.99	—	Unverified
4	VarGNet	Accuracy	0.9	—	Unverified
5	QMagFace	Accuracy	0.84	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	LightCNN-29 + DVG	TAR @ FAR=0.001	97.3	—	Unverified
2	DVR Wu et al. (2019)	TAR @ FAR=0.001	96.9	—	Unverified
3	W-CNN He et al. (2018)	TAR @ FAR=0.001	91.9	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	LightCNN-29 + DVG	TAR @ FAR=0.001	99.8	—	Unverified
2	DVR Wu et al. (2019)	TAR @ FAR=0.001	99.6	—	Unverified
3	W-CNN He et al. (2018)	TAR @ FAR=0.001	98.4	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	ArcFaceR50 + EM-FRR	BFAR	33.65	—	Unverified
2	ArcFaceR50 + EM-C	BFAR	2.44	—	Unverified
3	ArcFaceR50 + EM-FAR	BFAR	2.11	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	LightCNN-29 + DVG	TAR @ FAR=0.001	92.9	—	Unverified
2	DVR Wu et al. (2019)	TAR @ FAR=0.001	84.9	—	Unverified
3	W-CNN He et al. (2018)	TAR @ FAR=0.001	54.6	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	DiscFace	Accuracy	96.15	—	Unverified
2	SFace	Accuracy	93.95	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	DiscFace	Accuracy	93.37	—	Unverified
2	SFace	Accuracy	91.05	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	AdaFace+CSFM	Rank-1 (Video2Booking)	72.65	—	Unverified
2	ArcFace+CSFM	Rank-1 (Video2Booking)	65.95	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SphereFace	Accuracy	93.8	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	LightCNN-29 + DVG	TAR @ FAR=0.01	97.86	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	DeepId2+	Accuracy	96.5	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	DiscFace	TAR @ FAR=0.1	35.9	—	Unverified