List your company
ResearchFormerly WizWand

Face Verification on IJB-C (test)

98.01TAR@FAR=1e-4TopoFR
90.979692.804894.6396.4552Jan 18, 2022Oct 15, 2022Jul 13, 2023Apr 9, 2024Jan 4, 2025Oct 2, 2025Jun 30, 2026
Updated 4mo ago

Evaluation results

MethodLinks
TopoFRBackbone=R200, Training Data=NeoFace32M, Venue=NeurIPS242024.1098.01——
TopoFRTrain Data=WF42M, Backbone=R-200, Venue=NeurIPS24, Re-implemented status=false2026.0698.01—97.1
Partial FCBackbone=R200, Venue=CVPR222024.1097.97——
UniFaceBackbone=R200, Training Data=WebFace42M, Venue=ICCV232024.1097.91——
UniFaceTrain Data=WF42M, Backbone=R-200, Venue=ICCV23, Re-implemented status=false2026.0697.91—96.68
TransFace-LYear='25, Param (M)=271.6, FLOPs (M)=25.4G, Training Dataset=Glint360K, Latency (ms)=OOM2026.0497.85——
PartialFC (0.3)Train Data=WF42M, Backbone=R-100, Venue=CVPR22, Re-implemented status=false2026.0697.82—96.46
Q-MarginTrain Data=WF42M, Backbone=R-100, alpha=1.25, s=35, m=0.2, Re-implemented status=true2026.0697.78—96.35
TransFace-BYear='25, Param (M)=124.5, FLOPs (M)=11.5G, Training Dataset=Glint360K, Latency (ms)=18.202026.0497.59——
EntMaxTrain Data=WF42M, Backbone=R-100, alpha=1.25, s=64, m=0.5, Re-implemented status=true2026.0697.55—96.11
ArcFaceTrain Data=WF42M, Backbone=R-100, s=64, m=0.5, Re-implemented status=true2026.0697.54—95.7
CosFaceTrain Data=WF42M, Backbone=R-100, s=64, m=0.5, Re-implemented status=true2026.0697.49—95.82
AdaFaceTrain Data=WF4M, Backbone=R-100, Venue=CVPR22, Re-implemented status=false2026.0697.39——
TransFace-SYear='23, Param (M)=86.7, FLOPs (M)=5.8G, Training Dataset=Glint360K, Latency (ms)=14.312026.0497.33——
F2CTrain Data=WF42M, Backbone=R-100, Venue=CVPR22, Re-implemented status=false2026.0697.31——
CAFace + AdaFaceTrain Data=WF4M, Backbone=R-100, Venue=NeurIPS22, Re-implemented status=false2026.0697.3—95.96
QAFaceTrain Data=MS1MV2, Backbone=ResNet†, Venue=WACV23, Re-implemented status=false2026.0697.2——
SparseMaxTrain Data=WF42M, Backbone=R-100, alpha=2, s=1.9, m=0.2, Re-implemented status=true2026.0697.12—95.28
Teacher, SH-KDModel=R1002022.0197.0390.8895.58
Teacher, VanillaModel=R1002022.019791.4995.52
ResNet50-ArcFaceYear='22, Param (M)=43.6, FLOPs (M)=6.3G, Training Dataset=Glint360K, Latency (ms)=3.762026.0496.81——
ElasticFaceTrain Data=MS1MV2, Backbone=R-100, Venue=CVPR22, Re-implemented status=false2026.0696.65——
FaceLiVTv2-LParam (M)=8.52, FLOPs (M)=309, Training Dataset=Glint360K, Latency (ms)=0.71(17.2× ↓)2026.0496.59——
ResNet200-TopoFRYear='24, Param (M)=118.8, FLOPs (M)=23.5G, Training Dataset=Glint360K, Latency (ms)=12.242026.0496.56——
FaceLiVTv2-MParam (M)=7.02, FLOPs (M)=258, Training Dataset=Glint360K, Latency (ms)=0.65(18.8× ↓)2026.0496.42——
CurricularFaceTrain Data=MS1MV2, Backbone=R-100, Venue=CVPR20, Re-implemented status=false2026.0696.1——
FaceLiVTv2-SParam (M)=4.62, FLOPs (M)=179, Training Dataset=Glint360K, Latency (ms)=0.54(22.7× ↓)2026.0495.99——
MagFaceTrain Data=MS1MV2, Backbone=R-100, Venue=CVPR21, Re-implemented status=false2026.0695.97—94.08
SH-KDModel=MobileFaceNet2022.0195.6490.2493.73
L2E+IS-samplingModel=MobileFaceNet2022.0195.49—93.25
L2EModel=MobileFaceNet2022.0195.4888.4793.49
TH-KDModel=MobileFaceNet2022.0195.4889.8293.5
VanillaModel=MobileFaceNet2022.0195.4288.7292.75
L2E+ES-samplingModel=MobileFaceNet2022.0195.39—93.2
AdafaceModel=R-50-512, Params (M)=49.02025.0295.34——
MartinezModel=MobileFaceNet2022.0194.7—92.2
AdafaceModel=R-18-512, Params (M)=29.42025.0294.03——
Efficient Global Neural Architecture Search (medium)Model=R-34-512, Params (M)=37.12025.0293.65——
Efficient Global Neural Architecture Search (small)Model=R-58-256, Params (M)=14.62025.0292.6——
FaceLiVTv2-XSParam (M)=2.9, FLOPs (M)=90, Training Dataset=Glint360K, Latency (ms)=0.43(28.5× ↓)2026.0491.25——