Face Verification on CFP-FP (test)
99.45AccuracyResNet200-TopoFR
Evaluation results
| Method | Links | |
|---|---|---|
| ResNet200-TopoFRYear='24, Param (M)=118.8, FLOPs (M)=23.5G, Training Dataset=Glint360K, Latency (ms)=12.242026.04 | 99.45 | |
| TransFace-LYear='25, Param (M)=271.6, FLOPs (M)=25.4G, Training Dataset=Glint360K, Latency (ms)=OOM2026.04 | 99.37 | |
| TransFace-BYear='25, Param (M)=124.5, FLOPs (M)=11.5G, Training Dataset=Glint360K, Latency (ms)=18.202026.04 | 99.24 | |
| TransFace-SYear='23, Param (M)=86.7, FLOPs (M)=5.8G, Training Dataset=Glint360K, Latency (ms)=14.312026.04 | 98.91 | |
| ResNet50-ArcFaceYear='22, Param (M)=43.6, FLOPs (M)=6.3G, Training Dataset=Glint360K, Latency (ms)=3.762026.04 | 98.77 | |
| MagFaceBackbone=ResNet-1002022.03 | 98.62 | |
| ArcFaceBackbone=ResNet-1002022.03 | 98.6 | |
| Arc+UNPGBackbone=ResNet-1002022.03 | 98.6 | |
| Mag+UNPGBackbone=ResNet-1002022.03 | 98.52 | |
| Cos+UNPGBackbone=ResNet-1002022.03 | 98.5 | |
| MagFace*Backbone=ResNet-1002022.03 | 98.46 | |
| FaceLiVTv2-LParam (M)=8.52, FLOPs (M)=309, Training Dataset=Glint360K, Latency (ms)=0.71(17.2× ↓)2026.04 | 98.26 | |
| FaceLiVTv2-MParam (M)=7.02, FLOPs (M)=258, Training Dataset=Glint360K, Latency (ms)=0.65(18.8× ↓)2026.04 | 97.93 | |
| CosFaceBackbone=ResNet-1002022.03 | 97.72 | |
| FaceLiVTv2-SParam (M)=4.62, FLOPs (M)=179, Training Dataset=Glint360K, Latency (ms)=0.54(22.7× ↓)2026.04 | 97.47 | |
| Circle-loss*Backbone=ResNet-1002022.03 | 96.02 | |
| FaceLiVTv2-XSParam (M)=2.9, FLOPs (M)=90, Training Dataset=Glint360K, Latency (ms)=0.43(28.5× ↓)2026.04 | 95.23 |