Face Identification on IJB-B 1:N Identification
95.52Rank-1Prodpoly-ResNet50
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| Prodpoly-ResNet50Backbone=ResNet-502020.06 | 95.52 | — | — | 97.16 | — | 85.58 | 94.69 | |
| ResNet50Backbone=ResNet-502020.06 | 95.29 | — | — | 97.14 | — | 84.7 | 94.01 | |
| SE-GV-3Training dataset=VF2, D=1282018.10 | 92.1 | 76.4 | 88.5 | 95.5 | 96.2 | — | — | |
| SE-GV-4-g1Training dataset=VF2, D=1282018.10 | 92.1 | 77.6 | 88.8 | 95.6 | 96.4 | — | — | |
| SETraining dataset=MS+VF2, D=20482018.10 | 90.2 | 74.3 | 86.3 | 94.6 | 95.9 | — | — | |
| SETraining dataset=VF2, D=20482018.10 | 90.1 | 70.6 | 83.9 | 94.5 | 95.8 | — | — |