Table 10.
Detection errors (APCER, BPCER, ACER, and HTER) of our proposed method with the Replay-mobile dataset using two types of PAI (unit: %).
| PAI | EER | Matte-Screen Attack | Print Attack | Overall | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| APCER | BPCER | ACER | APCER | BPCER | ACER | APCER | BPCER | ACER | HTER | ||
| Using CNN Features [27] | 0.067 | 0.000 | 0.009 | 0.0045 | 0.000 | 0.009 | 0.0045 | 0.000 | 0.009 | 0.0045 | 0.0045 |
| Using CNN-RNN Features | 0.002 | 0.000 | 0.003 | 0.0015 | 0.000 | 0.003 | 0.0015 | 0.000 | 0.003 | 0.0015 | 0.0015 |
| Using only MLBP features | 4.659 | 8.820 | 1.937 | 5.379 | 2.451 | 1.937 | 2.194 | 8.820 | 1.937 | 5.379 | 5.684 |
| FLF | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 |
| SLF | 0.000 | 0.000 | 0.003 | 0.0015 | 0.000 | 0.003 | 0.0015 | 0.000 | 0.003 | 0.0015 | 0.0015 |