Research Article
MAM: Multiple Attention Mechanism Neural Networks for Cross-Age Face Recognition
Table 5
The performance of the improved model with self-attention mechanism in different deployment modes on the CACD-VS and MPROH datasets.
| | Model | ACC on CACD-VS | ACC on MORPH |
| | ResNet50(fine-tuned by MORPH) | 95.31% | 95.83% | | Residual attention ResNet50 with HRC | 98.38% | 98.54% | | Residual attention ResNet50 with HRC + self-attention (stage 1) | 98.74% | 98.86% | | Residual attention ResNet50 with HRC + self-attention (stage 2) | 98.87% | 99.02% | | Residual attention ResNet50 with HRC + self-attention (stage 3) | 98.69% | 98.82% | | Residual attention ResNet50 with HRC + self-attention (stage 4) | 98.56% | 98.69% | | Residual attention ResNet50 with HRC + self-attention (stage 5) | 98.43% | 98.51% |
|
|
