Research Article
Directional Correlation Filter Bank for Robust Head Pose Estimation and Face Recognition
Table 1
The results obtained by the different competing methods for head pose estimation based on three feature representations.
| Method | Features | Face Database | PIE | HPI | UMIST |
| PCA | Gray | 57.56% | 78.04% | 52.34% | Gabor | 43.73% | 75.71% | 32.74% | HOG | 89.33% | 89.68% | 86.55% |
| LDA | Gray | 62.20% | 82.34% | 64.09% | Gabor | 62.20% | 75.60% | 61.02% | HOG | 93.54% | 90.64% | 87.28% |
| QRLDA | Gray | 77.05% | 76.00% | 48.48% | Gabor | 60.62% | 70.75% | 16.04% | HOG | 94.98% | 78.00% | 60.91% |
| CFA | Gray | 93.23% | 64.36% | 46.74% | Gabor | 73.63% | 56.09% | 6.47% | HOG | 93.94% | 87.18% | 84.54% |
| DCFB | Gray | 94.11% | 76.55% | 70.56% | Gabor | 75.08% | 87.09% | 79.37% | HOG | 100.00 | 99.61 | 100.00 |
|
|