Research Article
Discriminative Label Relaxed Regression with Adaptive Graph Learning
Table 4
Average recognition rate of different methods on the CMU PIE database (%).
| Methods | The number (or proportion) of the training samples per class | 4 (19.05%) | 5 (23.81%) | 6 (28.57%) | 7 (33.33%) | 8 (38.10%) |
| CRC | 79.43 | 84.27 | 87.68 | 88.99 | 92.20 | LRC | 70.72 | 79.61 | 82.86 | 87.57 | 89.62 | FLRC | 78.22 | 80.23 | 84.36 | 88.45 | 91.33 | FME | 70.93 | 77.52 | 81.33 | 85.07 | 86.96 | JGLDA | 78.55 | 79.78 | 79.95 | 88.63 | 88.69 | DLSR | 74.25 | 80.29 | 85.98 | 88.74 | 91.95 | DLRR_AG | 81.50 | 84.71 | 88.15 | 89.75 | 92.59 |
|
|
Bold values indicate the highest recognition rate of all methods.
|