Research Article
Bio-Inspired Structure Representation Based Cross-View Discriminative Subspace Learning via Simultaneous Local and Global Alignment
Table 2
Classification results (%) of all methods on 10% corrupted CMU-PIE dataset.
| Methods | C1 | C2 | C3 | C4 | C5 | C6 |
| LDA | 20.45 ± 0.04 | 21.01 ± 0.06 | 19.16 ± 0.17 | 16.42 ± 0.12 | 13.64 ± 0.15 | 14.96 ± 0.05 | PCA | 48.09 ± 0.74 | 46.90 ± 0.25 | 47.74 ± 0.31 | 47.99 ± 0.08 | 46.90 ± 0.10 | 46.63 ± 0.06 | LatLRR | 59.14 ± 1.82 | 60.38 ± 1.10 | 57.94 ± 0.81 | 58.10 ± 1.24 | 57.03 ± 0.56 | 56.98 ± 0.96 | LPP | 38.47 ± 0.27 | 34.97 ± 0.80 | 37.85 ± 0.47 | 38.06 ± 0.24 | 33.16 ± 0.63 | 35.77 ± 0.57 | SRRS | 69.54 ± 0.41 | 66.91 ± 0.96 | 70.02 ± 0.96 | 68.89 ± 0.31 | 70.28 ± 0.97 | 65.64 ± 0.51 | RCVL | 87.14 ± 0.04 | 78.10 ± 0.07 | 85.52 ± 0.12 | 77.75 ± 0.07 | 86.87 ± 0.10 | 76.09 ± 0.05 | Ours | 88.68 ± 0.05 | 79.22 ± 0.02 | 87.31 ± 0.04 | 79.75 ± 0.11 | 90.69 ± 0.09 | 78.39 ± 0.04 |
|
|