Research Article
Incremental Graph Regulated Nonnegative Matrix Factorization for Face Recognition
Table 1
The recognition rates (%) for the FERET database during adding 5 batches of images belonging to one class.
| ā | Supervised methods | Unsupervised methods | ā | BIGNMF-S | GNMF-S | CINMF | BIGNMF-U | GNMF-U | IOPNMF | NMF |
| Start | 70.77 | 70.77 | 67.18 | 68.72 | 68.72 | 61.67 | 62.41 | Add 1 | 70.28 | 69.47 | 66.84 | 68.37 | 65.5 | 61.52 | 61.12 | Add 2 | 69.86 | 69.04 | 67.89 | 67.89 | 66.24 | 61.42 | 60.58 | Add 3 | 69.44 | 70.54 | 67.3 | 67.3 | 67 | 61.4 | 61.89 | Add 4 | 69.22 | 69.1 | 67.09 | 67 | 69 | 61.58 | 61.98 | Add 5 | 70.28 | 69.47 | 66.84 | 68.37 | 65.5 | 61.52 | 61.12 |
|
|