Research Article
Multilevel and Multiscale Feature Aggregation in Deep Networks for Facial Constitution Classification
Table 2
Classification results of different classifiers under different feature extraction.
| ā | SVM (%) | RF (%) | KNN (%) | NB (%) | Softmax (%) | DT (%) | Catb (%) | Lgb (%) | Xgb (%) |
| HSV | 64.91 | 62.47 | 23.26 | 17.82 | 17.63 | 59.66 | 27.20 | 26.45 | 55.91 | LBP | 64.91 | 60.78 | 24.02 | 14.82 | 16.14 | 56.85 | 58.54 | 23.45 | 57.79 | HOG | 19.14 | 60.78 | 25.7 | 16.32 | 14.07 | 57.22 | 34.52 | 32.65 | 59.29 | VGG16-CI | 67.17 | 66.04 | 66.79 | 66.60 | 67.72 | 64.72 | 66.42 | 68.67 | 67.35 |
|
|