Research Article
Indian Classical Dance Action Identification and Classification with Convolutional Neural Networks
Table 5
Recognition rates of online dance data identification with different classifiers.
| Classifier | Average recognition rates (%) of online dance data | Batch I training | Batch II training | Batch III training | Testing with the same dataset | Testing with different dataset | Testing with the same dataset | Testing with different dataset | Testing with the same dataset | Testing with different dataset |
| ANN [15] | 82.82 | 79.13 | 84.16 | 80.46 | 85.95 | 81.19 | Deep ANN [16] | 85.27 | 81.26 | 86.45 | 82.91 | 87.43 | 84.87 | SVM [13] | 74.12 | 70.66 | 77.63 | 73.45 | 79.51 | 74.23 | Adaboost [14] | 80.01 | 74.54 | 79.49 | 76.62 | 80.55 | 78.10 | AGM [17] | 85.11 | 80.09 | 86.09 | 81.31 | 86.88 | 88.41 | LeNet [25] | 86.43 | 81.79 | 88.26 | 80.49 | 87.73 | 85.66 | VGG [27] | 88.52 | 84.54 | 89.15 | 85.45 | 88.19 | 86.76 | Our proposed CNN architecture | 90.34 | 88.98 | 91.52 | 89.87 | 91.96 | 89.92 |
|
|