Research Article
Instructor Activity Recognition through Deep Spatiotemporal Features and Feedforward Extreme Learning Machines
Table 1
Performance of the proposed technique at different network depth on the IAVID-I dataset using LOAO validation scheme.
| Deep spatiotemporal features | Features | No. Layers | Hidden Node | LOAO Accuracy % | Time Sec |
| x17 | 1x4096 | 17 | 1300 | 67.98 | 0.00159 | x 20 | 1x 4096 | 20 | 1300 | 46.25 | 0.00203 | x 23 | 1x 1000 | 23 | 1300 | 43.10 | 0.00108 | x 39 | 1x 4096 | 39 | 500 | 55.74 | 0.00059 | x 42 | 1x 4096 | 42 | 500 | 44.96 | 0.001302 | x 45 | 1x 1000 | 45 | 500 | 43.13 | 0.000811 |
|
|