Enhanced Human Action Recognition Using Fusion of Skeletal Joint Dynamics and Structural Features

<div>(a) Confusion matrix for temporal features. Dataset: UTKinect, #bins <i>k</i> = 5, features considered: orientation angle only. (b) Confusion matrix for the score-level fusion using neural network. Dataset: UTKinect, number of hidden layers = 3, epochs = 50.</div>

Journal of Robotics

fig6

Figure 6

Figure 6: Enhanced Human Action Recognition Using Fusion of Skeletal Joint Dynamics and Structural Features 