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

Figure 5

(a) Confusion matrix for structural features. Dataset: KTH, #bits used for encoding b = 8, total number of joints considered: 25, 2D location of joints. (b) Confusion matrix for temporal features. Dataset: KTH, #bins k = 5, features considered: orientation angle only. (c) Confusion matrix for the score-level fusion using neural network. Dataset: KTH, number of hidden layers = 3, epochs = 20. (d) Confusion matrix for structural features. Dataset: UTKinect, #bits used for encoding b = 8.

(a)

(b)

(c)

(d)