Using a Selective Ensemble Support Vector Machine to Fuse Multimodal Features for Human Action Recognition

Figure 1

The proposed system configuration.