Skip to main content
. 2022 May 21;19:48. doi: 10.1186/s12984-022-01025-3

Fig. 2.

Fig. 2

Overview of the multi-stage graph convolutional neural network architecture (MS-GCN). MS-GCN generates an initial prediction with multiple blocks of spatial-temporal graph convolutional neural network (ST-GCN) layers and refines the predictions over several stages with multiple blocks of temporal convolutional (TCN) layers. An ST-GCN block is visualized in blue and a TCN block in gray