Skip to main content
. 2020 Jul 13;20(14):3901. doi: 10.3390/s20143901

Figure 4.

Figure 4

The proposed global perception fusion module based on spatial pyramid pooling. When the features from the two branches of the Siamese encoder are fused, four scale (1 × 1, 2 × 2, 3 × 3, 6 × 6) pooling operations follow. Then these feature maps are upsampled to the same spatial resolution and cascaded together.