Towards Fine Whole-Slide Skeletal Muscle Image Segmentation through Deep Hierarchically Connected Networks

Figure 2

The illustration of the network architecture. The input image has a ground truth segmentation mask and a boundary map. Black boxes indicate the encoder module while colored boxes indicate the decoder module. One decoder takes the feature maps of one encoder layer as input and outputs one segmentation results. The multiscale outputs of all encoders are concatenated to generate the final segmentation result.