Figure 8
From: Semantic segmentation of COVID-19 lesions with a multiscale dilated convolutional network
Lung infection segmentation results of different networks, where the ground-glass, consolidation, and pleural effusion are marked in dark gray, bright gray, and white, respectively. The first column is the ground truth. The second column is the segmentation result of the ResNet-50 network using dilated convolution. The last column is the segmentation result of the baseline based on ResNet-50.
