J-Net: Asymmetric Encoder-Decoder for Medical Semantic Segmentation

Table 1

Segmentation results produced by different methods on the liver datasets.

Method	Backbone	Precision	Dice	Recall
FCN-8s	VGG16	0.8787	0.8537	0.8420
DeepLab-v3+	ResNet101	0.8773	0.8642	0.8605
U-Net	—	0.8774	0.9172	0.9622
J-Net	ResNet101	0.8836	0.9118	0.9637