README.md

Training results (50 epochs)

Model	accuracy
FCN	0.889
SegNet	0.906
U-Net	0.901
VGG16 U-Net	0.962
ResNet101 U-Net	0.977
VGG16 UNet 3+	0.977
ResNet101 UNet 3+	0.978

Results. Model: VGG16 UNet 3+

Comparison of result (left image) and true segmentation (right image). Model: VGG16 UNet 3+