Table of contents
FCN network
In short, the difference between FCN and CNN is: the CNN convolutional layer is connected to the fully connected layer; the FCN convolutional layer is still connected to the convolutional layer, and the output is a feature map of the same size as the input
network structure
full convolution part
upsampling part
Skip connection
It can be seen that the more upsampling is done, the finer the segmentation result
Summarize
Unet network
The Unet network is based on the FCN network. Its network architecture is shown in the figure below. Generally speaking, it is very similar to the idea of FCN.