Now, using the second special technique, we can create decoder of FCN’s using transposed convolution. A transpose convolution is essentially a reverse convolution in which the forward and the backward passes are swapped. Hence, we call it transpose convolution. Some people may call it deconvolution because it undoes the previous convolution. Since all we’re doing is swapping the order of forward and backward passes, the math is actually exactly the same as what we’ve done earlier. The property of differentiability is thus retain and training is simply the same as previous neural networks.