I am trying to run a semantic segmentation model. The problem is, my data has different resolution for every image and for corresponding segmentation map (image and corresponding segmentation map have same aspect ratio. But aspect ratio changes for every image).
This is problematic because, if I want to feed it to the network, usually network will expect the size of the image to remain same.
I want to resize the images. It is simple for regular images. But, for segmentation masks, if done incorrectly, it will lead to some kind of loss of information at the edges of the objects.
I was searching for techniques to change the sizes of the masks properly but did not find anything too useful. Can someone tell me how to do this resizing properly?
