I have a CNN that is image to image and maps a binary image input to a binary image output. These are usually simple shapes, like a rectangle or a circle. Usually they become smoothed a bit (the effect of lithography.)
However, for my workflow I need to keep everything in vector defined polygons - again, relatively simple polygons. I need to use a similar transform as that provided by this CNN, but for VECTORS.
Is this something that is possible? I am avoiding rasterizing anything since rasterizing is usually not differentiable.
