I would like to write a function (I use Haskell) that can display the actual contents of the screen, if you look at the user, to pre-develop (granulate and / or some screen convolution through the repa package) and feed it into a neural network or other AI architecture.
The type will look like watch :: IO img Is there a library that provides such functions? And if I need not the whole screen, but only the visual output of the program, is there any other way than manually cutting the image?
source share