Neither is simple.
If you have monocular footage, and apply a "depth map" or, which is the same thing, shear parts of the image but not others, you're going to "uncover" pixels that aren't there in the monocular footage, and that need to be filled somehow. Unless, of course, you always shear...