I’m looking for a way to get the 3d world orientation vector components from an arbitrary camera translated into 2d camera space (per pixel, translated into 3 TOPs - one per coordinate). What would be the workflow for this?
