I assume you're talking of the GH-like game?

Yes, a background process certainly could assume all presses are in 4 carefully-selected (to remove ambiguity) positions and resolve them accordingly -- but why in the world you would want that I don't know. It's clever, to be sure, but it's not general-purpose.

How hard is it to understand that the single pressure-weighted average of n touch locations (exactly what a standard resistive screen measures) contains less information than a list of each position? You can't just summon the missing data from nowhere!