Resource consumption
I think you're considering the 3-D modelling of the collected data as being too resource intensive for the phone, not the collection of the data. Collecting the actual data is a piece of cake. Take video of surrounding area, and add rough positional data. That is, if you even really need that positional data in the first place.
If you just want to spy on your target and get as much information as possible about things in the room and potentially theft-worthy information on their desks, you just need to take video without even bothering with positional data.
But then, once you've broken into your target's phone, you can also do things like find out the wireless password for the corporate network, and then snoop the network from afar (or even from the phone), collecting passwords, financial information, or any amount of other data that's being sent on the wireless connection.
