Impostor Seperation #7
Description
If we take the vector of each persons movements, we should be able to predict which impostor is which, thus we should be able to train multiple people at once.
The prediction would us the following:
- Distance (from last position compared to all other faces)
- Direction (compared to last movement direction)
- Face size (hence we get some idea of the z-distance of each face)