Single molecule fluroescence has been around for a while; when I was in grad school, students in another lab were doing this. They'd label motor proteins (whic

Single molecule fluroescence has been around for a while; when I was in grad school, students in another lab were doing this. They'd label motor proteins (which use energy to move in a specific direction) to visualize them on a surface and calculate their velocity. The particle can be much smaller than the resolution, as long as it's really bright, it will just sort of "smear out" over multiple adjacent pixels and it's possible, with some arcane trickery, to then localize to a sub-pixel.