Separating images

This looks interesting: using a kind of dual measurement to try to distinguish two very close sources. Rayleigh showed that two point sources couldn't be distinguished if their angular separation were less than a factor times the wavelength divided by the lens diameter. But their research suggests that if you are not doing direct imaging, it may be possible to do much better. Unfortunately I don't understand their paper well enough to say how practical this is likely to be, but I'll keep an eye on it. Maybe we'll be able to--well, "get a look" is direct imaging. How about "detect" binary star systems? OK, we detect them with doppler shifts and with occlusion effects if we're lucky. You know what I mean.