For class matching, Hietanen in  compares several
binary descriptors and SIFT with different detectors, including
a dense grid. SIFT preformed better than other descriptors and
dense grid responds very well.
This result leads us to test dense detector for different
It seems that a dense detector is one that uses a dense grid on which to evaluate some feature metric.
This is opposed to "classical" feature detectors that search the image for points with features.
Frankly, spanning a grid over an image to reduce it to a lower resolution does sound like what I'd call a subsampling and image transform, but maybe I'm misunderstanding this!