The best I can figure out the illusion is this:

The film is a genuine time lapse of an outdoor scene. Several elements have been manipulated to make it appear as if it were a scale mockup. The use of a tilting lens [as claimed] renders the upper part of the image out of focus, which creates the illusion of macrophotography with the corresponding shallow depth-of-field. Normal filming of such a scene would result in everything being in sharp focus. Further, it appears that some manipulation of colors was performed, adding to the illusion of unreality. We can only go by our experience of observation, and these elements were manipulated to suggest other situations than the ones actually present.

It also reminds me of the age old practice of hyper-stereo, where stereo pairs of scenic shots were made where the individual exposures were separated by distances far greater than the normal interocular distance of approximately 2.5 in. This practice resulted in depth information far beyond the usual limit, but had the, sometimes, unintended effect of making the scene resemble a tabletop tableau.