So, if I understand from the abstract, for every pixel they look through the image for similar "pixel neighbourhoods" of the same or varying size, and collect t

So, if I understand from the abstract, for every pixel they look through the image for similar "pixel neighbourhoods" of the same or varying size, and collect them into a kind of database of examples to be used for scaling up that pixel. Pretty cool idea!