Slide 10 of 42
So the basic technique of our paper is simple. We're going to be
using ideas from edge detection, but in a different way. Computer
vision edge detectors tell you which pixels in an image belong to
edges. So here's a cross-section of a simple object, and we're
cutting through the boundary here, and here we're plotting the data
value as a function of position along that cut. Through out this
talk I'm going to be using f as the function which gives data value
as a function of position, and v as a variable for data value. In
answer to the question, "where's the edge", and edge detector will
tell you "here", at this position, and the result of running it
over the whole image might be this set of edge pixels.
