A simple greedy clustering algorithm which puts windows in the same cluster if they are close to each other, and combines clusters if they are close the goal is to find a set of windows which all represent a good location for the object
This state class holds the results of
The detector cascade prepares, inititates and controls the 3 underlying detection steps in the TLD algorithm.
The ensemble classifier implements a forest of random binary pixel comparisons.
a 2D double array such that each element is the sum of each value to the top left of the element,
The third and most powerful, but equally most slow parts of the
Defines a intensity normalised patch extracted from an image.
A rectangle that knows which scale index it belongs to
The variance filter forms the first step of the