This function calculates the sequence difficulty from the results of the trackers on the sequences using accuracy and robustness measures. These two measures are quantized into ten levels. The quantized robustness was calculated by clipping the raw robustness at nine failures and the quantized accuracy was computed by 9 - floor(10*Phi), where Phi is the VOT accuracy. This function also print the inner-cluster performance variation, to compare and validate affine propagation and k-means clustering quality.
Input:
experiments (cell): An array of experiment structures.
sequences (cell): an array of sequence structures.
trackers (struct): An array of tracker structures.
clusters_ap (structure): clustering structure of affine propagation clustering alg.
clusters_kmeans (structure): clustering structure of k-means clustering alg.
Output: