Given an arff formatted dataset of continuous image features and a double array containing the desired
percentiles, this method calculates an returns the variance values corresponding to these percentiles.
This class creates a codebook from a set of local features using a slightly modified (to produce some
additional output) version of Weka's SimpleKMeans class.
This method takes as input a folder containing .surf or .sift files extracted from a set of images,
quantizes all the descriptors against a given codebooks and writes in a file the distribution of
distances in each centroid.
Takes a query vector as input and returns a look-up table containing the distance between each
sub-vector of the query vector from each centroid of each sub-quantizer.
Takes a query residual vector as input and returns a look-up table containing the distance between each
sub-vector of the query vector from each centroid of each sub-quantizer.
This method returns an array of Result objects, corresponding to the k nearest neighbors' ids, names
and their distances from the query vector, ordered by lowest distance.
This method returns a bounded priority queue of Result objects, corresponding to the k nearest
neighbors' ids and their distances from the query vector, ordered by lowest distance.