Because digital images are not meaningful by themselves, images are often coupled with some descriptive or qualitative data in an image database. Moreover the division of these da...
We present a novel multiscale clustering algorithm inspired by algebraic multigrid techniques. Our method begins with assembling data points according to local similarities. It us...
We present a new histogram distance family, the Quadratic-Chi (QC). QC members are Quadratic-Form distances with a cross-bin χ2 -like normalization. The cross-bin χ2 -like normal...
Clustering, in data mining, is useful for discovering groups and identifying interesting distributions in the underlying data. Traditional clustering algorithms either favor clust...
In this paper we propose diffusion distance, a new dissimilarity measure between histogram-based descriptors. We define the difference between two histograms to be a temperature f...