A non-parametric statistical method for unsupervised learning of
visual taxonomies is presented. The method automatically organizes a
collection of images into a tree-shaped hierarchy. Each image is
associated with a path through this hierarchy. Similar images
share initial segments of their paths and
therefore have a smaller distance from each
other in the tree. Each internal node in the
hierarchy represents information that is common to images whose paths
pass through that node, thus providing a more compact image
representation. Our experiments show that a
disorgainzed collection of images will be
organized into an intuitive taxonomy. 
Furthermore, we find that the taxonomy allows good image classification
and, in this respect, is superior to the
popular LDA model.