Sequential superparamagnetic clustering

Clustering is defined as a task of grouping of similar objects. As the concept of similarity between objects is ambiguous, the clustering task is inherently ill-defined. The Sequential Super-paramagnetic Clustering (SSC) algorithm approaches the clustering task by providing a unique clustering hierarchy along with a measure of the naturalness of the cluster (where naturalness is defined as a group without any significant substructure). The SSC algorithm does not assume any a priori information about shape or internal distribution of the clusters, nor is the number of clusters predefined. Moreover, the SSC algorithm can easily deal with clusters of different shapes, densities and largely unequal distances between clusters. Interrelation SSC- visual perception: When viewing a scene that is larger than the human eyes field of view, the eye will perform rapid eye movements or saccades, followed by time intervals where the eye is stationary on areas of interest which are defined as visual fixations. These combined saccadic motions and intervals of visual fixations are collectively known as the visual scan path. This project involves the use of SSC algorithms to cluster images into areas of interest and relate the visual scan paths to the measure of the naturalness of each cluster.


© 2018 Institut für Neuroinformatik