Abstract

We describe a clustering approach with the emphasis on de- tecting coherent structures in a complex dataset, and illustrate its effec- tiveness with computer vision applications. By complex data, we mean that the attribute variations among the data are too extensive such that clustering based on a single feature representation/descriptor is insuffi- cient to faithfully divide the data into meaningful groups. The proposed method thus assumes the data are represented with various feature rep- resentations, and aims to uncover the underlying cluster structure. To that end, we associate each cluster with a boosting classifier derived from multiple kernel learning, and apply the cluster-specific classifier to feature selection across various descriptors to best separate data of the cluster from the rest. Specifically, we integrate the multiple, correlative training tasks of the cluster-specific classifiers into the clustering procedure, and cast them as a joint constrained optimization problem. Through the op- timization iterations, the cluster structure is gradually revealed by these classifiers, while their discriminant power to capture similar data would be progressively improved owing to better data labeling.