Abstract

Spectral clustering requires robust and meaningful affifinity graphs as input in order to form clusters with desired structures that can well support human intuition. To construct such affifinity graphs is non-trivial due to the ambiguity and uncertainty inherent in the raw data. In contrast to most existing clustering methods that typically employ all available features to construct affifinity matrices with the Euclidean distance, which is often not an accurate representation of the underlying data structures, we propose a novel unsupervised approach to generating more robust affifinity graphs via identifying and exploiting discriminative features for improving spectral clustering. Specififically, our model is capable of capturing and combining subtle similarity information distributed over discriminative feature subspaces for more accurately revealing the latent data distribution and thereby leading to improved data clustering, especially with heterogeneous data sources. We demonstrate the effificacy of the proposed approach on challenging image and video datasets