What is subspace clustering?

Subspace clustering is an extension of traditional cluster- ing that seeks to find clusters in different subspaces within a dataset. Often in high dimensional data, many dimen- sions are irrelevant and can mask existing clusters in noisy data.

Why subspace clustering?

Often in high dimensional data, many dimensions are irrelevant and can mask existing clusters in noisy data. Subspace clustering algorithms localize the search for relevant dimensions allowing them to find clusters that exist in multiple, possibly overlapping subspaces.

What is clique clustering?

Clique clustering is the problem of partitioning the vertices of a graph into disjoint clusters, where each cluster forms a clique in the graph, while optimizing some objective function. The goal is to maintain a clustering with an objective value close to the optimal solution.

How is data distributed in a subspace cluster?

If the data within each subspace is distributed around a cluster center and the cluster centers for different subspaces are far apart, then the subspace clustering problem reduces to the simpler and well studied central clustering problem, where the data is distributed around multiple cluster centers.

How is subspace clustering used in computer vision?

This problem, known as subspace clustering, has found numerous applications in computer vision (e.g., image segmentation [1], motion segmentation [2] and face clustering [3]), image pro- cessing (e.g., image representation and compression [4]) and systems theory (e.g., hybrid system identification [5]).

How is ordered subspace clustering used in seg-ment?

We propose Ordered Subspace Clustering (OSC) to seg- ment data drawn from a sequentially ordered union of sub- spaces. Current subspace clustering techniques learn the relationships within a set of data and then use a separate clustering algorithm such as NCut for final segmentation.