Algorithm 2 CSC for Segmentation

Input: A set of points $Y={\left\{y_i\right\}}_{i=1}^N\in ℝ$ lying in a union of N subspaces $X={\left\{x_i\right\}}_{i=1}^N$ .

Solve the sparse optimization problem by Equation (7).

Normalize the columns of $X$ .

Obtain a similarity graph (matrix) with N nodes representing the data points by Equation (8).

Apply spectral clustering to the similarity graph.

Output: Segmentation of the data.