Form a sparse a low rank factorization of a dataset using samples from the data
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.


Read Me !!

1. Setup
- Download and install OMPBox from Rob Rubinstein:

- Download and install oASIS solver:

- Add all folders in oASIS, OMPbox, and SEED to your path

2. To get started
- To run SEED on a synthetic dataset consisting of a union of subspaces, run uos_demo.m
- To run SEED on a real dataset consisting of a collection of face images under different illumination conditions, run face_demo.m (This face data is taken from a subset of the YaleB face database,

3. References
The paper associated with SEED:
E.L. Dyer, T.A. Goldstein, R. Patel, K.P. Kording, R.G. Baraniuk,"Self-Expressive Decompositions for Matrix Approximation and Clustering",

The paper associated with our column sampling method oASIS:
R. Patel, T.A. Goldstein, E.L. Dyer, A. Mirhoseini, R.G. Baraniuk, "oASIS: Adaptive Column Sampling for Kernel Matrix Approximation",

** Code and examples for the sampling method (oASIS) used in SEED can be found at: