praneethmurthy/ReProCS: MATLAB implementation of "Provable Dynamic Robust PCA or Robust Subspace tracking", IEEE Transactions on Information Theory, 2019. · GitHub

[1] "Provable Dynamic Robust PCA or Robust Subspace Tracking", Praneeth Narayanamurthy and Namrata Vaswani, IEEE Trans. Info. Theory, 2019 (available at arXiv:1705.08948).

1. DemoFB.m - Wrapper containing the real video foreground background separation. The sparse recovery step here uses ell-1 minimization. 
2. DemoDynRPCA.m - Wrapper containing the simulated data experiments. The sparse recovery step here uses CoSAMP. We can use ell-1 too if necessary.
3. AutoReProCS - main function which implements the ReProCS algororithm using CoSaMP.
4. ReProCS_real - main function which implements the ReProCS algorithm using ell-1.

1. Data folder (120 MB) contains .mat files for videos. Pushing this too, because the original source web-page (http://perception.i2r.a-star.edu.sg/bk_model/bk_index.html) is down.
2. YALL1 - folder containing files to implement ell-1 minimization.

1. ncrpca -- code implemented Non-convex Robust PCA, NIPS 14 downloaded from authors' website and its accompaniments
2. cgls -- fast method to implement least squares