The introduction of CTV-RPCA

The simulation experiments are placed in folder "Simulation"

### Run "test_demo.m" to get metrics
  The mean value of the 50 results
      rank_ratio  sparse_ratio  error of CTV-RPCA     error of RPCA
        0.05          0.05          0.001142            0.002047
        0.10          0.10          0.000593            0.000893
        0.30          0.10          0.011400            0.018900
        0.10          0.30          0.008380            0.021556

### run "lambda_selection.m" to get visualizations
    we can verify that lambda= 3/sqrt(M) is reasonale, 
     where M,N are size of data, and M > N. 

Real experiments

Run "Demo_of_HSI_denoising.m" to get metrics and visualizations

    one result of small Pure DC mall dataset is:

     gaussian_level sparse_level  psnr/ssim of CTV   psnr/ssim of RPCA
          0.10             0.10       32.18/0.9697     30.44/0.9691 
          0.00             0.20       48.46/0.9923     43.54/0.9983
          0.20             0.00       27.82/0.9317     26.72/0.9335

Run "Demo_of_Video_Extraction.m" to get metrics and visualizations

     result of airport dataset is:

                auc of CTV   auc of RPCA
                  0.9156         0.8721

If you have any questions, please contact the author,

  the author's email address is：andrew.pengjj@gmail.com.