Super-resolution on test images

Reset to Original vs Set as Left Image

IID Gaussian AR Gaussian IID Sparse AR Sparse λ=0.01 λ=0.001 Estimated kernel h=1 h=1.5

Super-resolution performed on test images described in the paper, at factors 3x and 2x.

λ is the regularization parameter; higher values give more importance to the prior. The sparse prior refers to the hyper-Laplacian distribution for image gradients with parameter α = 0.8. The AR prior assumes that the image gradients are correlated (according to a simple independent 2-D auto-regressive model with correlation parameters 0.3 and 0.6). The blur kernel is either estimated from the input image resized using linear interpolation (using a Fourier domain computation which assumes the AR model above), or fixed (from the Epanechnikov family) with h giving the bandwidth (see code for details). The estimated kernels are cropped to a size of 5 x 5 pixels, and entries less than 1% of the maximum are set to 0.