These results illustrate the process of denoising a blurred images, allowing for different kernels in different regions. The denoising is performed on overlapping patches (sub-images), and the kernel is estimated separately from each patch using a Fourier domain computation which assumes that image gradients are correlated (according to the AR model described below).
The estimated kernels are cropped to a size of 11 x 11 pixels, and entries less than 1% of the maximum are set to 0.
λ is the regularization parameter; larger values give more importance to the prior and less to image fidelity. The sparse prior refers to the hyper-Laplacian distribution for image gradients with parameter α = 0.8. The AR priors assume 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 IID priors assume that the gradients are independent.
Compare with the results of denoising with a common kernel.