Deblurring-by-Denoising using Spatially Adaptive Gaussian Scale Mixtures in Overcomplete Pyramids

摘要: In a previous work, we presented an extension of the original Bayes Least Squares - Gaussian Scale Mixtures (BLS-GSM) denoising algorithm that also compensated blur. However, method had some problems: a) it could not compensate for blurring kernels; b) its performance depended critically on having accurate estimation power spectral density (PSD); and c) be easily adapted to spatially variant description image statistics. this work propose two-step restoration overcomes these problems by first performing global blur compensation, then applying adaptive local denoising, in overcomplete pyramid. Our is efficient, robust non-iterative. We demonstrate through simulations provides state-of-the-art performance.