Class-specific image denoising using importance sampling
This addresses denoising for specific image classes where clean datasets are available, representing an incremental improvement over existing methods.
The paper tackles image denoising for specific classes like faces and text by proposing a method that uses importance sampling to compute weighted averages of non-local patches, approximating minimum mean squared error estimates, and it outperforms state-of-the-art methods in these classes.
In this paper, we propose a new image denoising method, tailored to specific classes of images, assuming that a dataset of clean images of the same class is available. Similarly to the non-local means (NLM) algorithm, the proposed method computes a weighted average of non-local patches, which we interpret under the importance sampling framework. This viewpoint introduces flexibility regarding the adopted priors, the noise statistics, and the computation of Bayesian estimates. The importance sampling viewpoint is exploited to approximate the minimum mean squared error (MMSE) patch estimates, using the true underlying prior on image patches. The estimates thus obtained converge to the true MMSE estimates, as the number of samples approaches infinity. Experimental results provide evidence that the proposed denoiser outperforms the state-of-the-art in the specific classes of face and text images.