Patch reprojections for Non-Local methods

Since their introduction in image denoising, the family of non-local methods, whose Non-Local Means (NL-Means) is the most famous member, has proved its ability to challenge other powerful methods such as wavelet based approaches or variational techniques. Though simple to implement and efficient in practice, the classical NL-Means algorithm suffers from several limitations: noise artifacts are created around edges and regions with few repetitions in the image are not treated at all. In this paper, we present an easy to implement and time efficient modification of the NL-Means based on a better reprojection from the patches space to the original pixel space, specially designed to reduce the artifacts due to the rare patch effect. We compare the performance of several reprojection schemes on a toy example and on some classical natural images.

► We study one of the step required in patch-based denoising algorithms.
► This step links the information gained in the patch domain to the pixel domain.
► We call it the “reprojection” process.
► We show non-uniform reprojections lower visual artefacts of patch-based methods.