Image Restoration and Inpainting Using Belief Propagation with Dynamic Pruning Optimized Exemplar Algorithm

One of the most important premises of research in image processing happens to be Image inpainting which is a technique to fill missing region or reconstruct damaged area in an image. Through image inpainting one can also remove an undesirable object from an image in visually plausible way. For filling the part of image, it uses information from the neighboring area. In this paper, we present an Exemplar based method for filling in the missing information in an image, and its subsequent optimization through belief propagation which takes structure synthesis and texture synthesis together for achieving its goal. The algorithm is based on priority patch based filling procedure.