Fast Digital Image Inpainting

Abstract

Reconstruction of missing or damaged portions of images is an ancient practice used extensively in artwork restoration. Recently, a few digital inpainting models based on the use of partial differential equations have been proposed. Unfortunately, these algorithms are computationally expensive, usually taking a few minutes to restore small portions of an image, which makes them inappropriate for interactive applications. We discuss the causes of inefficiency of these algorithms and present a simple inpainting model that is two to three orders of magnitude faster, while producing results comparable the ones obtained with current methods.