A Directional Diffusion Algorithm for Inpainting

TL;DR

This paper introduces a directional diffusion algorithm that enhances image inpainting by better reconstructing edges, outperforming traditional diffusion methods especially in images with obfuscations like text masks.

Contribution

The paper proposes a novel directional diffusion algorithm that improves edge propagation in image inpainting over standard diffusion techniques.

Findings

Better edge reconstruction in inpainting tasks.

Outperforms regular diffusion in images with text obfuscation.

Effective in reconstructing damaged or obscured images.

Abstract

The problem of inpainting involves reconstructing the missing areas of an image. Inpainting has many applications, such as reconstructing old damaged photographs or removing obfuscations from images. In this paper we present the directional diffusion algorithm for inpainting. Typical diffusion algorithms are bad at propagating edges from the image into the unknown masked regions. The directional diffusion algorithm improves on the regular diffusion algorithm by reconstructing edges more accurately. It scores better than regular diffusion when reconstructing images that are obfuscated by a text mask.