High-Order Nonparametric Belief-Propagation for Fast Image Inpainting

TL;DR

This paper introduces a fast image inpainting method using belief-propagation with Gaussian mixture approximations for high-order models, achieving quick convergence and high-quality visual results.

Contribution

It presents a novel belief-propagation algorithm that efficiently handles high-order models in image inpainting through Gaussian mixture approximation.

Findings

Achieves competitive inpainting results after few iterations

Maintains good visual quality with high-order model approximation

Runs faster than traditional gradient-based methods

Abstract

In this paper, we use belief-propagation techniques to develop fast algorithms for image inpainting. Unlike traditional gradient-based approaches, which may require many iterations to converge, our techniques achieve competitive results after only a few iterations. On the other hand, while belief-propagation techniques are often unable to deal with high-order models due to the explosion in the size of messages, we avoid this problem by approximating our high-order prior model using a Gaussian mixture. By using such an approximation, we are able to inpaint images quickly while at the same time retaining good visual results.