Image Inpainting via Generative Multi-column Convolutional Neural Networks

TL;DR

This paper introduces a multi-column neural network for image inpainting that synthesizes image components in parallel, utilizing a confidence-driven loss and MRF regularization to improve global and local detail reconstruction.

Contribution

It presents a novel multi-column network architecture with specialized loss functions for improved image inpainting without post-processing.

Findings

Produces visually compelling inpainting results

Effective on diverse image types including street view and faces

Outperforms existing methods in quality and detail

Abstract

In this paper, we propose a generative multi-column network for image inpainting. This network synthesizes different image components in a parallel manner within one stage. To better characterize global structures, we design a confidence-driven reconstruction loss while an implicit diversified MRF regularization is adopted to enhance local details. The multi-column network combined with the reconstruction and MRF loss propagates local and global information derived from context to the target inpainting regions. Extensive experiments on challenging street view, face, natural objects and scenes manifest that our method produces visual compelling results even without previously common post-processing.