Image Inpainting for High-Resolution Textures using CNN Texture Synthesis

TL;DR

This paper introduces a CNN-based patch inpainting method for high-resolution textures that effectively maintains global coherence and detail, outperforming existing techniques in filling large regions.

Contribution

It presents a novel patch-based CNN inpainting approach that optimizes global and detail texture statistics for high-resolution images, addressing computational limitations.

Findings

Capable of filling large inpainting regions with high quality

Outperforms comparable methods in high-resolution texture inpainting

Uses texture summary statistics for reference patch look-up

Abstract

Deep neural networks have been successfully applied to problems such as image segmentation, image super-resolution, coloration and image inpainting. In this work we propose the use of convolutional neural networks (CNN) for image inpainting of large regions in high-resolution textures. Due to limited computational resources processing high-resolution images with neural networks is still an open problem. Existing methods separate inpainting of global structure and the transfer of details, which leads to blurry results and loss of global coherence in the detail transfer step. Based on advances in texture synthesis using CNNs we propose patch-based image inpainting by a CNN that is able to optimize for global as well as detail texture statistics. Our method is capable of filling large inpainting regions, oftentimes exceeding the quality of comparable methods for high-resolution images. For reference patch look-up we propose to use the same summary statistics that are used in the inpainting process.