Where is the Fake? Patch-Wise Supervised GANs for Texture Inpainting

TL;DR

This paper introduces a patch-wise supervised GAN approach for texture inpainting that effectively uses local information to improve the realism and consistency of generated textures, outperforming existing methods.

Contribution

A novel segmentor discriminator with patch-wise supervision that enhances local texture continuity in GAN-based inpainting.

Findings

Achieves state-of-the-art performance on DTD dataset.

Better handles local consistency in texture inpainting.

Outperforms existing methods in visual realism.

Abstract

We tackle the problem of texture inpainting where the input images are textures with missing values along with masks that indicate the zones that should be generated. Many works have been done in image inpainting with the aim to achieve global and local consistency. But these works still suffer from limitations when dealing with textures. In fact, the local information in the image to be completed needs to be used in order to achieve local continuities and visually realistic texture inpainting. For this, we propose a new segmentor discriminator that performs a patch-wise real/fake classification and is supervised by input masks. During training, it aims to locate the fake and thus backpropagates consistent signal to the generator. We tested our approach on the publicly available DTD dataset and showed that it achieves state-of-the-art performances and better deals with local consistency than existing methods.