Restore from Restored: Single-image Inpainting

TL;DR

This paper introduces a self-supervised fine-tuning method for pre-trained image inpainting networks that leverages internal image information to significantly enhance inpainting quality without requiring ground-truth images.

Contribution

It proposes a novel self-supervised algorithm that adapts pre-trained inpainting models using internal self-similar patches, improving results without altering network architecture.

Findings

Achieves state-of-the-art inpainting results on benchmark datasets.

Significantly improves inpainting quality through internal information exploitation.

Demonstrates effectiveness of self-supervised fine-tuning in image inpainting.

Abstract

Recent image inpainting methods have shown promising results due to the power of deep learning, which can explore external information available from the large training dataset. However, many state-of-the-art inpainting networks are still limited in exploiting internal information available in the given input image at test time. To mitigate this problem, we present a novel and efficient self-supervised fine-tuning algorithm that can adapt the parameters of fully pre-trained inpainting networks without using ground-truth target images. We update the parameters of the pre-trained state-of-the-art inpainting networks by utilizing existing self-similar patches (i.e., self-exemplars) within the given input image without changing the network architecture and improve the inpainting quality by a large margin. Qualitative and quantitative experimental results demonstrate the superiority of the proposed algorithm, and we achieve state-of-the-art inpainting results on publicly available benchmark datasets.