Restore from Restored: Single-image Inpainting
This addresses the limitation of existing inpainting methods in exploiting internal image information, offering a solution for computer vision applications, though it is incremental as it builds on pre-trained networks without architectural changes.
The paper tackles the problem of image inpainting by proposing a self-supervised fine-tuning algorithm that adapts pre-trained networks using internal self-similar patches in the input image, improving inpainting quality by a large margin and achieving state-of-the-art results on benchmark datasets.
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.