High-quality image inpainting requires filling missing regions in a damaged image with plausible content. Existing works either fill the regions by copying image patches or generating semantically-coherent patches from region context, while neglect the fact that both visual and semantic plausibility are highly-demanded. In this paper, we propose a Pyramid-context ENcoder Network (PEN-Net) for image inpainting by deep generative models. The PEN-Net is built upon a U-Net structure, which can restore an image by encoding contextual semantics from full resolution input, and decoding the learned semantic features back into images. Specifically, we propose a pyramid-context encoder, which progressively learns region affinity by attention from a high-level semantic feature map and transfers the learned attention to the previous low-level feature map. As the missing content can be filled by attention transfer from deep to shallow in a pyramid fashion, both visual and semantic coherence for image inpainting can be ensured. We further propose a multi-scale decoder with deeply-supervised pyramid losses and an adversarial loss. Such a design not only results in fast convergence in training, but more realistic results in testing. Extensive experiments on various datasets show the superior performance of the proposed network
@article{arxiv.1904.07475,
title = {Learning Pyramid-Context Encoder Network for High-Quality Image Inpainting},
author = {Yanhong Zeng and Jianlong Fu and Hongyang Chao and Baining Guo},
journal= {arXiv preprint arXiv:1904.07475},
year = {2019}
}
Comments
Accepted as a CVPR 2019 poster paper; update SUPP;update Eq5;