Generative Image Inpainting with Contextual Attention
J. Yu, Z. Lin, J. Yang, X. Shen, X. Lu, and T. Huang. (2018)cite arxiv:1801.07892Comment: Accepted in CVPR 2018; add CelebA-HQ results; open sourced; interactive demo available: http://jhyu.me/demo.
Abstract
Recent deep learning based approaches have shown promising results for the
challenging task of inpainting large missing regions in an image. These methods
can generate visually plausible image structures and textures, but often create
distorted structures or blurry textures inconsistent with surrounding areas.
This is mainly due to ineffectiveness of convolutional neural networks in
explicitly borrowing or copying information from distant spatial locations. On
the other hand, traditional texture and patch synthesis approaches are
particularly suitable when it needs to borrow textures from the surrounding
regions. Motivated by these observations, we propose a new deep generative
model-based approach which can not only synthesize novel image structures but
also explicitly utilize surrounding image features as references during network
training to make better predictions. The model is a feed-forward, fully
convolutional neural network which can process images with multiple holes at
arbitrary locations and with variable sizes during the test time. Experiments
on multiple datasets including faces (CelebA, CelebA-HQ), textures (DTD) and
natural images (ImageNet, Places2) demonstrate that our proposed approach
generates higher-quality inpainting results than existing ones. Code, demo and
models are available at: https://github.com/JiahuiYu/generative_inpainting.
%0 Generic
%1 yu2018generative
%A Yu, Jiahui
%A Lin, Zhe
%A Yang, Jimei
%A Shen, Xiaohui
%A Lu, Xin
%A Huang, Thomas S.
%D 2018
%K fans inpainting
%T Generative Image Inpainting with Contextual Attention
%U http://arxiv.org/abs/1801.07892
%X Recent deep learning based approaches have shown promising results for the
challenging task of inpainting large missing regions in an image. These methods
can generate visually plausible image structures and textures, but often create
distorted structures or blurry textures inconsistent with surrounding areas.
This is mainly due to ineffectiveness of convolutional neural networks in
explicitly borrowing or copying information from distant spatial locations. On
the other hand, traditional texture and patch synthesis approaches are
particularly suitable when it needs to borrow textures from the surrounding
regions. Motivated by these observations, we propose a new deep generative
model-based approach which can not only synthesize novel image structures but
also explicitly utilize surrounding image features as references during network
training to make better predictions. The model is a feed-forward, fully
convolutional neural network which can process images with multiple holes at
arbitrary locations and with variable sizes during the test time. Experiments
on multiple datasets including faces (CelebA, CelebA-HQ), textures (DTD) and
natural images (ImageNet, Places2) demonstrate that our proposed approach
generates higher-quality inpainting results than existing ones. Code, demo and
models are available at: https://github.com/JiahuiYu/generative_inpainting.
@misc{yu2018generative,
abstract = {Recent deep learning based approaches have shown promising results for the
challenging task of inpainting large missing regions in an image. These methods
can generate visually plausible image structures and textures, but often create
distorted structures or blurry textures inconsistent with surrounding areas.
This is mainly due to ineffectiveness of convolutional neural networks in
explicitly borrowing or copying information from distant spatial locations. On
the other hand, traditional texture and patch synthesis approaches are
particularly suitable when it needs to borrow textures from the surrounding
regions. Motivated by these observations, we propose a new deep generative
model-based approach which can not only synthesize novel image structures but
also explicitly utilize surrounding image features as references during network
training to make better predictions. The model is a feed-forward, fully
convolutional neural network which can process images with multiple holes at
arbitrary locations and with variable sizes during the test time. Experiments
on multiple datasets including faces (CelebA, CelebA-HQ), textures (DTD) and
natural images (ImageNet, Places2) demonstrate that our proposed approach
generates higher-quality inpainting results than existing ones. Code, demo and
models are available at: https://github.com/JiahuiYu/generative_inpainting.},
added-at = {2019-05-23T19:02:26.000+0200},
author = {Yu, Jiahui and Lin, Zhe and Yang, Jimei and Shen, Xiaohui and Lu, Xin and Huang, Thomas S.},
biburl = {https://www.bibsonomy.org/bibtex/2a7c780d10c20c98cb1cda69ccfedf7f6/dickscheid},
interhash = {06a90ac49d75b245dc495aefabbb8ff2},
intrahash = {a7c780d10c20c98cb1cda69ccfedf7f6},
keywords = {fans inpainting},
note = {cite arxiv:1801.07892Comment: Accepted in CVPR 2018; add CelebA-HQ results; open sourced; interactive demo available: http://jhyu.me/demo},
timestamp = {2019-05-23T19:02:26.000+0200},
title = {Generative Image Inpainting with Contextual Attention},
url = {http://arxiv.org/abs/1801.07892},
year = 2018
}