DRFNet-Inpainting

to be published.

Prerequisites

• Python 3.7
• NVIDIA GPU + CUDA cuDNN 10.1
• PyTorch 1.8.1

TODO

• Releasing evaluation code.
• Releasing inference codes.
• Releasing pre-trained weights.
• Releasing training codes.

Download Datasets

We use Places2, CelebA-HQ, and Paris Street-View datasets. Liu et al. provides 12k irregular masks as the testing mask.

Run

1. train the model

to be published.

2. test the model

to be published.

Citation

to be published.

Acknowledgments

This code based on LGNet. The evaluation code is borrowed from TFill. Please consider to cite their papers.

@ARTICLE{9730792,
  author={Quan, Weize and Zhang, Ruisong and Zhang, Yong and Li, Zhifeng and Wang, Jue and Yan, Dong-Ming},
  journal={IEEE Transactions on Image Processing}, 
  title={Image Inpainting With Local and Global Refinement}, 
  year={2022},
  volume={31},
  pages={2405-2420}
}
@InProceedings{Zheng_2022_CVPR,
    author    = {Zheng, Chuanxia and Cham, Tat-Jen and Cai, Jianfei and Phung, Dinh},
    title     = {Bridging Global Context Interactions for High-Fidelity Image Completion},
    booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    month     = {June},
    year      = {2022},
    pages     = {11512-11522}
}