An Implement of MADF (Image Inpainting by Cascaded Refinement with Mask Awareness). This is a temporally anonymous repo for reproducing results of our submission.
- Pytorch 1.0.0
- python 3.*
- numpy 1.13.1
-
download this repo by the following instruction:
$ git clone https://github.com/MADF-inpainting/Pytorch-MADF.git $ cd Pytorch-MADF-master
- Mask datasets can be download here. We train all the datasets with the same mask datasets. Notice that mask images must be 0-255 images.
- You can download this dataset here. We use the High-resolution version in this repo.
- You can download this dataset here. We crop and align the images before training and testing.
- You can download this dataset here.
$ python train.py --train_root train_root --mask_root mask_root --test_root test_root --use_incremental_supervision
$ python test.py --list_file test_list --snapshot model_path
- All the released models are trained with images of resolution 512 x 512 and the training mask dataset.
GT | maked input | inpainting result |
---|---|---|