Skip to content

Latest commit

 

History

History
30 lines (24 loc) · 1.33 KB

ImageReconstruction.md

File metadata and controls

30 lines (24 loc) · 1.33 KB

Dataset preparation

If you finish the testing of stage 1, the warped images and corresponding masks will be generated in the 'ImageAlignment/output' folder. And these images are the input of stage 2.

VGG19 pretrained model

Download the pretrained model of vgg19 from:

https://github.com/machrisaa/tensorflow-vgg

Then move the vgg model to 'ImageReconstruction/vgg19/'.

Training

Modidy the 'ImageReconstruction/Codes/constant.py' to set the 'TRAIN_FOLDER'/'ITERATIONS'/'GPU'. In our experiment, we set 'ITERATIONS' to 200,000.

Train the reconstruction network on UDIS-D:

cd ImageReconstruction/Codes/
python train.py

Testing

Our pretrained reconstruction model can be available at Google Drive or Baidu Cloud(Extraction code: 1234).

Modidy the 'ImageReconstruction/Codes/constant.py' to set the 'TEST_FOLDER'/'GPU'. The path for the checkpoint file can be modified in 'ImageReconstruction/Codes/inference.py'.

Generate the stitched image:

python inference.py

About the pretrain reconstruction model

We are sorry that we delete the pretrained reconstruction model whose result is the same as that in the paper by mistake. We retrain the reconstruction model and release it.