This repository provides code to reproduce results from the paper: Compressed Sensing using Generative Models.
Here are a few example results:
Reconstruction | Super-resolution | Inpainting |
---|---|---|
Here we show the evolution of the reconstructed image for different number of iterations:
NOTE: Please run all commands from the root directory of the repository, i.e from csgm/
- Python 2.7
- Tensorflow 1.0.1
- Scipy
- PyPNG
- (Optional : for lasso-wavelet) PyWavelets
- (Optional) CVXOPT
Pip installation can be done by $ pip install -r requirements.txt
-
Clone the repository and dependencies
$ git clone https://github.com/AshishBora/csgm.git $ cd csgm $ git submodule update --init --recursive
-
Download the datasets:
$ ./setup/download_data.sh
-
Download pretrained models or train your own!
- To download pretrained models:
$ ./setup/download_models.sh
- To train your own
- VAE on MNIST:
$ ./setup/train_mnist_vae.sh
- DCGAN on celebA, see https://github.com/carpedm20/DCGAN-tensorflow
- VAE on MNIST:
- To download pretrained models:
-
To use wavelet based estimators, you need to create the basis matrix:
$ python ./src/wavelet_basis.py
The following are the supported experiments and example commands to run the demos:
[Note: For celebA experiments, we run very few iterations per experiment (30) to give a quick demo. To get results with better quality, increase the number of iterations to at least 500 and use at least 2 random restarts.]
- Reconstruction from Gaussian measurements
$ ./quick_scripts/mnist_reconstr.sh
$ ./quick_scripts/celebA_reconstr.sh "./images/182659.jpg"
- Super-resolution
$ ./quick_scripts/mnist_superres.sh
$ ./quick_scripts/celebA_superres.sh "./images/182659.jpg"
- Reconstruction for images in the span of the generator
$ ./quick_scripts/mnist_genspan.sh
$ ./quick_scripts/celebA_genspan.sh
- Quantifying representation error
$ ./quick_scripts/mnist_projection.sh
$ ./quick_scripts/celebA_projection.sh "./images/182659.jpg"
- Inpainting
$ ./quick_scripts/mnist_inpaint.sh
$ ./quick_scripts/celebA_inpaint.sh "./images/182659.jpg"
-
Create a scripts directory
$ mkdir scripts
-
Identfy a dataset you would like to get the quantitative results on. Locate the file
./quant_scripts/{dataset}_reconstr.sh
. -
Change
BASE_SCRIPT
insrc/create_scripts.py
to be the same as given at the top of./quant_scripts/{dataset}_reconstr.sh
. -
Optionally, comment out the parts of
./quant_scripts/{dataset}_reconstr.sh
that you don't want to run. -
Run
./quant_scripts/{dataset}_reconstr.sh
. This will create a bunch of.sh
files in the./scripts/
directory, each one of them for a different parameter setting. -
Start running these scripts.
- You can run
$ ./utils/run_sequentially.sh
to run them one by one. - Alternatively use
$ ./utils/run_all_by_number.sh
to create screens and start proccessing them in parallel. [REQUIRES: gnu screen][WARNING: This may overwhelm the computer]. You can use$ ./utils/stop_all_by_number.sh
to stop the running processes, and clear up the screens started this way.
- You can run
-
Create a results directory :
$ mkdir results
. To get the plots, seesrc/metrics.ipynb
. To get matrix of images (as in the paper), run$ python src/view_estimated_{dataset}.py
. -
You can also manually access the results saved by the scripts. These can be found in appropriately named directories in
estimated/
. Directory name conventions are defined inget_checkpoint_dir()
insrc/utils.py
For a complete list of images not used while training on celebA, see here.