Code accompanying CVPR'20 paper of the same title. Paper link: https://arxiv.org/abs/2003.03808
We have noticed a lot of concern that PULSE will be used to identify individuals whose faces have been blurred out. We want to emphasize that this is impossible - PULSE makes imaginary faces of people who do not exist, which should not be confused for real people. It will not help identify or reconstruct the original image.
We also want to address concerns of bias in PULSE's outputs. It does appear that PULSE is producing white faces much more frequently than faces of people of color. This bias is likely inherited from the dataset StyleGAN was trained on (see Salminen et al., 2020), though there could be other factors that we are unaware of. We recognize that bias like this is a critical issue in the fields of machine learning and computer vision. We’ve reached out to the original creator of StyleGAN and FFHQ, NVIDIA, about this issue. Our hope is that this will lead to the development of methods that don’t display such behavior. We will also be including a new section in our paper directly addressing this bias in more detail.
- PULSE: Self-Supervised Photo Upsampling via Latent Space Exploration of Generative Models
- Table of Contents
Given a low-resolution input image, PULSE searches the outputs of a generative model (here, StyleGAN) for high-resolution images that are perceptually realistic and downscale correctly.
The main file of interest for applying PULSE is run.py. A full list of arguments with descriptions can be found in that file; here we describe those relevant to getting started.
You will need to install cmake first (required for dlib, which is used for face alignment). Currently the code only works with CUDA installed (and therefore requires an appropriate GPU) and has been tested on Linux and Windows. For the full set of required Python packages, create a Conda environment from the provided YAML, e.g.
conda create -f pulse.yml
or (Anaconda on Windows):
conda env create -n pulse -f pulse.yml
conda activate pulse
In some environments (e.g. on Windows), you may have to edit the pulse.yml to remove the version specific hash on each dependency and remove any dependency that still throws an error after running conda env create... (such as readline)
dependencies
- blas=1.0=mkl
...
to
dependencies
- blas=1.0
...
Finally, you will need an internet connection the first time you run the code as it will automatically download the relevant pretrained model from Google Drive (if it has already been downloaded, it will use the local copy). In the event that the public Google Drive is out of capacity, add the files to your own Google Drive instead; get the share URL and replace the ID in the https://drive.google.com/uc?=ID links in align_face.py and PULSE.py with the new file ids from the share URL given by your own Drive file.
By default, input data for run.py should be placed in ./input/ (though this can be modified). However, this assumes faces have already been aligned and downscaled. If you have data that is not already in this form, place it in realpics and run align_face.py which will automatically do this for you. (Again, all directories can be changed by command line arguments if more convenient.) You will at this stage pic a downscaling factor.
Note that if your data begins at a low resolution already, downscaling it further will retain very little information. In this case, you may wish to bicubically upsample (usually, to 1024x1024) and allow align_face.py to downscale for you.
The dataset we evaluated on was CelebA-HQ, but in our experience PULSE works with any picture of a realistic face.
Once your data is appropriately formatted, all you need to do is
python run.py
Enjoy!