/NSRR-PyTorch

PyTorch version of the paper 'Neural Supersampling for Real-time Rendering' by Facebook Reality Labs (2020)

Primary LanguagePythonMIT LicenseMIT

image-interpolation

Deep learning project to create new images based on the previous ones

Usage

Dataset

In order to be loaded using NSRRDataLoader, the dataset should be structured like so:

[root_dir]
│
└───View
│   │   img_1.png
│   │   img_2.png
│    ...
│   
└───Depth
│   │   img_1.png
│   │   img_2.png
│    ...
│   
└───Motion
│   │   img_1.png
│   │   img_2.png
│    ...

Where root_dir can be given as an argument, and View, Depth and Motion are static members of NSRRDataLoader.

Note that corresponding tuples of (view, depth, motion) images files should share the same name, as they cannot be grouped together otherwise.

Unit testing

python3 unit_test.py --directory [path_to_root_dir] --filename [image_name]

Miscellaneous information

Using :