The way to read the title of this project is, "A function that maps the set of two-dimensional real numbers to the set of three-dimensional real numbers." You see this kind of mathematical shorthand used to provide a generic definition of an image. It makes sense; an image is either a 2D scalar field (monochrome images) or a mapping from a 2D space to a 3D space (colour images).

The fr2r3 application was designed to generate a small set of visualizations that demonstrates how the exact same image can be viewed in completely different ways. Put a little bit more formally, it's possible to define a set of operators that map the image from one high-dimensional space to another high-dimensional space. The specifics depend on the transform, but they are chosen to pull out something interesting, or useful, about an image.

You can find out more details about the algorithms here: An Example

The supported method for installing fr2r3 is to use conda. First, make sure that you have conda installed locally (development was done with miniconda but the full Anaconda distribution will work as well). Next, check out the repository from GitHub,

$ git clone https://github.com/richengguy/fr2r3.git

Setting up the environment to use fr2r3 is then simply

$ conda env create
$ conda activate fr2r3

The main entrypoint is the fr2r3 command. Use --help to see what subcommands are available and how to use them.

Usage: fr2r3 [OPTIONS] COMMAND [ARGS]...

  Apply a transform operator onto an image.

  The fr2r3 application will apply a transformation (e.g. Fourier transform)
  onto the provided image.  The transforms are meant for visualization
  purposes and have been modified to be visually pleasing.

Options:
  --help  Show this message and exit.

Commands:
  filter      Apply the transform on to the image.
  ls          List the available transforms.
  synthesize  Generate a synthetic image.

All source code is licensed under a BSD 3-Clause Licence. The contents of the docs folder are licensed under a CC-BY-SA-4.0 licence.