README.md
In d-bohn/quantIm: Quantitative Assesment of Facial Images

quantIm

An R package for the quantitative assessment of images.

DISCLAIMER ~~This package is currently in pre-pre-beta. It is a host of personal functions I wrote for specific projects that I find myself reusing over and over again. It is very much a work in progress. I have only tested it on UNIX systems.~~

This package has been superseded by the package faceplyr.

In order to use this package, a valid installation of the package manager software conda must be installed. quantIm sets up a specific enviornment for the package (much like RStudio's tensorflow, and will install all of the necessary packages for the various functions (mostly python packages).

install.packages('devtools')
devtools::install_github('d-bohn/quantIm')

library(quantIm)
install_quantIm()

The motivation to compile this package are entirely selfish. The R ecosystem has a number of excellent image manipulation packages (e.g., imager, EBImage and magick), however, they all do different things. This leaves the user to navigate different packages for specific tasks, running the risk of incompatible data structures and non-reproducibility. I wanted all of the functions I use in a single space for easier data analysis traversing.

quantIm attempts to provide convenient wrappers around some common functions data scientists and computer vision researchers might utilize often when manipulating/assessing images. Additionally, with the implementation of RStudio's wonderful reticulate package, quantIm attempts to provide wrappers around some common computer vision and image functions written in python.

The functions in quantIm can largely be broken into the following categories:

Image manipulation
1. Padding images
2. Subtracting two images
3. "gleam"ing and image
Image analysis
1. t/z-test between images
2. Plot heat map of values
3. Distance metrics between images