EyeTrackingUtils.jl

Robust package for preparing and analysing eye tracking data
Author jakewilliami
Popularity
3 Stars
Updated Last
1 Year Ago
Started In
September 2020

Eye Tracking Utils

Code Style: Blue Build Status Project Status

Note

THIS PACKAGE IS IN PRE-ALPHA AND IS UNDER DEVELOPMENT! IT IS NOT READY FOR USE

Though we have high hopes for this package, it is currently not finished being developed at this point in time. This project is currently worked on by Alexandros Tantos, Jake Ireland, and others.

Introduction

This is a Julia implementation of a robust data-preparation and -analysis package using data from eye tracking experiments. This package is designed modelling an R package called eyetrackingR. This package is not to be confused with the "EyeTracking.jl" package (at time of writing; October, 2020), which seems to be a GUI-style experimental interface (at first glance, perhaps similar to PyGaze).

Installation and Examples

julia> using EyeTrackingUtils

How it works

This package has two main steps in the workflow, whose second step has three paths. This workflow is drawn from eyetrackingR.

  1. Data cleaning — obtain data and information about data from various eye-tracking sources (this step should be extremely robust), and puts data into a standardised format for analyses.
  2. Analyses —:

a) Overall Looking;

b) Onset-Contingent;

c) Time-Course of Looking.

Obtaining Test Data

# PATH_TO_EYETRACKING="/some/path" # PLEASE FILL IN THIS LINE
$ cd /tmp/
$ wget https://raw.githubusercontent.com/jwdink/eyetrackingR/master/data/word_recognition.rda
$ Rscript -e 'load("word_recognition.rda"); write.csv(word_recognition, "word_recognition.csv")'
$ mv word_recognition.csv "$PATH_TO_EYETRACKING"/data/
$ curl https://raw.githubusercontent.com/jwdink/eyetrackingR/master/tests/testthat/tb_output_between_subj.txt > "$PATH_TO_EYETRACKING"/data/tb_output_between_subj.txt
$ curl https://raw.githubusercontent.com/jwdink/eyetrackingR/master/tests/testthat/tb_output_between_subj.txt > "$PATH_TO_EYETRACKING"/data/tb_output_between_subj.txt
$ curl https://raw.githubusercontent.com/jwdink/eyetrackingR/master/tests/testthat/tb_output_interaction.txt > "$PATH_TO_EYETRACKING"/data/tb_output_interaction.txt
$ curl https://raw.githubusercontent.com/jwdink/eyetrackingR/master/tests/testthat/tb_output_within_subj.txt > "$PATH_TO_EYETRACKING"/data/tb_output_within_subj.txt
$ curl https://raw.githubusercontent.com/jwdink/eyetrackingR/master/tests/testthat/tclust_tb_anal.txt > "$PATH_TO_EYETRACKING"/data/tclust_tb_anal.txt

Timeline of Progression

To Do

  • Summarise (Query.jl?)
  • Read from EDF file.
    • Is this needed? As long as we support dataframes, the rest is up to other pacakges.

A Note on running on BSD:

The default JuliaPlots backend GR does not provide binaries for FreeBSD. Here's how you can build it from source..

Used By Packages

No packages found.