Pipeline for Processing Eye-tracking Data

Overview

(click on the image for better quality)

Notebooks and Procedures

1. Gaze Projections and AOIs Mapping:

   • Gaze Data import and computation of X and Y gaze coordinates
   • Data exploration and visual quality assessment (projecting gaze points on AOIs)
   • Assign gaze points to AOIs

2. Event Detection

   • Oculomotor event detection (i.e., Fixations and saccades)

3. Fixation, Scan-path, Heatmap plots

   • Fixations plots
   • Scan-path plots
   • Heatmap plots

4. Mapping Fixations and Saccades to AOIs

   • Mapping Fixations and Saccades to AOIs

5. AOI Visits

   • Identify Dwells

6. AOI Sequence Charts

7. Fixation, Saccade, Dwell Measures (See [documentation/ETmetrics.csv] (https://github.com/aminobest/ETLecture/blob/main/jupyter/workspace/documentation/ETmetrics.csv) for metrics and definions)

   • Fixation measures at stimulus and AOI levels
   • Saccade measures at stimulus and AOI levels
   • Dwell measures

8. Transition Matrix and Markov Model

   • Identify transitions (including self-transitions e.g., transitions from AOI1 to AOI1)
   • Identify transitions (with no self-transitions)

Installation

Prerequisites

1. Docker https://docs.docker.com/get-docker/
2. Git https://git-scm.com/downloads

Procedure

git clone https://github.com/aminobest/ETLecture.git

cd ETLecture
docker compose up --build

The build process can take some minutes, once it's over, a link to the jupyter lab web-application (starting with https://127.0.0.1:8888/lab?token=) will be showned in the terminal (cf. screnshot below)

Copy this link and paste it in your browser

Use

Browse to "ETLecture" and start the docker instance using the following command

docker compose up

A new link to the jupyter lab web-application (starting with https://127.0.0.1:8888/lab?token=) will be showned in the terminal (similar to the screenshot above)

Update notebooks

Browse to "ETLecture" and run the following

git pull

docker compose up --build