wiki:ProjectWebGazerjs

Version 5 (modified by Monika Rizova, 7 years ago) ( diff )

--

WebGazer.js - Eye-tracking library for common webcams

Description

WebGazer.js an online eye tracker that uses common webcams already present in laptops and mobile devices to infer the eye-gaze locations of web visitors on a page in real time. With this it e a natural experience to

everyday users that is not restricted to laboratories and highly controlled user studies. WebGazer.js has two key components: a pupil detector that can be combined with any eye detection library, and a gaze estimator using regression analysis informed by user interactions.

Advantages of WebGazer.js:

  • Real time gaze prediction on most major browsers

  • Self-calibration from clicks and cursor movements
  • Swappable components for eye detection
  • Multiple gaze prediction models

URL:

  1. https://webgazer.cs.brown.edu/#home
  2. http://jeffhuang.com/Final_WebGazer_IJCAI16.pdf
  3. https://github.com/brownhci/WebGazer

Prerequisites for using this software - Software/Hardware : No special hardware - WebGazer.js uses common webcams

Project Anatomy

Community: Alexandra Papoutsaki, James Laskey, Aaron Gokaslan, Yuze He, Jeff Huang.

Leadership: Dr. Gerald Weber and his team Dr. Clemens Zeidler and Kai-Cheung Leung

Forking: Fork your own copy at this address: https://github.com/brownhci/WebGazer, for which you will need a GitHub account.

Communication: There is info for contact on the personal profiles of two members, Alexandra Papoutsaki: http://www.cs.pomona.edu/~apapoutsaki/ and Jeff Huang: http://jeffhuang.com/.

Roadmaps:

Original goal:

It was designed as a new approach to browser-based eye tracking for common webcams.

Future goals:

Short term:

WebGazer aims to overcome the accuracy problems that webcams typically face, by adopting user interactions to continuously self-calibrate during regular web browsing.

Source: http://jeffhuang.com/Final_WebGazer_IJCAI16.pdf

Releases:

  1. WebGazer: Scalable Webcam Eye Tracking Using User Interactions - Alexandra Papoutsaki, Nediyana Daskalov, Patsorn Sangkloy, Jeff Huang, James Laskey, James Hays - July 2016 Conference: 25th International Joint Conference on Artificial Intelligence (IJCAI 2016)At: New York City, New York

Repositories:

The main repository of PyGaze is: https://github.com/esdalmaijer/PyGaze , where you can download the source code, clone it to desktop, or even make your own fork. There is also a repository for PyGazeAnalizer available at: https://github.com/esdalmaijer/PyGazeAnalyser

Packaging:

Different packages for PyGaze, including Windows, Linux and Mac OS X packages can be found at the downloads page of PyGaze: http://www.pygaze.org/downloads/

Upstream/downstream:

So far, from 2013 when it was originaly published, PyGaze has 12 contibutors, and it is open for contributing for upstream, after of course your pull request is revised and approved.

Version Control:

version 0.6.0 (07-09-2017) - Latest version control

Trackers:

You can see commits and verified changes at this link: https://github.com/esdalmaijer/PyGaze/commits/master

Project Evaluation

Fieldtrips

Github: https://github.com/esdalmaijer/PyGaze

Openhub: / (PyGaze is not listed here)

Source Forge: / (PyGaze is not listed here)

Evaluation

Licensing: GPLv3

Language: Javascript

Activity: Active

Number of contributors: There are two developers and a couple of official contributors. They are listed on this page: http://www.pygaze.org/contributors/

Size: There aren't any official releases on GitHub, you can find out the size of the packages from the download page after you download them. http://www.pygaze.org/downloads/

Issue tracker: There is a forum where you can write about an issue: http://forum.cogsci.nl/index.php?p=/categories/pygaze and there is the issue tracker on GitHub: https://github.com/esdalmaijer/PyGaze/issues

New contributor: If you want to be a contributor to PyGaze one way is through the GitHub page, where you can make your own fork and pull request and wait for an approval.

Community norms: You can report an issue through the forum page , help to fix it by forking in the GitHub repository and commit fixes and if you prefer to work via different channels (contact form, or via carrier pigeon), that’s fine too.

User base: Their user base are the developers and the contributors: http://www.pygaze.org/contributors/

...Review

Note: See TracWiki for help on using the wiki.