Eye Tribe Eye Tracker Dev Kit, $99; Open Source ITU Gaze Tracker Grows Up!

The Eye Tribe Eye Tracker developer kit is available for pre-order for $99.00. The kit comes with an SDK with C++, C#, and Java, full source code included.  

I've been waiting for a while to see this happen! 

The Eye Tribe Eye Tracker is an outgrowth of the work of a group of researchers at the IT University of Copenhagen.  At the time, it was known as the open-source ITU Gaze Tracker. 
I came across it a few years ago in a NUI-Group forum, and later wrote a post about it when the 2.0 version was released. 

Although the Eye Tribe Tracker was originally developed to meet the needs of people with disabilities who could not access computers, it was found to have potential for a number of other uses that were not really possible before the spread of mobile technologies such as touch-screen tablets and smart phones. 

To get a better understanding of eye-gaze/tracking technology, take a look at the following videos and follow the related links.



Below is a demonstration of the gaze UI on an Android smartphone:


Here is another look at this technology running on a Windows 8 Tablet:





RELATED
The Eye Tribe (website)
Eye Tribe starts taking pre-orders for $99 Windows eye tracker
Senseye will let you control your mobile phone with your eyes
Martin Bryant, The Next Web, 12/2/11
Open-source Eye-tracking: The ITU Gaze Tracker 2.0 Beta Via Martin Tull, NUI-Group Member
Lynn Marentette, Interactive Multimedia Technology, 11/1/10
ITU GazeGroup
Gaze Tracker Development
GazeGroup Forum
Martin Tall


RELATED VIDEOS
Eye Tribe was formally known as Senseye. Below is an earlier video that shows how it worked with a web-cam on a mobile device:



Open-Source ITU Gaze Tracker

ITU Gaze Tracker from ITUcph on Vimeo.


Earlier Videos of the ITU Gaze Tracker:
Technical Demonstration 




Seeking Sustainable Innovation

EyeTube for YouTube! Eye-gaze interaction software, free and downloadable from GazeGroup

Gaze interaction systems provide access to computers and the rich content now available on the web for many people with disabilities.  Unfortunately, commercial gaze tracking systems are very expensive and at times, difficult to calibrate.  There is hope!


Following up on my recent post, "Open-source Eye-tracking: The ITU Gaze Tracker 2.0 Beta", I thought I'd share the GazeGroup's EyeTube for YouTube interface.  


What is great about EyeTube for YouTube is that it provides two different interfaces. The simplified version looks good for younger children or people with cognitive disorders, and is icon-based.  The second version is appropriate for people who can navigate through more complex visual representations of content. 


EyeTube requires a Windows-based system and .Net 3.5 at this time. It can be downloaded from the GazeGroup website.  If you plan to download the application, you must also make sure you have a YouTube account. To get the application up and running, you'll need to change the settings (EyeTubeSettings.xml) to match your account.   (If you don't know much about changing settings or xml, ask someone you know who works in IT.)


Below is the icon-based version of the eye-gaze interface for YouTube:



Feature-rich version of the EyeTube interface for YouTube:


From the GazeGroup site:

"The EyeTube prototype offers a feature rich eye controlled interface for the popular YouTube service. Instead of emulating a mouse pointer and interacting with a web browser the EyeTube interface is especially designed to be driven by gaze input. It offers a wide range of features such as keyword searching, popular video feeds, favorites and social aspects such as subscriptions, friends and commenting on videos.The highly optimized interfaces allows for a streamline interaction which is aleviated from the Midas Touch problem. In most previous gaze interfaces selection is made by a dwell time activator, e.g fixat a button for a specific amount of time and it will execute the function. In the EyeTube interface a fixation on a U.I element will highlight it and a second fixation on the activation button is required to execute the function. This removes the stress of having to constantly move the eyes to avoid unintentional activation."

"The EyeTube also exists in another simplified incarnation developed for users whom are distracted by a larger number of options. It supports basic features such as browsing categories, optional keyword searching and favorites."

RELATED
The GazeGroup
(The individuals mentioned below may be currently working elsewhere, but involved in the gaze research in some way.)

• John Paulin Hansen, Head of group (Ph.D. in Psychology)
• Dan Witzner Hansen, Associate professor
• Javier San Agustin, PhD student
• Sune Alstrup Johansen, PhD student
• Henrik Hegner Tomra Skovsgaard, PhD student
• Martin Tall, PhD student (Now working at Duke University, North Carolina)

GazeGroup Research Areas

Development of Gaze Tracking Technology

Application of Eye Tracking for User Experience studies and HCI

Gaze based Interaction

COGAIN (Communication by Gaze Interaction)

ACM CHI Conference Articles
San Agustin, J., Skovsgaard, H., Hansen, J. P., and Hansen, D. W. 2009. Low-cost gaze interaction: ready to deliver the promises. In Proceedings of the 27th international Conference Extended Abstracts on Human Factors in Computing Systems (Boston, MA, USA, April 04 - 09, 2009). CHI EA '09. ACM, New York, NY, 4453-4458. DOI= http://doi.acm.org/10.1145/1520340.1520682
San Agustin, J., Hansen, J. P., Hansen, D. W., and Skovsgaard, H. 2009. Low-cost gaze pointing and EMG clicking. In Proceedings of the 27th international Conference Extended Abstracts on Human Factors in Computing Systems (Boston, MA, USA, April 04 - 09, 2009). CHI EA '09. ACM, New York, NY, 3247-3252. DOI= http://doi.acm.org/10.1145/1520340.1520466 
Tall, M., Alapetite, A., San Agustin, J., Skovsgaard, H. H., Hansen, J. P., Hansen, D. W., and Møllenbach, E. 2009. Gaze-controlled driving. InProceedings of the 27th international Conference Extended Abstracts on Human Factors in Computing Systems (Boston, MA, USA, April 04 - 09, 2009). CHI EA '09. ACM, New York, NY, 4387-4392. DOI= http://doi.acm.org/10.1145/1520340.1520671

UPDATE

Eye-controlled games and leisure applications from the COGAIN wiki: http://www.cogain.org/wiki/Leisure_Applications

• EyeArt - EyeArt eye-drawing program, developed by Andre Meyer and Markus Dittmar, Technical University of Dresden, Applied Cognitive Research Unit, Germany.
• GazeTrain - Gaze-controlled action oriented puzzle game, developed by Lasse Farnung Laursen, Technical University of Denmark
• Puzzle - Simple puzzle game that can be played with eye movements, developed by Vytautas Vysniauskas, Siauliai University, Lithuania
• Road to Santiago - Gaze-controlled adventure game (full game), developed by Javier Hernandez Sanchiz, Universidad Publica de Navarra, Spain
• Snap Clutch - An application that uses eye gaze data to generate key and mouse events for playing games such as World of Warcraft and Second Life.
• ASE: Accessible Surfing Extension for Firefox - Follow this link to access ASE, an Accessible Surfing Extension for Firefox, developed by Emiliano Castellina and Fulvio Corno at Politecnico di Torino. (Note that this is a beta version.)
• Eye Gaze Music (SAW Selection Sets) - Point and Play – eye gaze (direct pointing) musical activities, developed by DART. Please note that SAW (Special Access to Windows) framework application is needed to play these 15 music selection sets. SAW is available for free athttp://www.oatsoft.org/Software/SpecialAccessToWindows
• EyeTube - Gaze interaction for YouTube - Follow this link to get more information and download EyeTube at ITU GazeGroup's web pages
• Eye3D and other head eye mouse software - Eye3D for education, and a collection of links to free software that works with head or eye mouse. Includes links to downloads and original sites.
• Gaze-controlled Breakout - Follow this link to access a modified version of the LBreakout2 game which can be operated by an SMI eye tracker, developed by Michael Dorr et al. at University of Luebeck
• Oleg Spakov's Freeware games for MyTobii - Follow this link to access MyTobii compatible games developed by Oleg Spakov, University of Tampere, Finland
• Free ITU Gaze Tracker and applications - Download a webcam based open-source gaze tracker and several applications that work with it, developed at IT University of Copenhagen
• GameBase - Check out the Eye-Gaze Games category at the SpecialEffect GameBase!
• More information about Gaze-Controlled Games - Follow this link to see a list of online information resources on using gaze for the control of games and other leisure applications

Open-source Eye-tracking: The ITU Gaze Tracker 2.0 Beta Via Martin Tall, NUI-Group Member

I came across the first version of the open-source ITU Gaze Tracker on the NUI Group forum in April of 2009 and played around with it a bit.  I was impressed.  I'm happy to say that the new version looks even better, although I haven't had the time to try it out.  Below are two recent videos that will give you a better understanding about gaze tracking.  


For the tech-curious, make sure you take the time to view the second video!  Links to info & code are below.


GT2 High speed remote eye tracking "Pushing the limits"


Technical Demonstration


Info about  the ITU Gaze Tracker 2.0 Beta from the NUI Group Forum, posted by Martin Tall:

Introducing the ITU Gaze Tracker 2.0 Beta

"We’ve made great progress since the initial release, today we open the doors for version 2.0. Internally we’ve rewritten major parts of the platform to gain flexibility and higher performance.  First version was DIY playtime, this version is nothing short of a screamer. High performance, very accuracy tracking. People are telling us we are crazy giving it away but we’re dedicated to the mission: Accessible eye tracking for all, regardless of nationality and means. We’re making it happen."

Important highlights for GT2.0b:
- Supports three modes of operation, head-mounted, remote mono/binocular
- Vastly improved performance, +500fps head mounted, +170fps remote binocular (both eyes)
- Awesome accuracy, avg. 0.3 - 0.7 degrees of visual angle (remote binocular)
- New U.I, looks so.. 2010
- Automatic tuning (optimization of algorithms parameters)
- Relatively low CPU-utilization and memory footprint (12%, 170Mb, core i7 860 win7-64)
- Many enhancements, bug-fixes etc.

More info:
http://forum.gazegroup.org
http://develop.gazegroup.org