Henrik Skovsgaard of the ITU Gaze Group successfully defended his PhD thesis “Noise Challenges in Monomodal Gaze Interaction” at the IT University of Copenhagen on the 13th December 2011. The PhD thesis can be downloaded here.
ABSTRACT
Modern graphical user interfaces (GUIs) are designed with able-bodied users in mind. Operating these interfaces can be impossible for some users who are unable to control the conventional mouse and keyboard. An eye tracking system offers possibilities for independent use and improved quality of life via dedicated interface tools especially tailored to the users’ needs (e.g., interaction, communication, e-mailing, web browsing and entertainment). Much effort has been put towards robustness, accuracy and precision of modern eye-tracking systems and there are many available on the market. Even though gaze tracking technologies have undergone dramatic improvements over the past years, the systems are still very imprecise. This thesis deals with current challenges of monomodal gaze interaction and aims at improving access to technology and interface control for users who are limited to the eyes only. Low-cost equipment in eye tracking contributes toward improved affordability but potentially at the cost of introducing more noise in the system due to the lower quality of hardware. This implies that methods of dealing with noise and creative approaches towards getting the best out of the data stream are most wanted. The work in this thesis presents three contributions that may advance the use of low-cost monomodal gaze tracking and research in the field:
An assessment of a low-cost open-source gaze tracker and two eye tracking systems through an accuracy and precision test and a performance evaluation.
Development and evaluation of a novel innovative 3D typing system with high tolerance to noise that is based on continuous panning and zooming.
Development and evaluation of novel selection tools that compensate for noisy input during small-target selections in modern GUIs.
This thesis may be of particular interest for those working on the use of eye trackers for gaze interaction and how to deal with reduced data quality. The work in this thesis is accompanied by several software applications developed for the research projects that can be freely downloaded from the eyeInteract appstore (http://www.eyeinteract.com).
SUPERVISORS
Associate professor John Paulin Hansen - ITU Copenhagen (main supervisor)
Associate professor Dan Witzner Hansen - ITU Copenhagen (secondary supervisor)
ASSESSMENT COMMITTEE
Professor Kasper Hornbæk - University of Copenhagen, Denmark
Associate professor Scott MacKenzie - York University, Canada
Associate professor Thomas Pederson - IT University of Copenhagen, Denmark (chairman)
Well, well, look here. A constellation of eye tracking manufacturers are joining in on the affordable market, perhaps defined some time ago by Mirametrix who launched at @ $5k. Tobii have their PC Eye, perfectly fine but at a cool $7k and is showcasing the new IS2 chipset but apparently can't do CEBIT12 demos. The new player is Sensomotoric Instruments, known for their high quality hardware and finely tuned algorithms. Their new contribution is the RED-M (M is for mini?). Even if the price hasn't been announced I would assume it's less than it's high speed fire-wire sibling, perhaps similar to the PCEye pricing?
The M-version is a small device made out of plastics that connects via USB 2.0 (assuming two plugs, one for power), it measures 240x25x33mm - that's pretty small and it's only 130 grams. This is a big difference from their prior models which have been very solid and made out of high quality materials and professional components. The accuracy is specified to 0.5deg, 50-75cm distance where the box is 320x210mm @ 60cm with a sample rate of 60/120Hz, in essence it's the low end version of the RED series where the top model is the super fast RED500 . Although it has yet to be demonstrated in operational state some material has appeared online. Below is the animated setup guide, you can find more information on their website. Looking good!
Just came across the RealGaze glasses which is being developed by Devon Greco et al. He's father was diagnosed with ALS some years ago and given that Devon has been tinkering with electronics since early on he set out to build an eye tracker. For a prototype the result looks good, I guess the form factor feels familiar. There isn't too much meat available at the moment other than big ambitions to manufacturer an affordable device. Most of us would love to see that happen!
The Eyewriter project which helped Tony 'TemptOne' Quan to draw again was originally document by Mick Ebeling. This material has been incorporated into a documentary called "Getting up" and recently won the audience award at the Slamdance. Movie buff Christopher Campbell wrote a short review on his blog. Great job on raising awareness, hope you guys find funding to further develop the software.
Abstract I (ETRA)
Remote pan-and-zoom control for the exploration of large information spaces is of interest for various application areas, such as browsing through medical data in sterile environments or investigating geographic information systems on a distant display. In this context, considering a user's visual attention for pan-and-zoom operations could be of interest. In this paper, we investigate the potential of gaze-supported panning in combination with different zooming modalities: (1) a mouse scroll wheel, (2) tilting a handheld device, and (3) touch gestures on a smartphone. Thereby, it is possible to zoom in at a location a user currently looks at (i.e., gaze-directed pivot zoom). These techniques have been tested with Google Earth by ten participants in a user study. While participants were fastest with the already familiar mouse-only base condition, the user feedback indicates a particularly high potential of the gaze-supported pivot zooming in combination with a scroll wheel or touch gesture.
To be presented at the ETRA12.
Abstract II
Since eye gaze may serve as an efficient and natural input for steering in virtual 3D scenes, we investigate the design of eye gaze steering user interfaces (UIs) in this paper. We discuss design considerations and propose design alternatives based on two selected steering approaches differing in input condition (discrete vs. continuous) and velocity selection (constant vs. gradient-based). The proposed UIs have been iteratively advanced based on two user studies with twelve participants each. In particular, the combination of continuous and gradient-based input shows a high potential, because it allows for gradually changing the moving speed and direction depending on a user's point-of-regard. This has the advantage of reducing overshooting problems and dwell-time activations. We also investigate discrete constant input for which virtual buttons are toggled using gaze dwelling. As an alternative, we propose the Sticky Gaze Pointer as a more flexible way of discrete input.
To be presented at the ETRA12.
Abstract III (CHI)
While eye tracking has a high potential for fast selection tasks, it is often regarded as error-prone and unnatural, especially for gaze-only interaction. To improve on that, we propose gaze-supported interaction as a more natural and effective way combining a user's gaze with touch input from a handheld device. In particular, we contribute a set of novel and practical gaze-supported selection techniques for distant displays. Designed according to the principle gaze suggests, touch confirms they include an enhanced gaze-directed cursor, local zoom lenses and more elaborated techniques utilizing manual fine positioning of the cursor via touch. In a comprehensive user study with 24 participants, we investigated the potential of these techniques for different target sizes and distances. All novel techniques outperformed a simple gaze-directed cursor and showed individual advantages. In particular those techniques using touch for fine cursor adjustments (MAGIC touch) and for cycling through a list of possible close-to-gaze targets (MAGIC tab) demonstrated a high overall performance and usability.
A video from EyeTech that features Michael who suffers from Thoracic Outlet Syndrome (TOS). Great little clip that shows what computer control without gaze-adapted interfaces comes down to. Luckily Michael can use voice recognition software for typing, text input using eye movements alone is a cumbersome process (source).
Demonstration for eye-gaze controlled navigation. The goal was to move the the point of gaze to the center of the screen, simply put "where you look is where it goes". This is the popular AR.Drone quadricopter controlled with this policy using an EyeLink 2000 eye-tracker. The drone pivots left and right when the operator looks into those directions. It tilts up and down (making it fly backwards and forwards) when the operator looks up or down. EyeDrone was implemented by Lucas C Parra in C++ with much help from Michael Quintian. The chin rest here is not really needed as the basic control algorithm is inherently stable and miscalibrations are of little concern.
Latest update from Stephen Vickers and Howell Istance at the Centre for Computational Intelligence, De Montfort University who have been doing research and development of gaze-based gaming for several years now. Their latest project has been shortlisted in the consumer category of The Enginner Awards 2011. Congratulations on an awesome job!
"Using your eyes and where you are looking to interact with computer games represents an exciting new direction that game play can take, following the success of whole body interaction enabled by the Kinect and the Wii. The Innovation Fellowship has supported the development and demonstration of a low-cost eye tracker by De Montfort University, in collaboration with Sleepy Dog, the East Midlands games company that produced the Buzz-it controller and games. The low-cost eye tracker utilised the ITU Gazetracking library and was produced as a fully working pre-production prototype. In the project, three different games were produced to demonstrate different ways in which eye gaze can be used to make game play more immersive and exciting.
This video demostrates two of them.
eyeAsteroids
The ship flies towards where you are looking and the space bar is used to fire.
eyeShoot in the Dark!
The torch shines at where you are looking and the mouse is used to move the cross-hair and fire.
The SMI Glasses, which we got a sneak-peak at before last Christmas and specs in April are now available for sale. SMI engineers have managed to squeeze two 30Hz eye tracking cameras and a high definition scene camera into the frame weighting only 75 grams. The advantage of tracking both eyes is that it enables parallax compensation where the accuracy is maintained regardless if you look close or far in the scene. In addition, binocular tracking most often yields better accuracy as it provides the opportunity to average both samples or to discard one with low validity. The HD scene camera captures 24 frames per second at 1280x960 pixels, a clear advantage. All in all, better late than never, SMI is back with a highly competitive product in the market.
