Question icon
Your Current Search
Choose below to refine your search
Research Topic
Download abstract book

Download the NI2012 abstract book here. The page numbers in the index are clickable for easy browsing.


Low-cost eyetrackers as assistive devices


Dawid Laszuk (University of Warsaw, Faculty of Physics, ul. Hoża 69, 00-681 Warszawa, Poland), Tomasz Spustek (University of Warsaw, Faculty of Physics, ul. Hoża 69, 00-681 Warszawa, Poland), Jarosław Rybusiński (University of Warsaw, Faculty of Physics, ul. Hoża 69, 00-681 Warszawa, Poland), Piotr Durka (University of Warsaw, Faculty of Physics, ul. Hoża 69, 00-681 Warszawa, Poland)

There is a huge gap between the possibilities offered by the cutting edge technologies and those available to the most of the target users; quoting William Gibson „the future is already here — it's just not very evenly distributed”. Advanced commercial eyetrackers, designed primarily for neuromarketing purposes, provide extended capabilities of compensating head movements, and can be effectively used for restoring communication in many cases of locked-in states -- but their price too often exceeds the available resources.
Decreasing cost of hardware, especially digital cameras, allowed for constructing simple eyetrackers from off-the-shelve parts -- like e.g. The EyeWriter Project -- with the aim of providing gaze-based communication. However, these simple, low-cost solutions are still sensitive to head movements. This may be less of a problem in late stages of neurodegenerative diseases like ALS, but other diseases leading to partly locked-in states are often accompanied by strong spasticity. Taking into account minimum requirements for an effective communication built upon a low-cost eyetracker, we propose an algorithm for detecting only large and quick, purposefully generated, eye movents in four directions (left/righ/up/down). Proper adjustment of parameters and thresholds allows for a basic yet robust communication in a cursor-like mode, for an overall cost over two orders of magnitude lower than the price of commercial systems.
The algorithm is implemented in the OpenBCI framework, which gives a direct access to all the features being implemented for the BCI systems, like the speller and speech synthesis. Such unified approach simplifies also user's transition to other modalities like BCI or switch-based interfaces.
Development of relevant signal processing and hardware solutions lies within the area of expertise of the graduates of the Warsaw Neuroinformatics course (, who will present their contributions to Open Hardware and Open Software.
Low-cost eyetrackers as assistive devices
Spelling a word by an eye tracker.
Preferred presentation format: Demo
Why demo: This demo has been presented many times - it's fairly popular among the audiance. Visitors can test system themselves by, for example, spelling sentences with a movement of their eyes.
Topic: Brain machine interface

Andrew Davison
Andrew Davison says:
May 11, 2012 01:34 PM
Seems more like engineering than science, but looks as though it would make a good demo.