Eye tracking hardware in use (via College of London & Institute of Physics)
People suffering from multiple sclerosis, Parkinson’s, muscular dystrophy, spinal cord injuries, amputees, etc could benefit immensely from modern advancements of computing power and robots. Their disabilities stand in the way of proper interaction or manipulation of most interfaces. Options have been explored that would allow people with disabilities to control computers using their eyes, but these can be highly invasive and can cost upwards of $20,000.
Researchers, AA Faisal and WW Abbott, from the Imperial College of London, have devised a system that uses cheap and readily available gaming cameras to track pupil movement and allow disabled patients to manipulate a cursor using their eyes. As mentioned, these systems have been made before, but this system costs less than $30, and performs at comparable levels of accuracy, which makes it unprecedented and extraordinarily exciting.
Using the video game cameras, at $10 per camera (it is not stated which specific cameras), the Imperial College team has been able to come up with a system that yields an information throughput of 43 bits/sec with 10 ms of latency. They have been able to generate more information for a richer signal by tracking gaze position in 3D rather than 2D. This means that the system is aware of the depth perception as well as where the pupils are pointing. The cheap materials provide a sampling rate of 120 Hz with a resolution in the range of 0.5-1 degree for the visual angle. The whole device uses just one watt of power.
The cameras are mounted on a pair of glasses outside the user’s line of vision. The cameras then take pictures of the eyes and relay the information via USB or Wi-Fi to any ordinary Windows or Linux system.
After a few minutes of calibration and use, six test patients were able to use the system to play pong at a 20% success rate of an able-bodied user. The Imperial college team was also able to solve the visual clicking problem by using single-eyed winks so involuntary blinking does not interfere.
The team would like to make use of the systems 3D depth perception awareness to create innovative graphical user interfaces and physical applications, like controlling a wheel chair or robotic arm, by tracking the use’s gaze.
There is no doubt that with cheaper tools and innovative software, people with motor disabilities will soon be able to use a computer in practically the same way as anyone else.
Cabe