| |
New Smartphone Eyetracking System Developed
A team of researchers at MIT and the University of Georgia have developed software which will allow ordinary smartphones to be used for eyetracking research
Graduate student Aditya Khosla and his colleagues taught their app using machine learning based on gaze patterns from 1,500 mobile device users - according to www.siliconindia.com .
The researchers used Amazon's Mechanical Turk crowdsourcing site to recruit subjects and paid them a small fee for each successful input, producing a data set with an average of 1,600 images for each user, and improving the error rate twofold by comparison with other experimental systems.
Khosla says the team were looking to break a chicken-and-egg style problem with the development of eye tracking: 'Since there are no applications, there's no incentive for people to buy the devices. We thought we should break this circle and try to make an eye tracker that works on a single mobile device, using just your front-facing camera'.
The new system is described in a paper due for presentation at the set to presented at the 'Computer Vision and Pattern Recognition' conference in Las Vegas next week.
|
