Rana Mansour

In our daily tasks we rely on computers or smartphones to browse internet or type email etc. and typing is the mostly used human computer interaction people use. And some people with limbs movement disabilities does rely on others to have these tasks done mostly. And with eye tracking emerging technology, it has been generally a quite beneficial addition to human computer interaction. It does help users generally for a fast and relatively accurate interactions. It is particularly beneficial for large screens and virtual environments and as eye trackers technology matures it has become more practical for particular uses. And as well one of the reasons that eye tracking being researched so heavily is to help people with disabilities to be more independent and have more ease of use for the technology. In this research lab we build a web based application that is fully functional with eye based gesture. And out of many methods for eye typing out there, three prominent eye typing methods were particularly picked and implemented for this research lab which are (dwell, gesture and swipe) keyboards based. The implemented web based application is scalable which can add many other eye typing methods and provide the similar environment to test all keyboards. Each eye typing method is implemented as a separate keyboard were the user can pick from. The experiment was set it up to have each user evaluate the three different keyboards based on “Latin square” method which make sure that during evaluation the order of the keyboards is different for each participant. Ten participants in total participated in the experiment each participant had for each keyboard one training session and three evaluation sessions where each session has five sentences. And accuracy and speed for each session was recorded and stored in the web based application database. Surprisingly although dwell time keyboard takes time to select each key, results has shown it maintained better accuracy and speed compared to gesture and swipe keyboards.

17.12.20 - 10:15
via Big Blue Button