Eye tracking is a truly great innovation utilized in lots of fields ranging from linguistics, human-computer interaction, as well as marketing. With a appropriate eye tracking setup, it’s possible for a web designer to see if their modifications to the design are effective, to determine exactly how quick somebody reads a page of text, as well as even diagnose medical disorders. Eye tracking configurations haven’t been cheap, though, at least up until now. pupil is a serious, research-quality eye tracking headset developed by [Moritz] as well as [William] for their thesis at MIT.
The fundamental concept behind pupil is to put one digital video camera dealing with the user’s eye while one more video camera looks out on the world. After calibrating the included software, the headset looks at the user’s pupil to identify where they’re really looking.
The hardware isn’t specialized whatsoever – just a pair of $20 USB webcams, a LED, an infrared filter made of subjected 35mm film negatives, as well as a 3D printed headset conveniently for sale at Shapeways.
The software application for pupil is based on OpenCV as well as OpenGL as well as is offered for Mac as well as Linux. Calibration is easy, as seen in the videos after the break, as well as the results are fantastic for an eye tracking headset thrown together for under $100.