Real-time eye movement analysis on a tablet computer
Vision is the primary sense of humans and the analysis of eye movements reveals much about cognitive processes. Eye tracking has provided insights into the perception of art, the viewing behavior of novice and expert athletes and the ability to drive in the elderly. In such cases, it is impertinent to record the participants’ behavior under natural conditions outside of laboratory settings. Therefore, mobile eye tracking devices which also record the participants view through a head-mounted scene camera are essential. Together with the Department of Neurology of the University Hospital Zurich, the IIS is developing such a mobile eye tracker which can be controlled with a table computer running a custom, android-based app from the IIS.
Your task consists of two steps: First, the existing routine for online calibration, e.g. the mapping between recorded eye movement signal (measured in millivolts) and eye position (measured in degrees), has to be further improved. This includes implementing and testing new methods (e.g. polynomial regression or Gaussian Process modelling) to increase accuracy of the calibrated signal as well as optimizing the interplay between the stimulus display and the app. When this is done, real-time methods for the detection of fast eye movements and fixations should be developed and ultimately implemented.
- 60% Programming (Java/Android/Matlab)
- 30% Image and Video Processing
- 10% Hardware modification
- Good programming skills (ideally Java/Android)
- Basic knowledge of image processing and GUI design