Eye Tracking Interaction on Unmodified Mobile VR Headsets Using the Selfie Camera

Input methods for interaction in smartphone-based virtual and mixed reality (VR/MR) are currently based on uncomfortable head tracking controlling a pointer the screen. User fixations fast natural input method VR/MR interaction. Previously, eye mobile VR suffered from low accuracy, long processing time, need hardware add-ons such as anti-reflective lens coating infrared emitters. We present an innovative methodology utilizing only images front-facing (selfie) camera through headset’s lens, without any modifications. Our system first enhances low-contrast, poorly lit by applying pipeline of customised low-level image enhancements suppressing obtrusive reflections. then propose iris region-of-interest detection algorithm that is run once. This increases speed reducing search space devices. iteratively fit geometric model to refine its coordinates. display thin bezel light at top edge screen constant illumination. A confidence metric calculates probability successful detection. Calibration linear gaze mapping between estimated centroid physical pixels results latency, real-time tracking. formal study confirmed our system’s accuracy similar trackers commercial headsets central part field-of-view. In game, gaze-driven user completion time was with head-tracked interaction, consecutive motions. panorama viewer, users could successfully switch panoramas using gaze.