Efficient Biologically Inspired Eye Tracking for Mixed Reality
- Benjamin Lundell ,
- Christopher Mei ,
- Paulo R. S. Mendonça
Eye tracking is a key component of mixed-reality headsets. Knowledge of eye pose allows for accurate rendering and greater user comfort, provides insights into user intent, and facilitates novel experiences. The challenges of building an eye tracker that works at scale comes from the difficulty in ensuring reliability for a large, diverse population. To this end, we developed a novel, easily interpretable and biologically realistic eye model, which is at the core of the eye tracking solution on Microsoft HoloLens 2.