Adaptive calibration method based on state space model for eye gaze HCI system

Abstract

Gaze-tracking is a direct and natural human-computer interaction(HCI) channel, which has a broad application prospect in intelligent interaction field. To improve the adaptability of Gaze-tracking HCI system, a method of adaptive gaze-tracking human-computer interaction based on the state space is proposed. The analysis and modeling of gaze tracking human-computer interaction system from the perspective of state space transformation has been done to research on the system working mechanism that the gaze resolution parameters can be corrected adaptively in the case of system state changes which includes the user's head movement, camera and interactive interface. The experimental system is established to study, and the results show that the proposed method has a good adaptability, which improves the problem of frequent system calibration without interruption during the HCI process.