Servo Control Based on Pupil Detection Eye Tracking
Constantin Barabasa, Radu Gabriel Bozomitu, Cristian Rotariu · 2018
The proposed article presents a proof of concept developed for the evaluation of eye tracking algorithms based on pupil detection, built around a servo-based pan/tilt platform controlled by pupil movement. An image of the eye is captured using a head mounted infrared video camera, and after the pupil center position is detected, its coordinates are mapped to control the pan and tilt of a servo motor-controlled miniature platform. The system allows efficient real-world evaluation of different eye tracking algorithms and it can have a wide range of application scenarios depending on the type of accessory mounted on the platform.