A free geometry model-independent neural eye-gaze tracking system

Massimo Gneo,Tommaso D’Alessio,Silvia Conforto,Maurizio Schmid

doi:10.1186/1743-0003-9-82

Massimo Gneo, Tommaso D’Alessio + Show 2 more

Open Access

https://doi.org/10.1186/1743-0003-9-82

Copy DOI

Abstract

BackgroundEye Gaze Tracking Systems (EGTSs) estimate the Point Of Gaze (POG) of a user. In diagnostic applications EGTSs are used to study oculomotor characteristics and abnormalities, whereas in interactive applications EGTSs are proposed as input devices for human computer interfaces (HCI), e.g. to move a cursor on the screen when mouse control is not possible, such as in the case of assistive devices for people suffering from locked-in syndrome. If the user’s head remains still and the cornea rotates around its fixed centre, the pupil follows the eye in the images captured from one or more cameras, whereas the outer corneal reflection generated by an IR light source, i.e. glint, can be assumed as a fixed reference point. According to the so-called pupil centre corneal reflection method (PCCR), the POG can be thus estimated from the pupil-glint vector.MethodsA new model-independent EGTS based on the PCCR is proposed. The mapping function based on artificial neural networks allows to avoid any specific model assumption and approximation either for the user’s eye physiology or for the system initial setup admitting a free geometry positioning for the user and the system components. The robustness of the proposed EGTS is proven by assessing its accuracy when tested on real data coming from: i) different healthy users; ii) different geometric settings of the camera and the light sources; iii) different protocols based on the observation of points on a calibration grid and halfway points of a test grid.ResultsThe achieved accuracy is approximately 0.49°, 0.41°, and 0.62° for respectively the horizontal, vertical and radial error of the POG.ConclusionsThe results prove the validity of the proposed approach as the proposed system performs better than EGTSs designed for HCI which, even if equipped with superior hardware, show accuracy values in the range 0.6°-1°.

Highlights

Eye Gaze Tracking Systems (EGTSs) estimate the Point Of Gaze (POG) of a user
Applications of EGTSs can be classified as diagnostic, where the user’s visual and attentional processes are quantified, or interactive where the user inter-acts with the EGTS [1]: in the first case, the obtained data are used to study oculomotor characteristics and abnormalities; in the second scenario, EGTSs are proposed as input devices for human computer interfaces (HCIs), e.g. to move a cursor on the screen when mouse control is not possible, such as in the case of assistive devices for people with motor disabilities or suffering from
If the user’s head remains still and the cornea rotates around its fixed centre, the pupil follows the eye in the captured images, whereas the outer corneal reflection generated by an IR light source, i.e. glint, can be assumed as a fixed reference point

Summary

Introduction

Eye Gaze Tracking Systems (EGTSs) estimate the Point Of Gaze (POG) of a user. In diagnostic applications EGTSs are used to study oculomotor characteristics and abnormalities, whereas in interactive applications EGTSs are proposed as input devices for human computer interfaces (HCI), e.g. to move a cursor on the screen when mouse control is not possible, such as in the case of assistive devices for people suffering from locked-in syndrome. If the user’s head remains still and the cornea rotates around its fixed centre, the pupil follows the eye in the images captured from one or more cameras, whereas the outer corneal reflection generated by an IR light source, i.e. glint, can be assumed as a fixed reference point. If the user’s head remains still and the cornea rotates around its fixed centre, the pupil follows the eye in the captured images, whereas the outer corneal reflection generated by an IR light source, i.e. glint, can be assumed as a fixed reference point. The glint appears in the IR band as a small intense spot whereas the pupil can be captured thanks to two distinct effects generated by IR: the dark pupil (Figure 1, left) if the IR light source is placed away from the camera (off-axis), and the bright eye (Figure 1 right) if the IR light source is close to the optical axis (on-axis) [6,7,8]

Methods

Results

Conclusion