Eye-Movements as a Biometric

Abstract

Abstract. We propose the use of eye-movements as a biometric. A case study investigating potentials of eye-movement data for biometric purposes was conducted. Twelve participants' eye-movements were measured during still and moving objects viewing. The measured data includes pupil sizes and their dynamics, gaze velocities and distances of infrared reflections of the eyes. For still object viewing of 1 second duration, identification rate of 60 % can be obtained by using dynamics of pupil diameters. We suggest an integration of the eyemovement-based identification into general video-based biometric systems.

On Biometrics With Eye Movements

“…Eye movements were measured from 9 subjects with an eye tracking device on the basis of two small video cameras catching the movements of the pupil of each eye. Both a moving spot and images as stimulations have been used for 12 subjects [21]. Nevertheless, eye movements were found to be less useful for biometric identification than other factors, particularly the distance between the eyes of the subject.…”

Towards task-independent person authentication using eye movement signals

“…Although the accuracy is not sufficient for a realistic application, it is clearly below the chance level (50 % EER), suggesting that there is individual information in the eye movements which can be modeled. The error rates are clearly higher than in [Kasprowski 2004;Bednarik et al 2005]. However, those studies used fixeddimensional templates that were carefully aligned whereas our system does not use any explicit temporal alignment.…”

“…customized user profiles in computers, adaptive user interfaces). We are aware of two prior studies that have utilized physical features of the eye movements to recognize persons [Kasprowski 2004;Bednarik et al 2005]. In [Kasprowski 2004], the author used a custommade head-mounted infrared oculography eye tracker ("OBER 2") with sampling rate of 250 Hz to collect data from N = 47 subjects.…”

“…In an independent study [Bednarik et al 2005], the authors used a commercially available eye tracker (Tobii ET-1750), built in a TFT panel, to collect data from N = 12 subjects at sampling rate of 50 Hz. The stimulus consisted of a single cross shown in the middle of the screen and displayed for 1 second.…”

“…Both of the studies [Kasprowski 2004;Bednarik et al 2005] are inherently task-dependent: they assume that the same stimulus appears in training and testing. This approach has the advantage that the reference template (training sample) and test sample can be accurately aligned.…”