Abstract
We study involuntary micro-movements of both eyes, in addition to saccadic macro-movements, as biometric characteristic. We develop a deep convolutional neural network that processes binocular eye-tracking signals and verifies the viewer’s identity. In order to detect presentation attacks, we develop a model in which the movements are a response to a controlled stimulus. The model detects replay attacks by processing both the controlled but randomized stimulus and the ocular response to this stimulus. We acquire eye movement data from 150 participants, with 4 sessions per participant and conduct experiments on this new and legacy data sets with varying tracker precision and sampling rate. We observe that the model detects replay attacks reliably. For identification and identity verification, the model attains substantially lower error rates than prior work. We explore the relationships between training population size, training data volume, types of visual stimuli, number of training and enrollment sessions, interval between enrollment and probe sessions on one hand and the model performance on the other hand.
Highlights
N O single biometric characteristic that is known today is by itself sufficiently reliable for all biometric applications, unique, collectible, convenient, and universally available
While identification based on fingerprint and iris tend to be more accurate than facial recognition, a good-quality fingerprint cannot be obtained for approximately 2-4% of the population due to degradation of the fingerprints from manual labor or hand-related disabilities, while long eyelashes, small eye apertures, cosmetic contact lenses, and conditions including glaucoma and cataract prevent the collection of good-quality images of the iris for an estimated 7% of the population [1]
As representatives for deep learning based methods we compare against the DeepEyedentification network, which differs from DeepEyedentificationLive in that it can only process monocular data and lacks presentation-attack detection; and Abdelwahab and Landwehr [29], who train a distributional sequence embedding on raw gaze sequences and pupil dilations
Summary
N O single biometric characteristic that is known today is by itself sufficiently reliable for all biometric applications, unique, collectible, convenient, and universally available. A neural network has been studied that processes a raw monocular eye tracking signal measured during reading [12] This approach does not rely on any prior detection of specific types of macro- or micro-movements. In order to detect replay attacks, we develop a model in which the eye movements are the ocular response to a challenge in the form of a controlled stimulus. In this setting, the identification task becomes more challenging as fixation durations and saccade amplitudes are largely determined by the stimulus, and their distributional properties are less likely to vary across individuals.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: IEEE Transactions on Biometrics, Behavior, and Identity Science
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
