A Semiparametric Model for Bayesian Reader Identification

Abstract

We study the problem of identifying individuals based on their characteristic gaze patterns during reading of arbitrary text. The motivation for this problem is an unobtrusive biometric setting in which a user is observed during access to a document, but no specific challenge protocol requiring the user's time and attention is carried out. Existing models of individual differences in gaze control during reading are either based on simple aggregate features of eye movements, or rely on parametric density models to describe, for instance, saccade amplitudes or word fixation durations. We develop flexible semiparametric models of eye movements during reading in which densities are inferred under a Gaussian process prior centered at a parametric distribution family that is expected to approximate the true distribution well. An empirical study on reading data from 251 individuals shows significant improvements over the state of the art.

A Discriminative Model for Identifying Readers and Assessing Text Comprehension from Eye Movements

“…We can conclude that the inclusion of lexical features leads to a significant improvement compared to the original generative model [7], and that a discriminative model using a Fisher kernel gives an additional considerable improvement over the generative model. We conclude that this model significantly outperforms the semiparametric model of [8] in some cases, which, to the best of our knowledge, is the best published biometric model that is based on eye movements. None of the considered models was able to reliably predict reading comprehension from a reader's eye movements.…”

“…The current gold-standard model for reader identification is the model of Abdelwahab et al, 2016 [8]. Note that no Fisher kernel can be derived from this non-parametric generative model for lack of explicit model parameters.…”

DeepEyedentificationLive: Oculomotoric Biometric Identification and Presentation-Attack Detection Using Deep Neural Networks

“…Spawned by the seminal works of Kasprowski and Ober [4] and Bednarik et al [5], and fueled by competitions in the following decade [14], [15], these methods can be subsumed into three categories: aggregational [16], [17], [18], statistical [19], [20], [21], [22] and generative methods. Suitable generative methods include Markov [23], [24] and graphical models [11], [25], [26].…”

“…The spectrum of visual stimuli that have been studied ranges from a static cross [5], images [32], faces [19], [33], [34], text [11], [12], [16], [26], video [35] and various implementations of jumping points [4], [17], [36], [37], [38]. Only a handful of studies evaluate their models on stimuli that have not also been shown to the respective user during enrollment [11], [12], [13], [16], [26], [31], [35].…”

“…The spectrum of visual stimuli that have been studied ranges from a static cross [5], images [32], faces [19], [33], [34], text [11], [12], [16], [26], video [35] and various implementations of jumping points [4], [17], [36], [37], [38]. Only a handful of studies evaluate their models on stimuli that have not also been shown to the respective user during enrollment [11], [12], [13], [16], [26], [31], [35]. Repetitions in stimulus sequences, unfortunately, enable an attacker to record the oculomotoric response to the known stimulus, and to perform a replay attack by presenting the recording to the system.…”