Building a Personalized, Auto-Calibrating Eye Tracker from User Interactions

Abstract

We present PACE, a Personalized, Automatically Calibrating Eye-tracking system that identifies and collects data unobtrusively from user interaction events on standard computing systems without the need for specialized equipment. PACE relies on eye/facial analysis of webcam data based on a set of robust geometric gaze features and a two-layer data validation mechanism to identify good training samples from daily interaction data. The design of the system is founded on an in-depth investigation of the relationship between gaze patterns and interaction cues, and takes into consideration user preferences and habits. The result is an adaptive, data-driven approach that continuously recalibrates, adapts and improves with additional use.Quantitative evaluation on 31 subjects across different interaction behaviors shows that training instances identified by the PACE data collection have higher gaze point-interaction cue consistency than those identified by conventional approaches. An in-situ study using real-life tasks on a diverse set of interactive applications demonstrates that the PACE gaze estimation achieves an average error of 2.56º , which is comparable to state-of-theart, but without the need for explicit training or calibration. This demonstrates the effectiveness of both the gaze estimation method and the corresponding data collection mechanism.

Moment-to-Moment Detection of Internal Thought during Video Viewing from Eye Vergence Behavior

“…Third, the in-situ experiment also exposed that vergence features depend on accurate gaze estimation. This issue could be addressed using implicit calibration approaches (Huang, Kwok, Ngai, Chan, andLeong (2016) Papoutsaki et al (2016)). Fourth, we hope to further explore vergence behaviour by representation learning from a large-scale in-the-wild data.…”

Unobtrusive and pervasive video-based eye-gaze tracking

“…One approach towards the reduction of calibration effort performs implicit calibration while the user is induced to execute a specific action, such as following a moving target [114] or gazing at a static visual stimulus [115], or while the user is interacting with a personal computer through mouse clicks [116][117][118][119] and keyboard presses [118,119]. Such user activities permit the unobtrusive collection of data in the background without the user being explicitly aware of the calibration task.…”

An Algorithm for Automatically Detecting Dyslexia on the Fly

“…Another study with self calibrating eye tracking technology which used a webcam consisted of pre-recorded gaze patterns [13]. Another eye tracking technology called PACE could self calibrate using the user interactions [14]. Unlike these previous technologies, our prototype uses a completely browser -based, self-calibrating and real-time technology which uses gaze-interaction relationships.…”