Please use this identifier to cite or link to this item: http://scholarbank.nus.edu.sg/handle/10635/92570
Title: VIP: A UNIFYING FRAMEWORK FOR COMPUTATIONAL EYE-GAZE RESEARCH
Authors: MA KENG TECK (MA QINGDE)
Keywords: Eye-tracking, Computational model, User Profiling, Personalisation, Dataset, Saliency
Issue Date: 8-Jul-2014
Source: MA KENG TECK (MA QINGDE) (2014-07-08). VIP: A UNIFYING FRAMEWORK FOR COMPUTATIONAL EYE-GAZE RESEARCH. ScholarBank@NUS Repository.
Abstract: Eye-gaze has been used in wide range of computer science research. For saliency research, the reference model is top-down and bottom-up. In biometric research, eye-gaze is a soft biometric. In human-computer interface, eye-gaze is a response of the interactions between tasks and visual stimulus. These models are incomplete and we propose the VIP framework. This framework captures the dependence of eye-gaze on Visual stimuli, Intent, and Person, subsuming all existing models. We conducted user experiments to collect the VIP and VVIP datasets. The utility of our framework is illustrated with 2 applications: (i) inferring viewer's personal traits and (ii) an implicit just-in-time profiling system: Eye-2-I. The first is a novel use of eye-gaze data; the second is the first system to use the eye-gaze data for comprehensive user profiling. Finally, we demonstrate the superiority of our framework over existing models by proposing a trait-specific fixation prediction approach.
URI: http://scholarbank.nus.edu.sg/handle/10635/92570
Appears in Collections:Ph.D Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
MaKT.pdf2.39 MBAdobe PDF

OPEN

NoneView/Download

Page view(s)

76
checked on Mar 10, 2018

Download(s)

53
checked on Mar 10, 2018

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.