Input precision for gaze-based graphical passwords

  • Authors:
  • Alain Forget;Sonia Chiasson;Robert Biddle

  • Affiliations:
  • Carleton University, Ottawa, ON, Canada;Carleton University, Ottawa, ON, Canada;Carleton University, Ottawa, ON, Canada

  • Venue:
  • CHI '10 Extended Abstracts on Human Factors in Computing Systems
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Click-based graphical passwords have been proposed as alternatives to text-based passwords, despite being potentially vulnerable to shoulder-surfing, where an attacker can learn passwords by watching or recording users as they log in. Cued Gaze-Points (CGP) is a graphical password system which defends against such attacks by using eye-gaze password input, instead of mouse-clicks. A first user study revealed that CGP's unique use of eye tracking required special techniques to improve gaze precision. In this paper, we present two enhancements that we developed and tested: a nearest-neighbour gaze-point aggregation algorithm and a 1-point calibration before each password entry. We found that these enhancements made a substantial improvement to users' gaze accuracy and system usability.