You are here: Home / InSight Out: object recognition for mobile eye-tracking data analysis

InSight Out: object recognition for mobile eye-tracking data analysis

Welcome to the homepage of the InSight Out project, a tech transfer project that focuses on the application of state-of-the-art vision techniques to the analysis of mobile eye-tracking data. Starting from October 2012, we will develop, implement and test a series of object recognition algorithms that will make the use of mobile eye-tracking technology both more flexible and user friendly for a range of applications and research questions.

Two observations have inspired the InSight Out project:

  • First, recent years have witnessed rapid technological advances in eye-tracking, the technology that allows for measuring of people’s eye movements and gaze during online behaviour. One of the most striking trends in this domain is the development and commercialization of unobtrusive, mobile eye-tracking systems. These light-weight and portable systems, typically integrated in a pair of glasses, in principle allow for tracking gaze behaviour in natural user environments, effectively taking eye-tracking out of the lab and into the wild.
  • Second, measuring eye gaze in real-life situations, with subjects moving around freely in constantly changing scenes with moving objects, typically requires a substantial amount of manual analysis, which is both a time-consuming and costly process. The result is that the efficient use of mobile eye-tracking to date still requires controllable (i.e. predefined) and static settings. This, however, does not do justice to the intrinsic potential of the technology to provide a versatile tool for fully mobile eye-tracking research.

The InSight Out project aims to extend the domain of application for mobile eye-tracking through the development of a (semi-)automatic video analysis module based on object recognition techniques. The technical details of the module are described on the project description page. In order to meet the requirement of researchers and professionals (interested in) working with mobile eye-tracking technology, the project is supported and guided by a diversified user commission . In collaboration with these partners, a series of specific user cases will be developed in the course of the project that test the potential of the newly developed tool.