Implicit Calibration Using Probable Fixation Targets.

Pawel Kasprowski,Przemysław Skurowski,Katarzyna Harȩżlak

doi:10.3390/s19010216

Pawel Kasprowski, Przemysław Skurowski + Show 1 more

Open Access

https://doi.org/10.3390/s19010216

Copy DOI

Abstract

Proper calibration of eye movement signal registered by an eye tracker seems to be one of the main challenges in popularizing eye trackers as yet another user-input device. Classic calibration methods taking time and imposing unnatural behavior on eyes must be replaced by intelligent methods that are able to calibrate the signal without conscious cooperation by the user. Such an implicit calibration requires some knowledge about the stimulus a user is looking at and takes into account this information to predict probable gaze targets. This paper describes a possible method to perform implicit calibration: it starts with finding probable fixation targets (PFTs), then it uses these targets to build a mapping-probable gaze path. Various algorithms that may be used for finding PFTs and mappings are presented in the paper and errors are calculated using two datasets registered with two different types of eye trackers. The results show that although for now the implicit calibration provides results worse than the classic one, it may be comparable with it and sufficient for some applications.

Highlights

With growing access to cheap eye-tracking devices using simple web cameras, and advancements in image processing and analysis, it has become possible to incorporate eye tracking as yet another human–computer interface.The accuracy and precision of cheap eye trackers are lower than the high-end eye trackers; there are numerous examples that such eye trackers may be used for simple tasks and for research [1]
The goal of the research summarized in this paper is to develop calibration methods that make using eye tracking possible for unattended and inexperienced end-users
The original idea, used during the research presented in this paper, is introducing an additional step of processing during which a saliency map is used to build a short list of probable fixation targets (PFTs)

Summary

Introduction

With growing access to cheap eye-tracking devices using simple web cameras, and advancements in image processing and analysis, it has become possible to incorporate eye tracking as yet another human–computer interface.The accuracy and precision of cheap eye trackers (priced less than $500) are lower than the high-end eye trackers (for more than $5000); there are numerous examples that such eye trackers may be used for simple tasks and for research [1]. With growing access to cheap eye-tracking devices using simple web cameras, and advancements in image processing and analysis, it has become possible to incorporate eye tracking as yet another human–computer interface. There are publications confirming that e.g., The Eye Tribe [2], Gazepoint GP3 [3] or Tobii EyeX [4] may give reliable eye-tracking results when certain conditions are met. One of the most important problems when considering eye-tracking usage on an end-user market is the necessity of calibration. Most eye trackers require to be calibrated prior to each usage. Implementation of all presented algorithms in a publicly available new version of the ETCAL library which has been created for calibration purposes and is available at GitHub [30]

Objectives

Methods

Results

Conclusion