Abstract
In this paper, a range of open-source tools, datasets, and software that have been developed for quantitative and in-depth evaluation of eye gaze data quality are presented. Eye tracking systems in contemporary vision research and applications face major challenges due to variable operating conditions such as user distance, head pose, and movements of the eye tracker platform. However, there is a lack of open-source tools and datasets that could be used for quantitatively evaluating an eye tracker’s data quality, comparing performance of multiple trackers, or studying the impact of various operating conditions on a tracker’s accuracy. To address these issues, an open-source code repository named GazeVisual-Lib is developed that contains a number of algorithms, visualizations, and software tools for detailed and quantitative analysis of an eye tracker’s performance and data quality. In addition, a new labelled eye gaze dataset that is collected from multiple user platforms and operating conditions is presented in an open data repository for benchmark comparison of gaze data from different eye tracking systems. The paper presents the concept, development, and organization of these two repositories that are envisioned to improve the performance analysis and reliability of eye tracking systems.
Highlights
Gaze data quality refers to the validity of the gaze data measured and reported by an eye tracker [1].The most common method of representing gaze data quality is by specifying gaze estimation accuracy, which refers to the difference between the true and the measured gaze positions [2]
There currently exists significant diversity in gaze accuracy measures as described in reference [3], which leads to ambiguity in interpretation of the quality of gaze data from different eye tracking systems and difficulty in comparison of two or more eye trackers
This paper describes the GitHub code repository named GazeVisual-Lib that contains the source codes for a complete graphical user interface (GUI) application tool and a range of numerical and visualization methods for quantitative and visual exploration of eye gaze data quality
Summary
Gaze data quality refers to the validity of the gaze data measured and reported by an eye tracker [1].The most common method of representing gaze data quality is by specifying gaze estimation accuracy, which refers to the difference between the true and the measured gaze positions [2]. With the growing applications of gaze information in consumer devices like augmented and virtual reality, smartphones, and smart TVs [4,5,6,7] the eye trackers used in such applications need to be thoroughly evaluated to ensure the high quality and consistency of their gaze data outputs. This calls for the development and adoption of homogeneous metrics for reporting gaze accuracy and a consistent set of methods for complete characterization of eye trackers’ data under different operating conditions [8]. The general focus of these software is toward determining eye movement characteristics (i.e., fixations, scanpath, saccades) and studying eye movement relationships with human cognitive process, such as creation of attention maps, understanding regions of user interests, and visual search
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
