Abstract

Remote eye tracking has become an important tool for the online analysis of learning processes. Mobile eye trackers can even extend the range of opportunities (in comparison to stationary eye trackers) to real settings, such as classrooms or experimental lab courses. However, the complex and sometimes manual analysis of mobile eye-tracking data often hinders the realization of extensive studies, as this is a very time-consuming process and usually not feasible for real-world situations in which participants move or manipulate objects. In this work, we explore the opportunities to use object recognition models to assign mobile eye-tracking data for real objects during an authentic students’ lab course. In a comparison of three different Convolutional Neural Networks (CNN), a Faster Region-Based-CNN, you only look once (YOLO) v3, and YOLO v4, we found that YOLO v4, together with an optical flow estimation, provides the fastest results with the highest accuracy for object detection in this setting. The automatic assignment of the gaze data to real objects simplifies the time-consuming analysis of mobile eye-tracking data and offers an opportunity for real-time system responses to the user’s gaze. Additionally, we identify and discuss several problems in using object detection for mobile eye-tracking data that need to be considered.

Highlights

  • In the case of you only look once (YOLO) v4, the true gaze was on the rail but predicted as "Other" 79 times, and it was predicted as paper 5 times

  • For the Faster R-Convolutional Neural Networks (CNN), the gaze was on the rail but was predicted as “Other” 125 times and predicted as paper 62 times

  • The results for all objects for YOLO v4 and Faster R-CNN can be seen in Tables 2 and 3; micro averages calculate the average of the contribution of all classes, which is useful for imbalanced classes and corresponds to the algorithm’s accuracy, whereas macro averages estimate the average across class averages, independently of their contribution, and weighted averages weigh the average by the number of samples in each class

Read more

Summary

Introduction

Mobile eye-tracking solutions became available, which offer the opportunity to extend the range of applications in comparison to those of stationary eye trackers to real field studies, such as lectures or classrooms conducting experimental work. Despite these compelling opportunities, and mobile eye-tracking has been applied in some cases (see Reference [5] for a review), quantitative data analysis is often absent in such studies. This is due to the fact that manual labelling is a very time-consuming process and, not feasible for real-world situations in which participants move or manipulate objects and their view changes with almost every frame

Objectives
Methods
Results
Conclusion
Full Text
Paper version not known

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

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.