Chapter 8
Automatic analysis of in-the-wild mobile eye-tracking experiments using object, face and person detection
In this chapter, we discuss a novel method for the analysis of mobile eye-tracking data in natural environments. Mobile eye-tracking systems generate large amounts of continuous data, making manual analysis extremely time-consuming. Available solutions provided by commercially available eye-tracking systems, such as marker-based analysis, minimize the manual labor but require experimental control, making real-life experiments practically unfeasible, and generally only apply to the analysis of objects. Here, we discuss a novel method for the processing of mobile eye-tracking data, based on the integration of computer vision techniques. Using such an approach allows us to automatically detect specific objects, faces and human bodies/body parts in images captured by a mobile eye-tracker. By mapping the gaze data on top of these detections, we gain insights into the visual behavior of recorded participants. As an important step in the integration of this method in the analysis of multimodal interaction, we developed an output format that is compatible with annotation tools such as ELAN, making our software integratable with existing annotations. In this chapter we give an overview of relevant image processing techniques and their application in interaction studies. We also present a thorough comparison between manual analysis and our automatic analysis in both speed and accuracy on challenging, real-life experiments.
Article outline
- 1.Introduction
- 2.Related work
- 2.1Object recognition techniques
- 2.2Object detection techniques
- 3.Recognition and detection solution for mobile eye-tracking data: A technical description
- 3.1Recognition of specific objects
- 3.2Detection of faces and bodies
- 3.3Person reidentification
- 4.Experimental results
- 4.1Object recognition results
- 4.2Results of face and body detections
- 4.3Combined results of objects, face and body detection
- 5.Conclusion and future work
-
Note
-
References