Using Mobile Eye Tracking to Evaluate the Satisfaction with Service Office

Abstract

Eye tracking can be applied to a variety of scenarios as a means of measuring visual attention and interpreting visual solution strategies. In this article, we use mobile eye-tracker to collect information to evaluate user satisfaction with the tax service office. Mobile eye tracking can collect precise information concerning the users’ visual attention and interactions in authentic environments. Unlike screen-based eye-tracker using a laboratory or stationary computer, mobile eye tracking also can be used effectively in a walk a round scene where users could walk around and interact with diverse resources. In the progress of eye tracking data analysis, fixations and gaze points, areas of interest (AOIs), heat-maps play an important role. This annotation is typically the most time-consuming step of the analysis process. To reduce processing time and human effort, we introduced the latest computer vision techniques (i.e., You Only Look Once, YOLO) based on a convolutional neural network (CNN) to detect and recognize specific objects in recorded video. We propose a new method to evaluate the user satisfaction of a service system by implicating mobile eye tracker. In addition we gave a new idea of using CNN-based object detection technique to annotate video data collected by mobile eye tracker, which could be followed up for further analysis.