In the literature, it has been proven that the distinction between relevant and irrelevant information through eye tracking is a technique that optimizes Robotic Process Automation (RPA) processes. To implement this, an eye tracking tool must be installed, both in terms of hardware and software. This paper addresses the need for accessible and cost-effective technologies to enhance productivity and efficiency through user-eye movement tracking. We review current webcam-based eye tracking software, assessing their availability, accessibility, and accuracy in estimating the user’s Point of Gaze (POG) on a standard computer setup. Through a systematic review and a pilot benchmarking study, we evaluate GazeRecorder, Webgazer.js, and EyeDid webcam-based eye tracking software on various metrics. Our findings reveal significant performance variations, indicating that while webcam-based eye tracking software may not match the high precision of specialized devices, they offer a viable alternative for RPA applications requiring user interaction analysis. This research improves our understanding of webcam eye tracking capabilities and limitations in RPA, providing insights that could aid the integration of eye tracking technologies into business processes.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Exploring Webcam Eye Tracking Software for Robotic Process Automation: A Pilot Benchmarking Study

  • Manuel García-Romero,
  • Antonio Martínez-Rojas,
  • José González Enríquez,
  • Andrés Jiménez-Ramírez

摘要

In the literature, it has been proven that the distinction between relevant and irrelevant information through eye tracking is a technique that optimizes Robotic Process Automation (RPA) processes. To implement this, an eye tracking tool must be installed, both in terms of hardware and software. This paper addresses the need for accessible and cost-effective technologies to enhance productivity and efficiency through user-eye movement tracking. We review current webcam-based eye tracking software, assessing their availability, accessibility, and accuracy in estimating the user’s Point of Gaze (POG) on a standard computer setup. Through a systematic review and a pilot benchmarking study, we evaluate GazeRecorder, Webgazer.js, and EyeDid webcam-based eye tracking software on various metrics. Our findings reveal significant performance variations, indicating that while webcam-based eye tracking software may not match the high precision of specialized devices, they offer a viable alternative for RPA applications requiring user interaction analysis. This research improves our understanding of webcam eye tracking capabilities and limitations in RPA, providing insights that could aid the integration of eye tracking technologies into business processes.