Webcam-based eye-tracking for reading


Current state-of-the-art webcam-based eye-trackers [1] are good enough to track web searches and to predict clicking behavior. However, their performance is not yet optimal for word-level eye-tracking when reading natural text in smaller font sizes. The goal of this project is to build or improve on an existing webcam-based eye-tracking system with a focus on on reading full sentences. The project includes developing and evaluating this new system.

Supervised by Nora Hollenstein

noraho@inf.ethz.ch

References

  1. [1] Papoutsaki, A., Sangkloy, P., Laskey, J., Daskalova, N., Huang, J., & Hays, J. (2016, January). Webgazer: Scalable webcam eye tracking using user interactions. In Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence-IJCAI 2016.