Gaze Gesture Recognition with Hierarchical Temporal Memory Networks
Eye movements can be consciously controlled by humans to the extent of performing sequences of predefined movement patterns, or ’gaze gestures’. Gaze gestures can be tracked non-invasively employing a video-based eye tracking system. Gaze gestures hold great potential in the context of Human Computer Interaction as low-cost gaze trackers become more ubiquitous. In this work, we build an original set of 50 gaze gestures and evaluate the recognition performance of a Bayesian inference algorithm known as Hierarchical Temporal Memory, HTM. HTM uses a neocortically inspired hierarchical architecture and spatio-temporal coding to perform inference on multi-dimensional time series. Here, we show how an appropiate temporal codification is critical for good inference results. Our results highlight the potential of gaze gestures for the fields of accessibility and interaction with smartphones, projected displays and desktop computers.
KeywordsNeural Network Architecture Soft Computing Pattern Recognition Time series analysis and prediction
Unable to display preview. Download preview PDF.
- 2.Mollenbach, E., Hansen, J.P., Lillholm, M., Gale, A.G.: Single stroke gaze gestures. In: Proceedings of the 27th International Conference Extended Abstracts on Human Factors in Computing Systems, CHI 2009, pp. 4555–4560. ACM, New York (2009), http://doi.acm.org/10.1145/1520340.1520699 Google Scholar
- 4.San Agustin, J., Skovsgaard, H., Mollenbach, E., Barret, M., Tall, M., Hansen, D.W., Hansen, J.P.: Evaluation of a low-cost open-source gaze tracker. In: ETRA 2010: Proceedings of the 2010 Symposium on Eye-Tracking Research & Applications, pp. 77–80. ACM, New York (2010)Google Scholar