The selected problem of multimodal interaction for multimedia is discussed. The application for visual analysis of eye winking is presented. The proposed algorithm allows interpreting “eye gestures” as appropriate system events, which corresponds to executing system commands. The main aim of this paper is to develop replacing the usage of mouse keys by eye winking. Such a solution can work in multimodal interaction for many different applications. The work of surgeon in the operating room is a good example of situation, where busy hands do not allow controlling the computer system by traditional devices. Introduced interaction based on eye winking can help to communicate with system. The prototype has been built and set of experiments have been carried out in a group of 30 people. Tests have shown high acceptance of the proposed solution.
CITATION STYLE
Kowalczyk, P., & Sawicki, D. (2017). Wink Detection on the Eye Image as a Control Tool in Multimodal Interaction. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10485 LNCS, pp. 353–362). Springer Verlag. https://doi.org/10.1007/978-3-319-68548-9_33
Mendeley helps you to discover research relevant for your work.