In this work we describe a set of visual routines, which support a novel sensor free interface between a human and virtual objects. The visual routines detect, track and interpret a gesture of pointing in real time. This is solved in the context of a scenario, which enables a user to activate virtual objects displayed on a projective screen. By changing a direction of pointing with an extended towards the screen arm, the user controls the motion of virtual objects. The vision system consists of a single overhead view camera and exploits a priori knowledge of the human body appearance, interactive context and environment. The system operates in real time on a standard Pentium-PC platform.
CITATION STYLE
Kolesnik, M., & Kuleßa, T. (2001). Detecting, tracking, and interpretation of a pointing gesture by an overhead view camera. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 2191, pp. 429–436). Springer Verlag. https://doi.org/10.1007/3-540-45404-7_57
Mendeley helps you to discover research relevant for your work.