A MultiTouch Surface Using Multiple Cameras （使用多个摄像头的多点触控表面）.pdfVIP

A Multi-Touch Surface Using Multiple Cameras Itai Katz, Kevin Gabayan, Hamid Aghajan Dept. of Electrical Engineering, Stanford University, Stanford, CA 94305 {katz,gabayan,aghajan}@ Abstract. In this project we present a framework for a multi-touch surface us- ing multiple cameras. With an overhead camera and side-mounted camera we determine the three dimensional coordinates of the fingertips and detect touch events. We interpret these events as hand gestures which can be generalized into commands for manipulating applications. We offer an example applica- tion of a multi-touch finger painting program. 1 Motivation Traditional human input devices such as the keyboard and mouse are not suffi- ciently expressive to capture more natural and intuitive hand gestures. A richer ges- tural vocabulary may be enabled through devices that capture and understand more gestural information. Interactive tables that detect fingertip touches and interpret these events as gestures have succeeded to provide a richer input alternative. Touch detection estimates in overhead camera based systems suffer in depth resolution, causing a system to report touches when a user’s finger hovers near the touch sur- face. In our multi-touch surface we strive to build a system that is low-cost, scaleable, and improves existing vision-based touch-table accuracy. 2 Background Multi-touch input devices have been explored by Bill Buxton [1] as early as 1984. Touch position sensing in touchpads is a common laptop input device [2], measuring the capacitance created between a user’s finger and an array of charged plates under the touch surface. Larger touch sensitive surfaces have been constructed from similar sensor arrays. The Mitsubishi DiamondTouch display s