Based On Computer Vision, Gesture Interaction Technology And Its In The Game

In modern society, the social function of digital games has been gradually recognized. With the rapid development of virtual reality, human computer interaction, and computer vision, great attention has been paid to public games for home and office. Aiming at social benefits rather than commercial ones seeking by Nintendo and Microsoft, in this thesis, we propose to develop next-generation public games on household personal computers. Natural interaction techniques, especially hand gesture interaction techniques, have been chosen as breakthrough entry.The main contents of the thesis consist of following 4 parts: 1) re-engineering of a low-cost hand-held interactive device, 2) study on computer-vision-based human hand tracking and its motion representation; 3)study on gesture intention recognizers, and 4) design and implementation of a prototype gesture interaction system and its application in a typical public game.The thesis introduces the research background, and briefly surveys related research on human computer interaction, especially gesture interaction techniques. A framework of our work is presented at the end of this chapter. Commercial hand-held interaction devices are firstly introduced, then the attributes of a hand-held gadget suitable for public games are discussed, and finally the re-engineering of a low-cost hand-held gadget is described. A two-pass method for detecting the hand-held gadget is proposed which synthesizes image luminance, color and shape features, and a determinate finite state automaton is developed for hand gesture interaction.The recognition of hand gesture intention is discussed in the thesis. The parameterization of hand gestures is presented, and a rule-based approach is proposed to classify the hand gestures.In the thesis, the design and implementation of a prototype hand gesture system is presented. The proposed methods are integrated into the prototype system and applied to the popular game Tetris on PC with a hand-held LED light pen being used to drive the game instead of traditional key strokes.Experimental results show that the proposed hand gesture interactions are natural and effective.