| Summary: | 碩士 === 國立臺北科技大學 === 創新設計研究所 === 101 === From command line interface (CLI) to graphical user interface (GUI), nature user interface (NUI) is the latest trend of Human-Computer Interaction. Such as touch panel, gesture recognition, posture recognition, face tracking, eye tracking, voice control and brainwave. All of this are attempt to replace present device, like mouse and keyboard, to achieve more nature and hand free control.
This paper proposes a rapid method for posture-sensed NI application production. This method can use to develop NI applications in many purposes. Use depth-sensing camera as input device, get transformations from human body, and define the rotation of joints as feature vector. Use multilayer perceptron artificial neural network as learning technique. Make computer to learn and to recognize human posture from statistical data. And verify this method by implement a full-body NI installation and a half-body NI game.
|
|---|
