Title :
A Hand Gesture Recognition System Based on GMM Method for Human-Robot Interface
Author :
Yihsin Ho ; Nishitani, Takashi ; Yamaguchi, Toru ; Sato-Shimokawara, Eri ; Tagawa, Norio
Author_Institution :
Fac. of Syst. Design, Tokyo Metropolitan Univ., Tokyo, Japan
Abstract :
This paper proposes a hand gesture recognition system for human-robot interface. Our research aims to provide users user-friendly operations in a more intuitive manner. We use the stereo camera to capture images as the primary source of information retrieval, and adapt Gaussian mixture model (GMM) method as the main method of image analysis. The GMM method we applied in this paper is a precise, stable and computationally efficient foreground segment method. Our system is mainly with the following three steps: take video by camera, obtain user´s images based on GMM method, and recognize hand gesture. In this paper, we will focus on describing the system´s overall concepts and GMM method. An experiment result of our prototype will also be discussed to show the research potential of our system.
Keywords :
Gaussian processes; gesture recognition; human computer interaction; human-robot interaction; image capture; image segmentation; mixture models; stereo image processing; GMM method; Gaussian mixture model method; foreground segment method; hand gesture recognition system; human-robot interface; image analysis; image capture; information retrieval; stereo camera; user-friendly operation; Cameras; Educational institutions; Gaussian distribution; Gesture recognition; Robots; System analysis and design; Transforms; Gaussian Mixture Model; Hand gesture recognition; Human-robot interface;
Conference_Titel :
Robot, Vision and Signal Processing (RVSP), 2013 Second International Conference on
Conference_Location :
Kitakyushu
Print_ISBN :
978-1-4799-3183-5
DOI :
10.1109/RVSP.2013.72
