Vision-based hand-gesture applications. Commun ACM Academic Article uri icon

abstract

  • The use of hand gestures for controlling computer games, robots, medical visualization devices and crisis management systems is a fairly new human-centered method providing an expressive, natural and intuitive way of interaction. However, to successfully implement hand gesture systems for diverse everyday environments, significant technical and usability challenges must be overcome. The systems must have fast response times and high recognition accuracy, must be easy to learn and must provide a high degree of user satisfaction. These issues are discussed here with an emphasis on machine vision techniques, since this modality of interaction presents most of the technical challenges and hence is the most difficult to implement. This paper presents a state-of-the-art review of real-world applications. These include four areas in which hand gesture interfaces show the fastest advances: health environments and assistive technologies, entertainment, crisis management/disaster relief systems, and human-robot interaction. These systems use hand gestures alone and in combination with body movements, gaze, facial expressions and voice (multimodal). Even excluding the use of gestures in games and virtual reality environments, hand gesture interfaces are an active research area and require more development before widespread deployment can be expected. This review highlights the potential of gestures as a natural modality for human-machine interaction, and indicates the need for additional research and evaluation procedures for such interfaces to gain widespread acceptance.

publication date

  • January 1, 2011