2014 | OriginalPaper | Buchkapitel
Micro-pose for Gesture Recognition with Bodily-Pose Decomposition
verfasst von : Jae-wan Park, Su-yong Park, Chil-woo Lee
Erschienen in: HCI International 2014 - Posters’ Extended Abstracts
Verlag: Springer International Publishing
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Because a pose is implemented with static configuration of bodily joints expressing an abstract meaning of human intention, a gesture can be thought as a sequential combination of specific poses of human body. Therefore we can understand the meaning of a gesture if we can analyze the arrangement of the poses in spatiotemporal space. In this paper we propose a novel gesture recognition algorithm in which a pose is decomposed into micro-poses that are constructed with a small number of bodily joints. A micro-pose is a smallest unit of joints which can include a physical mean of body. To obtain the location of body joint MS Kinect is used and the information extracted from the micro-poses is finally applied to code matrix as the code book of the Bag-of-Words to understand the meaning of the gesture.