作者: S Mohamed Mansoor Roomi , R Jyothi Priya , H Jayalakshmi
DOI: 10.3844/JCSSP.2010.1002.1007
关键词: Input device 、 Gesture 、 Ubiquitous computing 、 Computer science 、 Gesture recognition 、 Computer vision 、 Artificial intelligence 、 Event (computing) 、 Mixture model
摘要: Problem statement: With the development of ubiquitous computing, current user interaction approaches with keyboard, mouse and pen are not sufficient. Due to limitation these devices useable command set is also limited. Direct use hands can be used as an input device for providing natural interaction. Approach: In this study, Gaussian Mixture Model (GMM) was extract hand from video sequence. Extreme points were extracted segmented using star skeletonization recognition performed by distance signature. Results: The proposed method tested on dataset captured in closed environment assumption that should Field Of View (FOV). This study 5 different datasets varying lighting conditions. Conclusion: specifically a real time vision system gesture based computer control event like navigation slides Power Point Presentation.