A solution of computer vision based real-time hand pointing recognition

Gesture recognition is needed for a variety of applications. It develops a real-time arm pointing system. The main contribution of the system is using stereo vision pointing to track the pointing arm and identify several pointing targets in 3D space. Our system consists of (1) hand tracking (2) hand shape extraction (3) finger orientation calculation and (4) finger line finding. The system allows the user to make the arm pointing and the walking in a work space at the same time. The novelty of our method is directly tracking two 3D points representing the pointing line in 3D space and then refining the tracking results. We take advantages of PCA (principal component analysis) to extend the samples of particle filter to 3D space. Experiments show that our system will analyse each frame of video input within about 1/10 second and achieve a near-real-time system. The pointing accuracy of our system is measured by the success rate of 100 pointing and the success rate of our system is above 90%.