In this paper, we propose an automatic system that handles hand gesture spotting and recognition simultaneously in stereo color image sequences without any time delay based on Hidden Markov Models (HMMs). Color and 3D depth map are used to segment hand regions. The hand trajectory will determine in further step using Mean-shift algorithm and Kalman filter to generate 3D dynamic features. Furthermore, k-means clustering algorithm is employed for the HMMs codewords. To spot meaningful gestures accurately, a non-gesture model is proposed, which provides confidence limit for the calculated likelihood by other gesture models. The confidence measures are used as an adaptive threshold for spotting meaningful gestures. Experimental results show that the proposed system can successfully recognize isolated gestures with 98.33% and meaningful gestures with 94.35% reliability for numbers (0-9). © 2009 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Elmezain, M., Al-Hamadi, A., & Michaelis, B. (2009). Hand Gesture Spotting Based on 3D Dynamic Features Using Hidden Markov Models. In Communications in Computer and Information Science (Vol. 61, pp. 9–16). https://doi.org/10.1007/978-3-642-10546-3_2
Mendeley helps you to discover research relevant for your work.