Abstract | ||
---|---|---|
The salient feature points of hand gesture play an important role for its representation and recognition. In this paper, a novel hand gesture recognition method based on salient feature point selection is proposed. The raw data of hand gesture is captured by the Kinect sensor and the hand gesture is segmented from the cluttered background. The shape feature of hand gesture is extracted from the contour, and the salient feature points are selected by a new algorithm to represent the hand gesture. Finally, the Dynamic Time Warping algorithm is modified and employed to find the best correspondence between two gestures. Extensive experiments are implemented on three benchmark databases to validate the effectiveness of our method. The experimental results verified the invariance of our method to translation, rotation scaling and articulated deformation. The comparison with state-of-the-art methods demonstrates the accuracy and efficiency of our method. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1109/RCAR.2017.8311843 | 2017 IEEE International Conference on Real-time Computing and Robotics (RCAR) |
Keywords | Field | DocType |
salient feature point selection,hand gesture recognition method,image representation,Kinect sensor,hand gesture segmentation,Dynamic Time Warping algorithm,RGB-D | Computer vision,Dynamic time warping,Invariant (physics),Gesture,Computer science,Gesture recognition,Image segmentation,Feature extraction,RGB color model,Artificial intelligence,Salient | Conference |
ISBN | Citations | PageRank |
978-1-5386-2036-6 | 0 | 0.34 |
References | Authors | |
7 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yiwen, H. | 1 | 1 | 1.02 |
Jianyu Yang | 2 | 429 | 78.51 |
Zhanpeng Shao | 3 | 25 | 8.50 |
Y. F. Li | 4 | 1128 | 105.83 |