Gesture Recognition with Multiple Spatial Feature Fusion
- 10.2991/icmmct-16.2016.351How to use a DOI?
- Multiple Space, Gesture Recognition, Depth Image, Feature Filtering
Hand gesture recognition is an important topic in the field of computer vision, it has a wide range of applications, such as interactive games and sign language recognition, etc.. With the launch of the depth sensor, the task of hand gesture recognition becomes more simple. In recent years, there are a number of methods to try to extract features in the depth image, which is to be an effective expression of some kind of gesture. However, due to the inherent flexibility and complexity of the gesture, the recognition performance of the existing algorithms on large data sets is still not satisfactory. Local shape is presented in this paper a novel method based on multi spatial feature fusion to recognize static hand gesture depth image, namely on the 3D point cloud which were the local principal component analysis. It extract the local gradient information and local points cloud depth distribution, the effective information coding the gesture, we put local features which are the concatenation of the gesture image features and the classification results of the random forest classifier of features are filtered to remove the result of classification which did not influence the characteristics. We adopt filtered features to train the random forest again to recognize gestures. Compared with the existing algorithms, this method can effectively improve the recognition rate of the two large gesture data sets.
- © 2016, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - Meng Qin AU - Guoping He PY - 2016/03 DA - 2016/03 TI - Gesture Recognition with Multiple Spatial Feature Fusion BT - Proceedings of the 2016 4th International Conference on Machinery, Materials and Computing Technology PB - Atlantis Press SP - 1765 EP - 1769 SN - 2352-5401 UR - https://doi.org/10.2991/icmmct-16.2016.351 DO - 10.2991/icmmct-16.2016.351 ID - Qin2016/03 ER -