Abstract | ||
---|---|---|
In this paper we propose a new method for view-invariant gesture recognition, based on what we call nonparametric shape descriptor. We represent gestures as 3D motion trajectories and then we prove that the shape of a trajectory is equivalent to the Euclidean distances between all its points. The set of point-to-point distances description is mapped to a high-dimensional kernel space by kernel principal component analysis (KPCA), and then nonparametric discriminant analysis (NDA) is used to extract the view-invariant shape features as the input for pattern classification. The algorithm is performed on a public dataset, and shows better view-invariant performance than other state-of-the-art methods. |
Year | DOI | Venue |
---|---|---|
2014 | 10.1109/ICPR.2014.104 | ICPR |
Keywords | Field | DocType |
view-invariant shape feature extraction,human computer interaction,gesture and behavior analysis, human computer interaction, motion, tracking and video analysis,euclidean distances,pattern classification,motion,tracking and video analysis,nonparametric discriminant analysis,point-to-point distances description,nda,high-dimensional kernel space,3d motion trajectory,nonparametric shape descriptor,feature extraction,image classification,gesture recognition,public dataset,view-invariant gesture recognition,principal component analysis,kernel principal component analysis,kpca,gesture and behavior analysis,image motion analysis | Kernel (linear algebra),Computer vision,Pattern recognition,Gesture,Computer science,Gesture recognition,Feature extraction,Kernel principal component analysis,Nonparametric statistics,Artificial intelligence,Invariant (mathematics),Trajectory | Conference |
ISSN | Citations | PageRank |
1051-4651 | 3 | 0.39 |
References | Authors | |
13 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Xingyu Wu | 1 | 15 | 2.28 |
Xia Mao | 2 | 188 | 21.89 |
Lijiang Chen | 3 | 304 | 23.22 |
Yu-Li Xue | 4 | 83 | 8.29 |
Angelo Compare | 5 | 29 | 3.81 |