Abstract | ||
---|---|---|
With the widespread use of smart phones and tablet computers, it is necessary to develop algorithms to assist high throughout analysis of mobile videos. A novel method for automated segmentation on the mobile video scenery is proposed in this paper. It uses the kernelized random walks on the globe KNN graph and the Smooth Representation Clustering to improve the segmentation effectiveness. The high order transition probability matrix of the kernelized random walks is utilized for erasing the unreliable edge of the graph. Simultaneously kernel approach is used to assign different weights for neighbors to evaluate their contribution to the clustering. The method is evaluated on two public datasets and a real-world mobile video taken by a smart phone. The experimental results show that the proposed algorithm achieves better performance compared with the other representative algorithms. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1007/978-3-319-69781-9_1 | WEB AND BIG DATA |
Keywords | Field | DocType |
Subspace clustering, Smooth Representation Clustering, The random walks, KNN graph | Kernel (linear algebra),Data mining,Graph,Subspace clustering,Stochastic matrix,Computer science,Random walk,Segmentation,Cluster analysis,Smart phone | Conference |
Volume | ISSN | Citations |
10612 | 0302-9743 | 1 |
PageRank | References | Authors |
0.36 | 15 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Liping Chen | 1 | 60 | 10.10 |
Gongde Guo | 2 | 1 | 1.71 |
Lifei Chen | 3 | 25 | 4.98 |