Efficient Visual Saliency Detection in Video Based on Trajectory Clustering

DOI

10.2991/icmii-15.2015.84How to use a DOI?

Keywords

Saliency Detection, Trajectories Classification, Cluster Algorithm

Abstract

In this paper, we propose an efficient visual saliency detection method based on trajectory clustering. We group the corner point trajectories using a two stage clustering algorithm. The most stable trajectories are pre-clustered using mean shift in the first stage. Then, we proposed an unsupervised clustering method to cluster the trajectories and detect the number of motions automatically. At last, the motion saliency map is generated with the segmented spare feature points. Experimental results on different videos demonstrate the utility and performance of the proposed approach.

Copyright

© 2015, 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/).

Download article (PDF)

TY  - CONF
AU  - Man Hua
AU  - Ruichun Lin
PY  - 2015/10
DA  - 2015/10
TI  - Efficient Visual Saliency Detection in Video Based on Trajectory Clustering
BT  - Proceedings of the 3rd International Conference on Mechatronics and Industrial Informatics
PB  - Atlantis Press
SP  - 493
EP  - 497
SN  - 2352-538X
UR  - https://doi.org/10.2991/icmii-15.2015.84
DO  - 10.2991/icmii-15.2015.84
ID  - Hua2015/10
ER  -