Name
Abstract | ||
|---|---|---|
In this paper we propose a novel tracking method, which provides accurately segmented object boundaries. The first step of the proposed method is to model the object and background using Gaussian mixture model (GMM), and extract a rough contour according to the object edge features. And then the states of the object, including translation, rotation and scale, are estimated using shape context matching. Finally, we take an elastic shape matching method to extract the exact contour. The proposed method is robust enough for tracking object with translation, rotation, scale change and partial occlusion, and it can also be used for real-time tracking applications. Experiments on both synthetic and real world video sequences demonstrate the effectiveness of the proposed method. |
Year | DOI | Venue |
|---|---|---|
2012 | 10.1016/j.neucom.2011.11.012 | Neurocomputing |
Keywords | Field | DocType |
novel tracking method,object edge feature,gaussian mixture model,rough contour,exact contour,scale change,real-time tracking application,shape context matching,elastic shape,segmented object boundary,object tracking | Active shape model,Computer vision,Pattern recognition,Video tracking,Artificial intelligence,Shape context,Mixture model,Machine learning,Mathematics | Journal |
Volume | ISSN | Citations |
83, | 0925-2312 | 11 |
PageRank | References | Authors |
0.56 | 30 | 4 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Zhao Liu | 1 | 23 | 1.47 |
| Hui Shen | 2 | 134 | 15.32 |
| Guiyu Feng | 3 | 174 | 9.92 |
| Dewen Hu | 4 | 1290 | 101.20 |