Institutional Repository
| local background-aware target tracking | |
| Chu Jun; Du Li-Hui; Wang Ling-Feng; Pan Chun-Hong | |
| 2012 | |
| Source | Zidonghua Xuebao/Acta Automatica Sinica
 |
| ISSN | 0254-4156 |
| Volume | 38Issue:12Pages:1985-1995 |
| English Abstract | Classical visual tracking methods usually only adopt target information to describe the target. In practice, local background information around the target also influences the tracking process. In this paper, we first introduce the local background information into target description, and represent it as a set of weighted feature points. After that, the posterior probability of each point in the search-region is calculated by incorporating the target observation obtained by K nearest neighbor (KNN) algorithm and the Gaussian prior distribution of the target. Finally, the mean shift algorithm is used to estimate the target state. The proposed method has the following two advantages: 1) The local background information is integrated into the target description, which enhances the target model. Thereby, the discriminative ability is promoted in the tracking process, which further makes the tracker more robust and accurate. 2) In the initialization stage, the mean-shift is applied to relocating the target, which can solve the problem that the tracking algorithm is prone to failure in inexact initialization. Extensive experiments in different video sequences are conducted to evaluate our approach qualitatively and quantitatively. The results show that our method holds high tracking accuracy and stability, especially when the target is roughly initialized.; Classical visual tracking methods usually only adopt target information to describe the target. In practice, local background information around the target also influences the tracking process. In this paper, we first introduce the local background information into target description, and represent it as a set of weighted feature points. After that, the posterior probability of each point in the search-region is calculated by incorporating the target observation obtained by K nearest neighbor (KNN) algorithm and the Gaussian prior distribution of the target. Finally, the mean shift algorithm is used to estimate the target state. The proposed method has the following two advantages: 1) The local background information is integrated into the target description, which enhances the target model. Thereby, the discriminative ability is promoted in the tracking process, which further makes the tracker more robust and accurate. 2) In the initialization stage, the mean-shift is applied to relocating the target, which can solve the problem that the tracking algorithm is prone to failure in inexact initialization. Extensive experiments in different video sequences are conducted to evaluate our approach qualitatively and quantitatively. The results show that our method holds high tracking accuracy and stability, especially when the target is roughly initialized. |
| Indexed Type | EI |
| Keyword | Algorithms Motion Compensation Probability Distributions |
| Department | (1) School of Software Nanchang Hangkong University Nanchang 330063 China; (2) School of Information Engineering Nanchang Hangkong University Nanchang 330063 China; (3) National Laboratory of Pattern Recognition Institute of Automation Chinese Academy of Sciences Beijing 100190 China |
| Language | 中文 |
| Content Type | 期刊论文 |
| URI | http://ir.iscas.ac.cn/handle/311060/15454 |
| Collection | 中国科学院软件研究所 |
| Recommended Citation GB/T 7714 | Chu Jun,Du Li-Hui,Wang Ling-Feng,et al. local background-aware target tracking[J]. Zidonghua Xuebao/Acta Automatica Sinica,2012,38(12):1985-1995. |
| APA | Chu Jun,Du Li-Hui,Wang Ling-Feng,&Pan Chun-Hong.(2012).local background-aware target tracking.Zidonghua Xuebao/Acta Automatica Sinica,38(12),1985-1995. |
| MLA | Chu Jun,et al."local background-aware target tracking".Zidonghua Xuebao/Acta Automatica Sinica 38.12(2012):1985-1995. |
| Files in This Item: | There are no files associated with this item. | |||||
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.
Edit Comment