Title | ||
---|---|---|
Fast tracking via spatio-temporal context learning based on multi-color attributes and pca |
Abstract | ||
---|---|---|
At present, the effective tracking of pedestrians is still a challenging task due to factors such as illumination change, pose variation, motion blur and occlusion. In this paper, we propose a simple and effective tracking algorithm which exploits the spatio-temporal context. Based on a existing Bayesian framework, we take full advantage of the relevance of the region of interest to its local context, and model the correlation of the visual characteristics of the target and the surrounding area. Through computing a confidence map and maximizing the likelihood function, our tracker gets the real position of the target. Multi-dimensional variant of color attributes provides superior performance for visual tracking in recent years. Our tracker extracts 11 dimensional color names features from the target. However, considering the real time tracking, we reduce the dimensionality of features from 11 to 3 with PCA algorithm. In order to ensure robustness, we cascade the HSV histogram features. In the absence of optimization, our algorithm runs at more than 80 frames per second implemented in MATLAB. Extensive experimental results show that the proposed algorithm is more accurate and accurate than many state-of-the-art algorithms on multiple datasets. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1109/ICInfA.2017.8078941 | 2017 IEEE International Conference on Information and Automation (ICIA) |
Keywords | Field | DocType |
Spatio-temporal context,color name,HSV histogram,PCA | Histogram,Computer vision,Likelihood function,Pattern recognition,Computer science,Visualization,Motion blur,Robustness (computer science),Feature extraction,Eye tracking,Frame rate,Artificial intelligence | Conference |
ISBN | Citations | PageRank |
978-1-5386-3155-3 | 0 | 0.34 |
References | Authors | |
17 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yixiu Liu | 1 | 3 | 1.75 |
Yunzhou Zhang | 2 | 219 | 30.98 |
Meiyu Hu | 3 | 0 | 0.34 |
Pengju Si | 4 | 0 | 0.34 |
Chongkun Xia | 5 | 2 | 1.76 |