Abstract | ||
---|---|---|
In person tracking applications involving multiple cameras, person re-identification is an important step for ensuring accurate tracking of individuals as they move between camera views. However, changes in camera parameters and environmental conditions can make re-identification challenging. This is especially difficult in resource-constrained environments, as is often the case in many real-world intelligent applications. In this paper, we explore dimensionality reduction, metric learning, and classification for achieving re-identification in a computationally efficient way. We report that the covariance metric transformation is a sufficient distance metric for achieving good linear separability between identity classes, and produces better results than more complex approaches across two re-identification datasets. We also explore one-shot learning methods for performing classification, show that our Sequential k-Means algorithm outperforms other fast one-shot learning approaches, and discuss parameter tuning to improve accuracy. |
Year | DOI | Venue |
---|---|---|
2019 | 10.1016/j.compeleceng.2019.06.009 | Computers & Electrical Engineering |
Keywords | Field | DocType |
Computer vision,Person tracking,Person re-identification,Metric learning,Classification,Camera surveillance | Linear separability,Computer vision,Dimensionality reduction,Multi camera,Computer science,Metric (mathematics),Real-time computing,Artificial intelligence,Covariance | Journal |
Volume | ISSN | Citations |
77 | 0045-7906 | 1 |
PageRank | References | Authors |
0.35 | 0 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Andrew Y. Chen | 1 | 25 | 7.15 |
Morteza Biglari-Abhari | 2 | 100 | 19.47 |
Kevin I-Kai Wang | 3 | 167 | 29.65 |