In order to improve the performance of particle filter object tracking algorithm based on two-view geometry
cooperative object tracking algorithm based on epipolar geometry constrained particle filter(EGCPF) is proposed. Using the epipolar geometry constraint
on one hand
the object detecting region is restricted during the tracking process to reduce the search space; on the other hand
the state transfer model of particle filtering algorithm is modified to decrease the total number of particles required for each camera with object in the overlap region. Finally
the frame rate
center error and coverage is introduced to measure the performance of the proposed algorithm. The experimental results showed that EGCPF algorithm met requirement in real-time performance
and the accuracy was improved obviously compared with the classical particle filter based object tracking algorithm.
关键词
Keywords
references
薛明. 基于稀疏表示的在线目标跟踪研究[D]. 上海: 上海交通大学, 2014.
Doucet A, Godsill S, Andrieu C. On sequential Monte Carlo sampling methods for Bayesian filtering[J]. Statistics and computing, 2000, 10(3): 197-208. DOI: 197
Sanjeev A M, Maskell S, Gordon N, et al. A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking[J]. IEEE Transactions on Signal Processing, 2002, 50(2): 174-188. DOI: 174
Ross D A, Lim J, Lin R S, et al. Incremental learning for robust visual tracking[J]. International Journal of Computer Vision, 2008, 77(1): 125-141. DOI: 125