| 摘要: |
| 针对视频序列中目标的跟踪问题，提出了一种基于粒子滤波框架
的联合仿射和外貌模型的目标跟踪算法。该算法首先提取图像帧之间的相关特征点，通过求
解Sylvester方程得到仿射参数，然后将仿射参数嵌入到基于仿射群的粒子滤波框架中进行
平滑估计。利用基于仿射群的一阶自回归过程模拟状态的变化，联合仿射特征点模型和外貌
模型进行似然估计，得到粒子的最佳平均状态，进而对目标实施跟踪。实验结果表明，在目
标经历姿势和尺度变化、遮挡以及复杂背景等情况下，提出的算法能够有效地跟踪目标，较
之其他相关算法具有很强的鲁棒性。 |
| 关键词: 目标跟踪 Sylvester方程 粒子滤波 仿射群 仿射模型 |
| DOI： |
|
| 基金项目: |
|
| Object tracking based on particle filtering framework joint affine model and appearance model |
| XIANG Zhi-yan,CAO Tie-yong,PAN Jing-feng |
| (Institute of Communications Engineering,PLA University of Science and Technol
ogy,Nanjing 210007,China;Institute of Command Automation,PLA University of Science and Technology,Nanjing 210007,China) |
| Abstract: |
| An algorithm based on particle filtering framework joint affine model and appear
ance model is introduced in this paper for object tracking in video sequences. T
he affine parameters for pose estimation from the corresponding feature points,
which are extracted between two successive images, can be formed as a solution t
o Sylvester′s equation. Then, the affine parameters can be smoothly estimated w
ithin the particle filtering framework based on the affine group. The state dyna
mic is modeled via the first order autoregressive (AR) process on the affine gro
up. And the optimal mean state of particles is estimated through the total likel
ihood function which is a combination of affine feature model and appearance mod
el. Thus, object tracking can be actualized via particle filtering based on the
affine group. Experimental results demonstrate the proposed method is more effec
tive and robust compared with other algorithms when the tracked object undergoes
pose and scale changes, occlusion and complex background. |
| Key words: object tracking sylvester′s equation particle filteri
ng affine group affine model |
《电讯技术》编辑部 