摘要
在多摄像机跟踪中,主要的难点是如何在多个摄像机之间对相同的目标建立正确的对应关系。摄像机视野边界(FOV)线是近年来出现的一种新的有效的解决多摄像机跟踪中目标交接问题的工具。因此,精确生成视野边界线成了解决问题的关键所在。根据投影不变量FOV线生成法,在生成视野边界线时不依赖于场景中目标的运动,并且不需要摄像机参数的信息。应用PETS2001数据库里的两组不同序列进行尝试,实验结果表明,对于视野边界线恢复,其鲁棒性与可靠性都较好。
Establishing correspondences between moving objects is the most difficult problem to tackle in multiple cameras tracking. Field of view （FOV） lines have been introduced in literature as an efficient tool to resolve the camera handoff issue. Therefore, finding the FOV lines precisely becomes the key to the problem. The proposed method of recovering field of view lines by using projective invariants does not rely on the object movement in the scene and does not require information about camera parameters. Experiment results on different sequences obtained from the PETS2001 database show that the robustness of the algorithm in recovering all visible FOV lines of another camera in the current camera view is satisfactory.
出处
《桂林电子科技大学学报》
2007年第3期216-218,共3页
Journal of Guilin University of Electronic Technology
基金
广西科学研究与技术开发计划项目资助(063006-5G-4)
关键词
多摄像机跟踪
目标交接
视野边界线
投影不变量
multiple camera tracking
camera handoff
field of view lines
projective invariants
