Spatiograms versus histograms for region-based tracking
Spatiograms versus histograms for region-based tracking
复制标题
DOI：
10.1109/cvpr.2005.330
复制
发表时间：
2005-06
期刊：
影响因子：
--
通讯作者：
Stan Birchfield;S. Rangarajan
中科院分区：
文献类型：
--
作者：
Stan Birchfield;S. Rangarajan
我们引入空间图的概念，它是包含潜在高阶矩的直方图的概括。直方图是零阶空间图，而二阶空间图包含每个直方图箱的空间均值和协方差。这种空间信息仍然允许相当一般的转换，如直方图，但捕获更丰富的目标描述，以提高跟踪的鲁棒性。我们展示了如何在基于内核的跟踪器中使用空间图，推导均值平移过程，其中各个像素不仅对平移量进行投票，而且还对其方向进行投票。实验表明，使用均值平移和详尽的局部搜索，与直方图相比，跟踪结果得到了改善。
We introduce the concept of a spatiogram, which is a generalization of a histogram that includes potentially higher order moments. A histogram is a zeroth-order spatiogram, while second-order spatiograms contain spatial means and covariances for each histogram bin. This spatial information still allows quite general transformations, as in a histogram, but captures a richer description of the target to increase robustness in tracking. We show how to use spatiograms in kernel-based trackers, deriving a mean shift procedure in which individual pixels vote not only for the amount of shift but also for its direction. Experiments show improved tracking results compared with histograms, using both mean shift and exhaustive local search.