Dynamic texture segmentation
Dynamic texture segmentation
复制标题
DOI：
10.1109/iccv.2003.1238632
复制
发表时间：
2003-10
期刊：
影响因子：
--
通讯作者：
Gianfranco Doretto;D. Cremers;P. Favaro;Stefano Soatto
中科院分区：
文献类型：
--
作者：
Gianfranco Doretto;D. Cremers;P. Favaro;Stefano Soatto
我们解决了将自然场景图像序列分割成具有恒定时空统计特征的不相交区域的问题。利用高斯-马尔可夫模型对各区域进行时空动态建模，并在变分优化框架下推导模型参数和区域边界。数值结果表明，与纯粹基于纹理的分割方案相比，即使在空间统计数据相同的情况下，我们的方法也能有效地分割动态不同的区域。
We address the problem of segmenting a sequence of images of natural scenes into disjoint regions that are characterized by constant spatio-temporal statistics. We model the spatio-temporal dynamics in each region by Gauss-Markov models, and infer the model parameters as well as the boundary of the regions in a variational optimization framework. Numerical results demonstrate that - in contrast to purely texture-based segmentation schemes - our method is effective in segmenting regions that differ in their dynamics even when spatial statistics are identical.