Hua Gang, Liu Zicheng, Zhang Zhengyou, Wu Ying
Microsoft Live Labs, One Microsoft Way, Redmond, WA 98052, USA.
IEEE Trans Pattern Anal Mach Intell. 2006 Oct;28(10):1701-6. doi: 10.1109/TPAMI.2006.209.
We propose a novel global-local variational energy to automatically extract objects of interest from images. Previous formulations only incorporate local region potentials, which are sensitive to incorrectly classified pixels during iteration. We introduce a global likelihood potential to achieve better estimation of the foreground and background models and, thus, better extraction results. Extensive experiments demonstrate its efficacy.
我们提出了一种新颖的全局-局部变分能量，用于从图像中自动提取感兴趣的对象。先前的公式仅包含局部区域势，在迭代过程中这些局部区域势对错误分类的像素很敏感。我们引入了全局似然势，以更好地估计前景和背景模型，从而获得更好的提取结果。大量实验证明了其有效性。
