Separating background and foreground optical flow fields by low-rank and sparse regularization
Separating background and foreground optical flow fields by low-rank and sparse regularization
复制标题
DOI：
10.1109/icassp.2015.7178225
复制
发表时间：
2015-04
期刊：
影响因子：
--
通讯作者：
T. Sakai;Hiroki Kuhara
中科院分区：
文献类型：
--
作者：
T. Sakai;Hiroki Kuhara
提出了一种分离由观察者自身运动和物体运动分别引起的背景光流场和前景光流场的方法。光流是由连续图像计算得到的瞬时视运动矢量场。光流场可以假设为由平移和旋转自运动引起的几个基场的线性组合，以及由运动物体引起的空间稀疏光流场。我们将二维光流向量表示为复数，并将场叠加为复矩阵的列。低秩分量自然对应于自我情绪背景光流场，稀疏分量捕获移动的前景物体。通过对复矩阵的鲁棒主成分分析，我们成功地从光流序列中提取了这些成分。
We present a method for separating background and foreground optical flow fields induced by observer's egomotion and motion of objects, respectively. Optical flow is a vector field of instantaneous apparent motion computed from successive images. An optical flow field can be assumed as a linear combination with a few basis fields caused by translational and rotational egomotion and a spatially sparse optical flow field by the moving objects. We represent two-dimensional optical flow vectors as complex numbers and stack the fields as columns of a complex matrix. The low-rank component naturally corresponds to the egomotional background optical flow fields and the sparse component captures the moving foreground objects. We show that these components are successfully extracted from optical flow sequences by the robust PCA applied to the complex matrix.