首页>
外国专利>
HIERARCHICAL MOTION ESTIMATION AND VIDEO SEGMENTATION IN PRESENCE OF MORE THAN ONE MOVING OBJECT IN A BLOCK
HIERARCHICAL MOTION ESTIMATION AND VIDEO SEGMENTATION IN PRESENCE OF MORE THAN ONE MOVING OBJECT IN A BLOCK
展开▼
机译:在一个块中存在多个移动对象的情况下的分层运动估计和视频分割
展开▼
页面导航
摘要
著录项
相似文献
摘要
In hierarchical motion estimation motion vectors (dv1, dv2, ..., dvN-1) are refined in successive levels of increasing search window pixel density and/or decreasing search window size. Within each hierarchical level, after finding an optimum vector at at least one pixel, based on the optimum vector, the absolute displaced frame difference or displaced frame difference for each pixel of the search window is determined (72, 75, 78). Within the search window, two or more groups of these pixels are determined (72, 75, 78), wherein each of these pixel groups is characterised by a different range of absolute displaced frame difference values or displaced frame difference values for the pixels. A segmentation of the search window into different moving object regions is carried out (72, 75, 78) by forming pixel areas according to the groups, which areas represent a segmentation mask (si1, si2, ..., siN) for the search window. For at least one of the segmentation areas a corresponding motion vector (dvN) is estimated (72, 75, 78).
展开▼
机译:在分层运动估计中,运动矢量(dv 1 Sub>,dv 2 Sub>,...,dv N-1 Sub>)按连续的递增级别进行精炼搜索窗口像素密度和/或减小的搜索窗口大小。在每个分级级别内,在至少一个像素处找到最佳向量之后,基于最佳向量,确定搜索窗口的每个像素的绝对位移帧差或位移帧差(72、75、78)。在搜索窗口内,确定两组或更多组这些像素(72、75、78),其中,这些像素组中的每一个的特征在于像素的绝对位移帧差值或位移帧差值的不同范围。通过根据各组形成像素区域,将搜索窗口分割为不同的运动对象区域(72、75、78),这些区域代表分割掩模(si 1 Sub>,si 2 Sub>,...,si N Sub>)作为搜索窗口。对于分割区域中的至少一个分割区域,估计对应的运动矢量(dv N Sub>)(72、75、78)。
展开▼