实验结果表明，该算法可有效地提取视频对象和分割视频对象平面。
Experiments show that the described algorithm can extract video objects, leading to the segmentation of VOPs.
最后，利用边界图像模型从序列中提取出视频对象平面（VOP） 。
Finally, the VOPs are extracted from the edge image model sequence.
它把图像序列分解成视频对象平面(VOP)，每个VOP代表一个运动对象。
It decomposes sequences into video object planes (VOPs) so that each VOP represents one moving object.
由于输入视频序列的每一帧被分割成任意形状的视频对象平面(VOP)，这样每个VOP描述了一个语义意义的对象或所感兴趣的视频内容。
Each frame of the input sequence is segmented into arbitrarily shaped image regions (VOP's) such that each VOP describes one semantically meaningful object or video content of interest.
由于输入视频序列的每一帧被分割成任意形状的视频对象平面(VOP)，这样每个VOP描述了一个语义意义的对象或所感兴趣的视频内容。
Each frame of the input sequence is segmented into arbitrarily shaped image regions (VOP's) such that each VOP describes one semantically meaningful object or video content of interest.
应用推荐