【24h】

A block-based temporal segmentation algorithm for moving pictures

机译:基于块的运动图像时间分割算法

获取原文
获取原文并翻译 | 示例

摘要

In this paper, we propose a block-based temporal segmentation algorithn for the object-based video compression. It consists of three steps: the change detection, the block merge, and the block segmentation. The first step separates and change detected region from background. Here, a new method of removing the uncovered region without motion estimation is presented. The second step, composed of three substeps, estimates motions for the changed detected region and merges blocks with similar motions. Here, we provide merging ocnditions for each substep as criteria. In the final block, the boundary block is segmented on the pixel basis. After describing our algorithm in detial, we present some experimental results step by step along the processing steps. The result tells that the proposed algorithm removes the uncovered region effectively and produces well-segmented objects.
机译:在本文中,我们为基于对象的视频压缩提出了一种基于块的时间分割算法。它包括三个步骤:更改检测,块合并和块分段。第一步是将检测到的区域从背景中分离出来并进行更改。在此,提出了一种无需运动估计即可去除未覆盖区域的新方法。第二步由三个子步骤组成,它估计更改后的检测区域的运动,并合并具有相似运动的块。在这里,我们为每个子步骤提供合并条件。在最后一个块中,边界块是按像素分割的。在详细描述了我们的算法之后,我们将逐步介绍一些实验结果以及处理步骤。结果表明,所提出的算法有效地去除了未覆盖区域并产生了分割良好的对象。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号