基于压缩域的视频关注对象提取

潘琤雯; 张兆扬; 石旭利; 沈礼权

doi:10.11834/jig.20091039

“第4届全国信号处理联合会议”栏目 | 浏览量 : 0 下载量: 304 CSCD: 0

PDF
导出
分享
收藏
专辑

基于压缩域的视频关注对象提取
An Approach to Video Object Extraction in the Compressed Domain
2009年14卷第10期页码：2145
纸质出版：2009
DOI： 10.11834/jig.20091039
稿件说明：

移动端阅览

潘琤雯, 张兆扬, 石旭利, 沈礼权. 基于压缩域的视频关注对象提取[J]. 中国图象图形学报, 2009,14(10):2145. DOI： 10.11834/jig.20091039.

An Approach to Video Object Extraction in the Compressed Domain[J]. Journal of Image and Graphics, 2009, 14(10): 2145. DOI： 10.11834/jig.20091039.

摘要

针对视频压缩域的对象分割问题，提出了一种视频对象精确提取方法。先将视频对象的颜色空间进行均值偏移和区域生长，得到分块的视频图像。同时对视频编码过程中所产生的运动矢量统计熵值，提取出视频中能引起人眼关注的运动对象。最后利用提出的对象轮廓精确分割方法，提取视频关注对象。实验表明该算法能精确和完整地分割出视频关注对象，并对视频中关注对象的切换、物体的非刚性形变有很好的鲁棒性。

Abstract

In this paper

we propose a new method to segment video object. Firstly

we smooth each frame with the mean-shift and region-growing method. And then

the motion vectors which are released by video coding are used to extract a moving object. Finally

the object segmentation method proposed in this paper is used to extract the video attention object. Experimental results show that the proposed method can effectively separate the video object from the sequence and has the strong robustness in multi-targets tracking especially for switching targets and deformed objects.