摘要
视频浓缩是包含原视频有效信息的简短表示,以便于视频的存储、浏览和检索。然而,大部分视频浓缩方法得到的浓缩视频中会丢失少量目标,不能完整表达原始视频的全部内容。本文介绍了一种基于目标轨迹优化的视频浓缩方法。首先使用改进的目标轨迹提取算法提取原视频中目标的轨迹,然后利用马尔可夫随机场模型和松弛线性规划算法得到每条轨迹的最优时间标签,将其与背景序列和目标轨迹结合生成浓缩视频。实验结果表明,与传统的视频浓缩方法相比,本文方法生成的浓缩视频具有较高的浓缩比,保证了信息的完整性又具有良好的视觉效果。
Video synopsis is a temporally compact representation of the original video, which facilitates the subsequent video processing, such as video storage, browsing and retrieval. Most of conventional methods easily lose some important objects and can not represent the original videos completely. Therefore, this paper proposes a novel method based on object trajectory optimization. The method extracts object trajectories using an improved multi-object tracking method, and optimizes the temporal shift labels of those trajectories. The optimal labels are then formulated as the maximum a posteriori state of a special Markov random field, which can be solved by the relaxed linear programming method. The synopsis video is obtained by integrating the optimal labels into the background sequence. Extensive experi- ments on both public and collected video sequences suggest that our method outperforms other methods in accuracy. In particular, our method can retain most essential information of the video sources in shorter synopsis videos.
出处
《数据采集与处理》
CSCD
北大核心
2016年第1期108-116,共9页
Journal of Data Acquisition and Processing
基金
国家自然科学基金(61472002
61502006)资助项目
安徽省自然科学基金(1308085MF97)资助项目
关键词
视频浓缩
视频监控
马尔可夫随机场
松弛线性规划
video synopsis
video surveillance
Markov random filed
relaxed linear programming
