Efficient video copy detection using multi-modality and dynamic path search
Teng Li; Fudong Nian; Xinyu Wu; Qingwei Gao; Yixiang Lu
刊名Multimedia Systems
2014
英文摘要Efficient and robust video copy detection is an important topic for many applications, such as commercial monitoring and social media retrieval. In this paper, with the aim of handling large-scale video data, we propose an efficient and robust video copydetection method jointly utilizing the characteristics of temporal continuity and multi-modality of video. The video is converted to a continuous sequence of states, and both the visual and auditory features are extracted for temporal frames. To facilitate tolerance of the length variations caused during video re-targeting, an efficient dynamic path search method is proposed to detect the target video clips, and highly compact audio fingerprint and visual ordinal features are jointly utilized in a flexible frame. The proposed scheme not only achieves high computational efficiency but also guarantees effectiveness in real applications. Comparison experiments were conducted using video commercials and real television programs from four channels as well as a benchmark video copy detection dataset, and the results demonstrate both the high efficiency and high robustness of the proposed method. 
收录类别EI
原文出处http://link.springer.com/article/10.1007/s00530-014-0387-8
语种英语
WOS记录号WOS:000368828500004
内容类型期刊论文
源URL[http://ir.siat.ac.cn:8080/handle/172644/5447]  
专题深圳先进技术研究院_集成所
作者单位Multimedia Systems
推荐引用方式
GB/T 7714
Teng Li,Fudong Nian,Xinyu Wu,et al. Efficient video copy detection using multi-modality and dynamic path search[J]. Multimedia Systems,2014.
APA Teng Li,Fudong Nian,Xinyu Wu,Qingwei Gao,&Yixiang Lu.(2014).Efficient video copy detection using multi-modality and dynamic path search.Multimedia Systems.
MLA Teng Li,et al."Efficient video copy detection using multi-modality and dynamic path search".Multimedia Systems (2014).
个性服务
查看访问统计
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。


©版权所有 ©2017 CSpace - Powered by CSpace