知识组织

面向事件的视频语义表示方法

  • 李旭晖 ,
  • 吴青峰
展开
  • 武汉大学信息管理学院, 武汉, 430072
李旭晖(ORCID:0000-0002-1155-3597),副教授,硕士生导师,E-mail:lixuhui@whu.edu.cn;吴青峰(ORCID:0000-0002-3967-3187),硕士研究生。

收稿日期: 2019-12-04

  修回日期: 2020-02-23

  网络出版日期: 2020-05-20

基金资助

本文系国家自然科学基金重大研究计划"大数据驱动的管理与决策研究"重点支持项目"基于知识关联的金融大数据价值分析、发现及协同创造机制"(项目编号:91646206)研究成果之一。

Research on Video Semantic Representation for Events

  • Li Xuhui ,
  • Wu Qingfeng
Expand
  • School of Information Management, Wuhan University, Wuhan 430072

Received date: 2019-12-04

  Revised date: 2020-02-23

  Online published: 2020-05-20

摘要

[目的/意义] 视频内容正在影响着我国大量人口的信息生活,视频语义的良好表示是推动当前视频内容研究和视频应用服务向前发展的关键基础。现有的视频语义表示方法存在事件语义表示角度和粒度划分方式单一、缺少灵活的对象语义变化机制的问题,因此探究更有效的视频语义表示方法具有重要意义。[方法/过程] 提出面向事件的视频语义表示方法。此方法考虑人的双向认知过程,可以根据不同用户背景和需求从不同角度解读和生成事件语义,并定义相应的语义对象和角色的变化机制。[结果/结论] 面向事件的视频语义表示方法具有完整的语义表示框架,支持多角度的事件语义表示,可以灵活地进行属性级、对象级和事件级的语义拓展,能够表示更丰富的视频语义。

本文引用格式

李旭晖 , 吴青峰 . 面向事件的视频语义表示方法[J]. 图书情报工作, 2020 , 64(10) : 99 -108 . DOI: 10.13266/j.issn.0252-3116.2020.10.011

Abstract

[Purpose/significance] Video content is affecting the information life of a large number of people in China. The proper representation of video semantic is the key foundation for the current development of video content research and application. The existing methods of semantic representation of video only support the semantic representation of an event from one single perspective and lack the flexible change mechanism of relevant semantic objects, which results in insufficient semantic representation. So it is important to explore more effective video semantic representation methods. [Method/process] This paper proposed a video semantic representation method for events. This method considered the bidirectional nature of human cognitive processes and adopted a scalable way to support multi-perspective interpretation of event semantic. A change mechanism of number and semantic is designed to support relevant objects included in events. [Result/conclusion] This method has a complete semantic representation framework, which can effectively support multi-perspective interpretation of video events. It flexibly supports attribute-level, object-level, and event-level semantic extensions. Generally it can represent richer video semantics than existing methods.

参考文献

[1] CNNIC互联网研究. 第43次CNNIC中国互联网报告发布[J]. 中国广播, 2019(4):48.
[2] 邓珞华, 邓东宁, 陈晟. 论视频图书馆的建设[J]. 大学图书馆学报, 2010, 28(2):70-73.
[3] 赵琨. 大数据环境下图书馆音视频资源发展及建设研究[J]. 图书馆建设, 2015, 248(2):64-68.
[4] 朱智贤.现代认知心理学评述[J].北京师范大学学报,1985(1):1-6.
[5] 曹刘彬.基于深度学习的图像及视频描述方法研究[D].太原:山西大学,2018.
[6] 周教生.基于隐含语义分析的视频语义概念检测方法[J].信息通信,2018(2):141-143.
[7] 陈晨.基于动作语义关联规则挖掘的视频分类研究[D].镇江:江苏大学,2018.
[8] VIJAYAKUMAR V, NEDUNCHEZHIAN R. Mining video association rules based on weighted temporal concepts[J]. ProQuest, 2012,9(4):297-303.
[9] LI G R, ZHANG W G, PANG J B, et al. Online web video topic detection and tracking with semi-supervised learning[J]. Multimedia systems,2016,22(1):115-125.
[10] 栾悉道,谢毓湘,韩智广,等.新闻视频挖掘技术研究[J].计算机科学,2007,34(2):1-6.
[11] 王硕.篮球视频精彩事件检测方法研究[D].西安:西安电子科技大学,2015.
[12] GUPTA A, WEYMOUTH T, JAIN R. Semantic queries with pictures:the VIMSYS model[C]//Proceedings of the seventeenth international conference on Very Large Data Bases.San Francisco:Morgan Kaufmann,1991:69-79.
[13] 王昊冉,白亮,老松杨.基于图模型的足球视频语义建模方法[J].计算机科学, 2011,38(6):266-269,297.
[14] 张静,高伟,刘安安, 等.基于运动轨迹的视频语义事件建模方法[J].电子测量技术,2013,36(9):31-36,40.
[15] 刘晓璐.基于知识元的安防视频内容场景化表示及检索[D].大连:大连理工大学,2017.
[16] 谢潇,朱庆,张叶廷, 等.多层次地理视频语义模型[J].测绘学报,2015(5):555-562.
[17] ADALI S, CANDAN K, CHEN S S, et al. The advanced video information system:data structures and query processing[J]. Multimedia systems, 1996, 4(4):172-186.
[18] TUSCH R, KOSCH H, BOSZORMENYI L. VIDEX:an integrated generic video indexing approach[C]//ACM international conference on multimedia. Los Angeles:ACM,2000:448-451.
[19] 刘宏哲,鲍泓,须德.基于内容的视频分层语义联想模型[J].计算机应用, 2005,25(8):1797-1800.
[20] WANG Y, XING C X, ZHOU L Z. THVDM:a data model for video management in digital library[C]//Proceedings of the sixth international conference of Asian digital libraries. Berlin:Springer International Publishing, 2003:178-192.
[21] ALLEN J F. Maintaining knowledge about temporal intervals[J]. Readings in qualitative reasoning about physical systems, 1990, 26(11):361-372.
[22] LI J Z, OZSU M T, SZAFRON D. Modeling video temporal relationships in an object database management system[C]//Proceedings of the multimedia computing and networking. San Jose:SPIE,1997:80-91.
[23] EGENHOFER, MAX J, FRANZOSA, ROBERT D. Point-set topological spatial relations[J]. International journal of geographical information science, 1991, 5(2):161-174.
[24] 朱旭.挖掘短视频信息传播优势强化大学生意识形态教育[J].才智,2019(24):77.
[25] KILPATRICK C, STORR J, LIM K, et al. Exploring the use of entertainment-education YouTube videos focused on infection prevention and control[J]. American journal of infection control,2018,46(11):1218-1223.
[26] 高士杰,吴丽丽,郭宸.移动短视频广告创作与消费者心理研究[J].中国市场, 2019,(2):139-140.
[27] 陈春,李娜,马建霞.国外图书馆非文本资源建设与服务现状分析及对我国的启示[J].图书情报工作,2015,59(10):53-59.
文章导航

/