摘要
视频语义检索的研究是目前研究的热点之一。现有的视频检索系统技术多是基于底层特征的、非语义层次的检索。与人类思维中所能理解的高层语义概念相去甚远,这严重影响视频检索的实际效果。如何跨越底层特征和高层语义的鸿沟,用高层语义概念进行视频检索是当前研究的重点。通过对视频内容的语义理解、语义分析、语义提取的简要概述,试图构造一种视频语义检索模型。
Video semantic retrieval is one of the most popular search issue in video retrieval today.Most video retrieval techniques are low-level feature based and no-semantic.These feature are abstract and quite different from the semantic concepts in human thought.To go beyond low-level similarity and access video data content by semantics,how can we bridge the gap between the low-level features and high-level semantics.How can we develop the model of video semantic retrieval.In this paper,semantic video understand,semantic video analysis,semantic video extract are discussed,in order to design a model of semantic video retrieval.
出处
《计算机工程与应用》
CSCD
北大核心
2007年第18期168-170,180,共4页
Computer Engineering and Applications
基金
国家自然科学基金(the National Natural Science Foundation of China under Grant No.70503022) 。
关键词
高层语义
基于高层语义的视频检索
支持向量机
视频语义检索模型
high-level semantic
video retrieval using high-level semantic
Support Vector Machines(SVM)
model of video semantic retrieval