首页 | 本学科首页   官方微博 | 高级检索  
     

多媒体推理的概念框架
引用本文:庄越挺,潘云鹤,潘红.多媒体推理的概念框架[J].计算机科学,1999,26(6):15-19.
作者姓名:庄越挺  潘云鹤  潘红
作者单位:[1]浙江大学人工智能研究所,杭州310027 [2]杭州师范学院计算机系
基金项目:国家自然科学基金,浙江省科技计划项目
摘    要:Up to now, AI technology is dominated by the Physical Symbolic System(PSS), in which symbolic information is used as the medium for reasoning, In these approaches, information other than symbols, such as image, graphics, and even video should first be represented by symbols, and after reasoning, the symbolic result is again changed into its original media form. In this paper, we will propose a new form of reasoning method called multimedia reasoning (MR), a kind of reasoning that is based on the different media such as text, image, video, audio and so on. By introducing the concept of multimedia transformation theory (MTT), it presents a conceptual framework for multimedia reasoning, In the end, it discusses the importance and potentials in applications.

关 键 词:多媒体推理  人工智能  物理符号系统  概念框架  媒体转换理论
本文献已被 CNKI 维普 等数据库收录!
点击此处可从《计算机科学》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号