Semantic knowledge extraction incorporated with syntactic information in personal image collection개인화된 영상 데이터베이스의 신택틱 정보를 활용하여 의미 정보를 추출하는 방법에 관한 연구

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 570
  • Download : 0
Automated understanding and reasoning of semantic knowledge of multimedia are an intrinsic problem in the area of multimedia information analysis. Their ultimate goal is to come up with integral computational model of semantic knowledge understanding and reasoning in view of human visual perception. Multimedia contents are equivalent to human sensory data. Even though understanding and reasoning on those sensory data are innate skills in human beings, these become extremely difficult tasks for multimedia information analysis in analogous way. As digital multimedia is getting popular, a great amount of researches has been done for enhanced functionalities in specific applications. In particular, lots of archiving technologies have been raised for managing, organizing, and searching desired images in a fast and convenient way. In this thesis, we propose a schematic model for representation and understanding of semantic knowledge in personal image collection, mostly focused on photographs. The main contribution is that multiple types and levels of semantic knowledge of photos is represented and modeled with a variety of syntactic and semantic features that are from photo-acquisition devices, and also they are understood by aids of temporal and spatial context within a photo collection. Herein, the major types of photo semantic knowledge are defined as category, situation or event, and person-identity semantics, each of which is separately modeled by different methodology, but the basic idea behind the proposed semantic photo albuming method is semantic knowledge discovery incorporated with syntactic information. The semantic knowledge in photo is related to contexts in view of temporal and spatial domains. The detailed contributions are presented as follow: (1) a schematic model for representation and understanding semantic knowledge of photo in a semantic hierarchy; (2) enhanced photo categorization approach that integrates local and global semantic detection ...
Advisors
Ro, Yong-Manresearcher노용만researcher
Description
한국정보통신대학교 : 공학부,
Publisher
한국정보통신대학교
Issue Date
2007
Identifier
392806/225023 / 020025902
Language
eng
Description

학위논문(박사) - 한국정보통신대학교 : 공학부, 2007.2, [ xi, 172 p. ]

Keywords

Camera metadata; Content-based feature; Semantic knowledge extraction; Photo indexing; 사진 색인; 카메라 메타데이터; 내용 기반 특징; 의미 정보 추출

URI
http://hdl.handle.net/10203/54584
Link
http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=392806&flag=dissertation
Appears in Collection
School of Engineering-Theses_Ph.D(공학부 박사논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0