Semantic context detection based on hierarchical audio models

Wen-Huang Cheng, Wei Ta Chu, Ja Ling Wu

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

67 Scopus citations

Abstract

Semantic context detection is one of the key techniques to facilitate efficient multimedia retrieval. Semantic context is a scene that completely represents a meaningful information segment to human beings. In this paper, we propose a novel hierarchical approach that models the statistical characteristics of several audio events, over a time series, to accomplish semantic context detection. The approach consists of two stages: audio event and semantic context detections. HMMs are used to model basic audio events, and event detection is performed in the first stage. Then semantic context detection is achieved based on Gaussian mixture models, which model the correlations among several audio events temporally. With this framework, we bridge the gaps between low-level features and the semantic contexts that last in a time series. The experimental evaluations indicate that the approach is effective in detecting high-level semantics.

Original languageEnglish
Title of host publicationProceedings of the 5th ACM SIGMM International Workshop on Multimedia Information Retrieval, MIR 2003
PublisherAssociation for Computing Machinery, Inc
Pages109-115
Number of pages7
ISBN (Electronic)1581137788, 9781581137781
DOIs
StatePublished - 7 Nov 2003
Event5th ACM SIGMM International Workshop on Multimedia Information Retrieval, MIR 2003 - Berkeley, United States
Duration: 7 Nov 2003 → …

Publication series

NameProceedings of the 5th ACM SIGMM International Workshop on Multimedia Information Retrieval, MIR 2003

Conference

Conference5th ACM SIGMM International Workshop on Multimedia Information Retrieval, MIR 2003
CountryUnited States
CityBerkeley
Period7/11/03 → …

Keywords

  • Audio content analysis
  • Audio retrieval
  • Gaussian mixture model
  • Hidden markov model
  • Semantic context

Fingerprint Dive into the research topics of 'Semantic context detection based on hierarchical audio models'. Together they form a unique fingerprint.

Cite this