Extracting Salient Keywords from Instructional Videos Using Joint Text, Audio and Visual Cues

This paper presents a multi-modal featurebased system for extracting salient keywords from transcripts of instructional videos. Specifically, we propose to extract domain-specific keywords for videos by integrating various cues from linguistic and statistical knowledge, as well as derived sound classes and characteristic visual content types. The acquisition of such salient keywords will facilitate video indexing and browsing, and significantly improve the quality of current video search engines. Experiments on four government instructional videos show that 82% of the salient keywords appear in the top 50% of the highly ranked keywords. In addition, the audiovisual cues improve precision and recall by 1.1% and 1.5% respectively.