论文信息 - A Novel Method for Spoken Text Feature Extraction in Semantic Video Retrieval

A Novel Method for Spoken Text Feature Extraction in Semantic Video Retrieval

We propose a novel method for extracting text feature from the automatic speech recognition (ASR) results in semantic video retrieval. We combine HowNet-rule-based knowledge with statistic information to build special concept lexicons, which can rapidly narrow the vocabulary and improve the retrieval precision. Furthermore, we use the term precision (TP) weighting method to analyze ASR texts. This weighting method is sensitive to the sparse but important terms in the relevant documents. Experiments show that the proposed method is effective for semantic video retrieval.

[1] Djoerd Hiemstra,et al. A probabilistic justification for using tf×idf term weighting in information retrieval , 2000, International Journal on Digital Libraries.

[2] Stephen E. Robertson,et al. Some simple effective approximations to the 2-Poisson model for probabilistic weighted retrieval , 1994, SIGIR '94.

[3] Tobun Dorbin Ng,et al. Video retrieval using speech and image information , 2003, IS&T/SPIE Electronic Imaging.

[4] Gang Wang,et al. TRECVID 2004 Search and Feature Extraction Task by NUS PRIS , 2004, TRECVID.

[5] Clement T. Yu,et al. Term Weighting in Information Retrieval Using the Term Precision Model , 1982, JACM.

[6] M. F. Porter,et al. An algorithm for suffix stripping , 1997 .

[7] Stephen E. Robertson,et al. Microsoft Cambridge at TREC-10: Filtering and Web Tracks , 2001, TREC.

[8] Dennis Koelma,et al. The MediaMill TRECVID 2008 Semantic Video Search Engine , 2008, TRECVID.

[9] John R. Smith,et al. IBM Research TRECVID-2009 Video Retrieval System , 2009, TRECVID.