In this paper, we describe an audio fingerprinting-based commercial monitoring system for TV broadcasting. The goal of the commercial monitoring is to identify commercials being broadcasted and find out the time duration. If there are similar commercials in the fingerprint database, we sometimes have trouble in determining which commercial has been broadcasted. To solve this problem, we propose the partial distance comparison method which identifies commercials and discriminates them from background music. Experimental results show that the proposed method improves commercial-identification performance by discriminating different versions of the same commercial. © IFIP International Federation for Information Processing 2007.
CITATION STYLE
Lee, S., & Seo, J. S. (2007). A TV commercial monitoring system using audio fingerprinting. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4740 LNCS, pp. 454–457). Springer Verlag. https://doi.org/10.1007/978-3-540-74873-1_59
Mendeley helps you to discover research relevant for your work.