Fast Frame-Accurate Mining for Repeating Video Clips

  • In the broad range of multimedia content analysis tasks the detection of recurring video sequences plays an important role. We introduce an algorithm for recognizing recurring video sequences frame-accurately in a highly effective and efficient way. It does not require temporal segmentation by shot detection. A 24 hour live-stream can be processed in about 4 hours including the computational expensive video decoding. The algorithm uses an inverted index for identifying similar frames rapidly. Gradient-based image features are mapped to the index by means of a hash function. The search algorithm consists of two steps: firstly searching for recurring short segments (e.g., 1 second long) and secondly assembling these small segments into the set of repeated whole video clips. In our experiments we investigate the sensitivity of the algorithm concerning all system parameters and apply it to the detection of unknown commercials within 24 hours of two different TV channels. It is shown thatIn the broad range of multimedia content analysis tasks the detection of recurring video sequences plays an important role. We introduce an algorithm for recognizing recurring video sequences frame-accurately in a highly effective and efficient way. It does not require temporal segmentation by shot detection. A 24 hour live-stream can be processed in about 4 hours including the computational expensive video decoding. The algorithm uses an inverted index for identifying similar frames rapidly. Gradient-based image features are mapped to the index by means of a hash function. The search algorithm consists of two steps: firstly searching for recurring short segments (e.g., 1 second long) and secondly assembling these small segments into the set of repeated whole video clips. In our experiments we investigate the sensitivity of the algorithm concerning all system parameters and apply it to the detection of unknown commercials within 24 hours of two different TV channels. It is shown that the method is an excellent alternative for searching for unknown commercials.show moreshow less

Download full text files

Export metadata

Statistics

Number of document requests

Additional Services

Share in Twitter Search Google Scholar
Metadaten
Author:Ina Döhring, Rainer LienhartGND
URN:urn:nbn:de:bvb:384-opus4-10143
Frontdoor URLhttps://opus.bibliothek.uni-augsburg.de/opus4/1176
Series (Serial Number):Reports / Technische Berichte der Fakultät für Angewandte Informatik der Universität Augsburg (2008-14)
Type:Report
Language:English
Publishing Institution:Universität Augsburg
Release Date:2008/09/10
Institutes:Fakultät für Angewandte Informatik
Fakultät für Angewandte Informatik / Institut für Informatik
Fakultät für Angewandte Informatik / Institut für Informatik / Lehrstuhl für Maschinelles Lernen und Maschinelles Sehen
Dewey Decimal Classification:0 Informatik, Informationswissenschaft, allgemeine Werke / 00 Informatik, Wissen, Systeme / 004 Datenverarbeitung; Informatik