Audiovisual production invariant searching

Information searching in non-textual media is a fundamental point of interest, especially in the audiovisual industry where there is still an important need of tools for manipulating multimedia contents. In video documents, the style signature extraction is a highly interesting process since it provides a new feature for contents classification. Video documents may have very different characteristics and properties. However, we all agree that there are some common points between all political programs, or all football matches, or, time to time, between all the movies realized by a given director. These common points are what we call "invariants". An "invariant of production" characterizes a document or a set of documents belonging to a same "collection", of a same director, or produced following the same set of guidelines. In this paper, we present a hypothetical definition of what we call production invariant in a video segment. We propose an algorithm for the invariant segment extraction, applicable on all video features independently with the feature nature and meaning, and with this invariant length or type. MOTS-CLÉS : indexation audiovisuelle, comparaison de caractéristiques, extraction de caractéristiques vidéo.

[1]  Jean-Gabriel Ganascia.  Extraction of Syntactical Patterns from Parsing Trees , 2002 .

[2]  Eamonn J. Keogh,et al.  A Probabilistic Approach to Fast Pattern Matching in Time Series Databases , 1997, KDD.

[3]  Antonin Guttman,et al.  R-trees: a dynamic index structure for spatial searching , 1984, SIGMOD '84.

[4]  Nasser Yazdani,et al.  Matching and indexing sequences of different lengths , 1997, CIKM '97.

[5]  John C. Curlander,et al.  ψ-s correlation and dynamic time warping: two methods for tracking ice floes in SAR images , 1991, IEEE Trans. Geosci. Remote. Sens..

[6]  D. Guégan.  Séries chronologiques non linéaires à temps discret , 1994 .

[7]  Itheri Yahiaoui.  Construction automatique de résumés vidéos , 2003 .

[8]  Dimitrios Gunopulos,et al.  Time-series similarity problems and well-separated geometric sets , 1997, SCG '97.

[9]  Dimitrios Gunopulos.  Time Series Similarity Measures , 2005 .

[10]  S. Chiba,et al.  Dynamic programming algorithm optimization for spoken word recognition , 1978 .

[11]  Donald J. Berndt,et al.  Using Dynamic Time Warping to Find Patterns in Time Series , 1994, KDD Workshop.

[12]  Kyuseok Shim,et al.  Fast Similarity Search in the Presence of Noise, Scaling, and Translation in Time-Series Databases , 1995, VLDB.

[13]  Riccardo Leonardi,et al.  Comparing the quality of multiple descriptions of multimedia documents , 2002, 2002 IEEE Workshop on Multimedia Signal Processing..

[14]  David Sankoff,et al.  Time Warps, String Edits, and Macromolecules: The Theory and Practice of Sequence Comparison , 1983 .

[15]  E. Weisstein.  CRC Concise Encyclopedia of Mathematics, Second Edition , 2002 .