Extraction of Key Frames from Videos

<BODY> <H1>Extraction of Key Frames from Videos by Optimal Color Composition Matching and Polygon Simplification</H1> <B>Abstract:</B><BR> A video sequence is first mapped to a sequence of points<BR> in a semi-metric space that forms a polyline.<BR> We require only that a semi-distance<BR> between pairs of points be defined that <BR> need not satisfy the triangle inequality.<BR> By simplifying the polyline, we obtain <BR> a small set of the most relevant key frames<BR> that is representative of the whole video sequence.<BR> The degree of the simplification is either determined automatically<BR> or selected by the user.<BR> Using our technique, a viewer can browse a video at the level of<BR> summarization that suits his patience level.<BR> Applications include the<BR> creation of a smart fast-forward function for digital VCRs, and the automatic<BR> creation of short summaries or trailers<BR> that can be used as previews before videos are downloaded from the web.<BR> <BR> Here we shows some examples and experiments for our article in the preceedings of the<BR> 2001 Workshop on <BR> Multimedia Signal Processing<BR> </BODY>