METHOD AND APPARATUS FOR SUMMARIZING AND INDEXING THE CONTENTS OF AN AUDIO-VISUAL PRESENTATION

摘要

A method and an apparatus for summarizing and indexing the contents of an audio-visual presentation are provided to summarize the contents automatically in real time. A pixel related to a current frame of image data is identified. A three-dimensional neighbor is generated and neighbor pixels around the pixel related to the current frame are defined. Distances between the pixel related thereto and each pixel related to the neighbor pixels are compared to each other in order to determine the shortest distance. It is determined whether the pixel related to the current frame belongs to a current segment of the image data or not on a basis of the shortest distance.