Files
Abstract
Comparison of video sequences is an important operation in many multimedia information systems. The similarity measure for comparison is based on some measure of correlation with the perceptual similarity (or difference) among video sequences or with the similarity (or difference) in some measure of semantics associated with the video sequences. In content-based similarity analysis, the video data are expressed in terms of different features. The similarity matching is then performed by quantifying the feature relationships between target video and query video shots, with either an individual feature or with a feature combination. In this study, two approaches are proposed for the similarity analysis of video shots. In the first approach, mosaic images are created from video shots, and the similarity analysis is done by examining the similarity amongst the mosaic images. In the second approach, the key frames are extracted for each video shot, and the similarity amongst video shots is examined by comparing the key frames of the video shots. The features extracted include image histograms, slopes, edges, and wavelets. Both individual features and feature combinations are used in similarity matching using an artificial neural network models. The similarity rank of query video shots is determined based on the coefficients of determination and mean absolute errors.