Performance Characterization and Comparison of Video Indexing Algorithms

Temporal segmentation of video is a necessary rst step to indexing digital video for browsing and retrieval. A number of diierent video temporal segmen-tation algorithms have been published in the literature. There has been little eeort to evaluate and characterize their performance so as to deliver a single (or set of) algorithms that may be used by other researchers for indexing video databases. We present results of evaluating a number of these algorithms and characterizing their performance, speciically with respect to robustness to encoder and bitrate changes. The lessons learnt have relevance to algorithm development and evaluation in general.