Metrics for shot boundary detection in digital video sequences

The detection of shot boundaries in video sequences is an important task fo r generating indexed video databases. This paper provides a comprehensive q uantitative comparison of the metrics that have been applied to shot bounda ry detection. In addition, several standardized statistical tests that have not been applied to this problem, as well as three new metrics, are consid ered. A mathematical framework for quantitatively comparing metrics is supp lied. Experimental results based on a video database containing 39,000 fram es are included.