In this paper we show how the use of an automatic threshold selection technique can improve some critical video parsing operations. To this end, in the framework of an information-theoretic approach, we propose a technique based on entropy maximization. We also present an algorithm for detecting dissolves in Mpeg compressed videos, which demonstrates the reliability of the proposed technique. Experimental results are provided and discussed.
|Titolo:||Automated Threshold Selection for the Detection of Dissolves in Mpeg Video|
|Data di pubblicazione:||2000|
|Appare nelle tipologie:||4.1.1 Proceedings con DOI|