Shot boundary detection based on Image correlation features in video

TitleShot boundary detection based on Image correlation features in video
Publication TypeConference Papers
Year of Publication2004
AuthorsLuo M, DeMenthon D, Doermann D
Conference NameTREC-VID: Text Retrieval and Evaluation Conference
Date Published2004///
Abstract

This paper introduced a framework of shot boundary detection used by the LAMPLAB, University of Maryland, College Park. It deals with cuts and dissolves detection. Generally, it is based on image correlation features in the videos. The cut detection is based on the so-called 2max ratio criterion in a sequential image buffer. The dissolve detection is based on the skipping image difference and linearity error in a sequential image buffer. We also eliminated the effect of flashlights in our algorithms. We submitted 5 runs to TRECVID 2004. Their run-ids are “D0, D1, … , D4”. They use the same algorithms with different parameter settings. These runs show a tradeoff curve between precision and recall on the testing data of TRECVID 2004.