Abstract
The video segmentation is a fundamental tool for the video semantic content evaluation. In multimedia application, videos are often in MPEG-1 format. In this paper, an algorithm for the automatic shot segmentation of MPEG-1 sequences is presented. The adopted method is based on heuristic considerations concerning the characteristics of MPEG-1 video streams. In particular, the pattern structure and the I-, B- and P-frame sizes are taken in account. The proposed algorithm has been applied to MPEG-1 sequences and some results are reported.
Cite
CITATION STYLE
Ardizzone, E., Lodato, C., & Lopes, S. (1999). Temporal segmentation of MPEG video sequences. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1614, pp. 283–290). Springer Verlag. https://doi.org/10.1007/3-540-48762-x_36
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
