Making sense of video content

Authors:
A. Viranga Ratnaike;Bala Srinivasan;Surya Nepal
Affiliations:
Monash University, Caulfield, Victoria, Australia;Monash University, Caulfield, Victoria, Australia;CSIRO, North Ryde, NSW, Australia
Venue:
MULTIMEDIA '03 Proceedings of the eleventh ACM international conference on Multimedia
Year:
2003

Citing 3
Cited 0

The spatial semantic hierarchy

Artificial Intelligence
Computational Media Aesthetics: Finding Meaning Beautiful

IEEE MultiMedia
A Design Pattern-Based Video Summarization Technique: Moving from Low-Level Signals to High-Level Structure

HICSS '00 Proceedings of the 33rd Hawaii International Conference on System Sciences-Volume 3 - Volume 3

Quantified Score

Hi-index	0.00

Visualization

Abstract

Our aim in this research is to make sense of scenes in video. We expect this will enable us to identify different scenes sharing the same semantic, even if they do not share any multimedia cues. Our approach is based on emergence, and involves classification and reasoning. We use patterns of cues to synthesize semantic classifications. These classifications need to be consistent with the observed cues, and suggest other elements that might be present. The suggestions might be based on sets of related patterns, ontology, or other knowledge bases. With an idea of what we are looking for, we can re-examine the scenes for multimedia elements, which support our hypothesis, or at least are not inconsistent. We expect that sufficiently detailed semantic descriptions can be generated, by cycling through these steps.