Generation of Multimedia Artifacts: An Extractive Summarization-based Approach

TL;DR

This paper presents an extractive summarization approach for generating multimedia artifacts, focusing on content selection and coherence in film tributes and science talks using audio and video.

Contribution

It introduces a novel method combining centrality, diversity, and topic analysis for multimedia content selection and coherence enhancement.

Findings

Effective content selection improves multimedia coherence

Emotion-based and topic similarity techniques enhance artifact quality

Method applicable to diverse multimedia generation scenarios

Abstract

We explore methods for content selection and address the issue of coherence in the context of the generation of multimedia artifacts. We use audio and video to present two case studies: generation of film tributes, and lecture-driven science talks. For content selection, we use centrality-based and diversity-based summarization, along with topic analysis. To establish coherence, we use the emotional content of music, for film tributes, and ensure topic similarity between lectures and documentaries, for science talks. Composition techniques for the production of multimedia artifacts are addressed as a means of organizing content, in order to improve coherence. We discuss our results considering the above aspects.