Video Summarization using Text Subjectivity Classification

Abstract

Video summarization has attracted researchers’ attention because it provides a compact and informative video version, supporting users and systems to save efforts in searching and understanding content of interest. Current techniques employ different strategies to select which video segments should be included in the final summary. The challenge is to process multimodal data present in the video looking for relevance clues (like redundant or complementary information) that help make a decision. A recent strategy is to use subjectivity detection. The presence or the absence of subjectivity can be explored as a relevance clue, helping to bring video summaries closer to the final user’s expectations. However, despite this potential, there is a gap on how to capture subjectivity information from videos. This paper investigates video summarization through subjectivity classification from video transcripts. This approach requires dealing with recent challenges that are important in video summarization tasks, such as detecting subjectivity in different languages and across multiple domains. We propose a multilingual machine learning model trained to deal with subjectivity classification in multiple domains. An experimental evaluation with different benchmark datasets indicates that our multilingual and multi-domain method achieves competitive results, even compared to language-specific models. Furthermore, such a model can be used to provide subjectivity as a content selection criterion in the video summarization task, filtering out segments that are not relevant to a video domain of interest.