ABSTRACT
In this chapter we review the current state of the art in automatic video summarization. Summarization, be it for a video, audio, or text document, is a challenging and ill-defined task since it requires the processing system to make decisions based on high-level notions such as the semantic content and relative importance of the parts of the documents with respect to each other. Evaluating resulting summaries is also a problem since it is hard to derive quantitative measures of summary quality. We will examine some of the approaches that have been proposed to deal with these problems.
