Interpretable video representation

The immense amount of available video data poses novel requirements for video representation approaches by means of focusing on central and relevant aspects of the underlying story and facilitating the efficient overview and assessment of the content. In general, the assessment of content relevance and significance is a high-level task that usually requires for human intervention. However, some filming techniques imply importance and bear the potential for automated content-based analysis. For example, core elements in a movie (such as the main characters and central objects) are often emphasized by repeated occurrence. In this paper we present a new approach for the automated detection of such recurring elements in video sequences that provides a compact and interpretable content representation. Performed experiments outline the challenges and the potential of the algorithm for automated high-level video analysis.