Online Video Summarization Based on Local Features

Abstract

Digital video has become a very popular media in several contexts, with an ever expanding horizon of applications and uses. Thus, the amount of available video information is growing almost limitless. For this reason, video summarization continues to attract the attention of a wide spectrum of research efforts. In this work, the authors present a novel video summarization technique based on tracking local features among consecutive frames. The authors' approach operates on the uncompressed domain, and requires only a small set of consecutive frames to perform. Additionally, the authors' algorithm processes the video stream directly and produces results on the fly. The authors tested their implementation with standard available datasets, and compared the results with the most recently published work in the field. The achieved results are similar or better in quality than the best performing proposals, with the additional advantage of being able to process the stream directly in the uncompressed domain.