Skip to Main Content
Video summaries are becoming increasingly popular as a way to convert large chunks of video into smaller, more comprehensible units. Many video summarization techniques rely on finding features in the video or audio streams, using them to identify clips, and then assembling the most salient clips into a shorter version of the original. An alternative approach is to match and take advantage of the known structure of a video sequence to find salient elements, and then match those segments into a summary design pattern. With this techniques, the final product is more elegantly watchable, with fewer artifacts of concatenation, and the final product is usually more content bearing than summaries composed with simple signal analysis techniques.
Date of Conference: 4-7 Jan. 2000