作者: Liwei He , Elizabeth Sanocki , Anoop Gupta , Jonathan Grudin
关键词:
摘要: As streaming audio-video technology becomes widespread, there is a dramatic increase in the amount of multimedia content available on net. Users face new challenge: How to examine large amounts quickly. One technique that can enable quick overview video summaries; is, shorter version assembled by picking important segments from original.We evaluate three techniques for automatic creation summaries online presentations. These exploit information audio signal (e.g., pitch and pause information), knowledge slide transition points presentation, about access patterns previous users. We report user study compares automatically generated are 20%-25% length full presentations author summaries. learn computer-generated summaries, although less than authors' They initially find coherent, but quickly grow accustomed them.