First-Person Video Summarization.Summarizing first-person videos has attracted the computer vision community in recent years [2,4,18]. Most existing approaches follow a basic workflow consisting of (1) visual feature extraction and (2) keyframe selection or scene segmentation, while the latter is ...
In this paper, we address the problem of jointly summa- rizing large-scale Flickr images and YouTube user videos. Starting from the intuition that the characteristics of the two media are different yet complementary, we develop a fast ...