Story-driven summarization for egocentric video

We present a video summarization approach that discov-ers the story of an egocentric video. Given a long input video, our method selects a short chain of video subshots depicting the essential events. Inspired by work in text anal-ysis that links news articles over time, we define a random-walk based metric of influence between subshots that re-flects how visual objects contribute to the progression of events. Using this influence metric, we define an objective for the optimal k-subshot summary. Whereas traditional methods optimize a summary’s diversity or representative-ness, ours explicitly accounts for how one sub-event “leads to ” another—which, critically, captures event connectivity beyond simple object co-occurrence. As a result, our sum-maries provide a better sense of story. We apply our ap-proach to over 12 hours of daily activity video taken from 23 unique camera wearers, and systematically evaluate its quality compared to multiple baselines with 34 human sub-jects. 1