יום שלישי, 10.4.2018, 11:30
חדר 337, בניין טאוב למדעי המחשב
Video scene detection is the task of temporally dividing a video into its semantic sections called "scenes" - a series of video shots depicting a high-level concept or story (action/drama scene, news segment etc.). This is an important preliminary step for effective analysis of heterogeneous video content, and can help with building a table-of-contents, enabling fast browsing and skipping between scenes, and identifying the contextual boundaries of the content in the video. In this talk we will present our approach to video scene detection in the form of optimal sequential grouping - formulating the problem as a generic optimization problem and solving it efficiently. Our method outperforms the state of the art, and additionally has inherent advantages such as being parameter free and general thus being applicable to other domains as well, such as audio segmentation, change point detection, and even document parsing.