Wipe scene-change detector based on visual rhythm spectrum

Abstract

To store and retrieve large-scale video data sets effectively, the process of wipe detection is an essential step. In this paper, we propose a wipe scene-change detection algorithm based on visual rhythm spectrum (VRS). The VRS contains distinctive patterns or visual features for wipe effects. During a wipe, intensity change between incoming and the outgoing shots gives rise to abrupt intensity discontinuities on the VRS. The proposed algorithm is designed to detect such discontinuities.