1. INTRODUCTION
Digital video content can be easily edited and redistributed. As a result, a high number of identical and near-identical video clips can typically be found in personal video collections and on websites for video sharing. Identical video clips are known as duplicates whereas video clips that have been the subject of at least one transformation are referred to as Near-Duplicate Video Clips (NDVCs) [1]. The detection of NDVCs is a prerequisite for a plethora of multimedia applications, including media usage monitoring, content linking on the Web, metadata propagation for annotation purposes, protection of intellectual property, mitigation of visual redundancy in search results, and management of personal media libraries [2] [3].