Robust video fingerprinting using local spatio-temporal features

In this paper, we present a robust video fingerprinting scheme based on local spatio-temporal features. The spatio-temporal interest point detector is first exploited to detect local regions in the input video clips. For each local region, contrast content histogram (CCH) is then used to calculate the intensity differences, and a unit vector obtained by normalizing CCH is used as the local fingerprints. The experimental results show that the developed scheme improves the whole performance over the representative video fingerprinting method in terms of robustness.