A Rotation Invariant BSIF Descriptor for Video Copy Detection Using a Ring Decomposition

Recently, with the increasing number of internetusers, detecting non legal copy of video sequence is of outstandingimportance. This paper provides a new approach for contentbased video copy detection (CBVCD) which is invariant torotation and flipping attacks. The proposed scheme is based onbinary statistical image features (BSIF) descriptor using a newring decomposition. The ring partition is particularly suitablefor rotation/flipping attacks that affect the video frames. In fact, the visual content of each ring is kept constant when the videoframes are rotated or flipped. The proposed VCD system wasevaluated under TRECVID 2009 database and compared toothers algorithms based on local binary pattern (LBP), localphase quantization (LPQ) or histogram of oriented gradient(HOG) descriptors. The experimental results demonstrated thatthe proposed descriptor is effective for all the attacks that canaffect a video sequence and particularly in the case of rotationand flipping attacks.