An approach to detect video frame deletion under anti-forensics

As a simple yet effective operation, frame deletion is widely used in video forgery. Many video forensic techniques have been developed to detect this manipulation. Some inter-frame continuity based methods are capable to detect frame deletion as well as locate the frame deletion points precisely. However, due to the simple principle, this kind of inter-frame continuity based methods are vulnerable to anti-forensic strategies. In this paper, we first put forward an inter-frame interpolation as ananti-forensic operation which is easy to realize. Then, to detect video frame deletion under anti-forensics, the new artifacts introduced by interpolation are analyzed, and we present a global and local joint feature to distinguish the interpolated frames and pristine frames. The feature overcomes the problem of weak residual in HEVC videos, and low dimension ensures the real-time detection. Experimental results show that the anti-forensic view we proposed needs to be considered in frame deletion forensics. In addition, the proposed global and local joint feature can detect frame deletion under anti-forensic operation effectively.