Blind motion-compensated video watermarking

The temporal correlation between adjacent video frames poses a severe challenges for video watermarking applications. Motion-coherent watermarking has been recognized as a strategy to embed watermark information in video frames, resistant to collusion attacks. The motion-compensated temporal wavelet transform (MC-TWT) provides an efficient tool to separate static and dynamic components of a video scene and enables motion-coherent water-marking. In this paper, we extend a MC-TWT domain watermarking scheme with blind detection, i.e. motion estimation and watermark detection is performed without reference to the unwatermarked video. Our results show that motion-coherent watermarking can be combined with a blind detector, widening the applicability of MC-TWT domain watermarking beyond forensics (where the unwatermarked content is assumed to be available).