3D-DCT video watermarking using quantization-Based methods

Video watermarking methods operating in the three-dimensional discrete cosine transform (3D-DCT) domain using quantization-based embedding techniques are here presented. Specifically, after having partitioned a video sequence into spatio-temporal units, they are projected onto the 3D-DCT domain. A set of transformed coefficients are taken according to a selection rule, ordered according to a predefined scan order, and eventually marked. Two embedding methods are here considered: Quantization Index Modulation watermarking (QIM) and Rational Dither Modulation (RDM). A perceptual criterion is also used to define the “local” quantization steps to be used in QIM. Finally, the inverse 3D inverse DCT (IDCT) is performed thus obtaining the marked video shot. Robustness of the proposed method against a variety of both intentional and unintentional video attacks has been observed through experimental results.