H.264/AVC digital fingerprinting based on content adaptive embedding

Abstract

Digital fingerprinting is a technology for tracing the distribution of multimedia content and protecting them from unauthorized redistribution. Unique identification information is embedded into each distributed copy of multimedia signal and serves as a digital fingerprint. This paper presents active fingerprinting which combines robust watermarking and independent Gaussian fingerprints generated by Tardos algorithm to secure the distributed copy of the H.264 /AVC compressed video. This approach is derived from [1], It relies mainly on improving the visual quality by eliminating the perceivable distortion that exists in the fingerprinted copies as well as the copies after collusion attacks produced by using unbounded Gaussian fingerprints. In order to remove this perceptual distortion, in this paper, we propose a content adaptive watermaking method which takes full advantage of both Intra and Inter frames information of the video content. Linear and non linear collusion attacks are performed to show the robustness of the proposed technique against collusion attacks while maintaining visual quality unchanged.