Robust and discriminative zero-watermark scheme based on invariant features and similarity-based retrieval to protect large-scale DIBR 3D videos

Abstract

Digital rights management (DRM) of depth-image-based rendering (DIBR) 3D video is an emerging area of research. Existing schemes for DIBR 3D video cause video distortions, are vulnerable to severe signal and geometric attacks, cannot protect 2D frames and depth maps independently, or have difficulty handling large-scale videos. To address these issues, a novel zero-watermark scheme based on invariant features and similarity-based retrieval to protect DIBR 3D video (RZW-SR) is proposed in this study. In RZW-SR, invariant features are extracted to generate master and ownership shares to provide distortion-free, robust and discriminative copyright identification under various attacks. Different from conventional zero-watermark schemes, our proposed scheme stores features and ownership shares correlatively and designs a similarity-based retrieval phase to provide effective solutions for large-scale videos. In addition, flexible mechanisms based on attention-based fusion are designed to protect 2D frames and depth maps, either independently or simultaneously. The experimental results demonstrate that RZW-SR has superior DRM performance compared to existing schemes. First, RZW-SR can obtain the ownership shares relevant to a particular 3D video precisely and reliably for effective copyright identification of large-scale videos. Second, RZW-SR ensures lossless, precise, reliable and flexible copyright identification for 2D frames and depth maps of 3D videos.