On a self-recovery digital watermarking scheme robust against spatial and temporal attacks on compressed video

Abstract

In recent decades, the use of digital media has increased. With it, digital media exposure to unwanted manipulations that, although unintentional, can affect third parties. These manipulations in digital watermarking are known as attacks, consisting of intentional or unintentional editing of multimedia content such as images, videos, or audio streams. For video sequences, attacks can be performed in spatial and temporal domains. The former includes changes to individual frames, while the latter involves altering sub-sequences, such as inserting or deleting frames.This work aims to achieve approximate recovery from these attacks using digital watermarking. Unlike most existing works, this scheme uses the compression domain since most video applications use some compression standard to transmit and store these sequences. By adapting popular image block-based popular recovery schemes for large video sequences and using a novel scrambled watermark mapping, satisfactory recovery and visual quality results are achieved.