What is frame-based forensic watermarking?

Digital watermarking consists of embedding a key dependent signal into digital data in a robust and invisible way. While some watermarking techniques exploit the specificities of a compression standard, others embed watermarks in a 3D transform.

Digital video watermarking techniques in the present day mainly consist of watermarking a sequence of still images. In frame-based forensic watermarking techniques, imperceptible watermark information is embedded in video frames throughout the media content. In the case of content leakage or attacks, such as re-compression, geometrical deformations adding noise, etc., the watermarks can be extracted from a video segment. Two major embedding strategies are used in frame based video watermarking: Either a different watermark is inserted in each video frame, or the same watermark is embedded in all the video frames. From a security standpoint, repeating the same sequence of watermarks lets attackers group frames carrying the same watermark and perform a watermark estimation remodulation (WER) attack.

The process of watermarking is usually embedded into a soft or hard transcoder, such as SDK or a standalone command line program. The watermarking SDK provides a video frame-based data interactive interface which makes asset management easier while reducing the performance required for the watermarking process. Transcoding is then carried out to decode each video frame from media content which can be an offline video file, DRM protected content, or even live TV. Each frame is converted to an RGB format and fed frame by frame into the SDK, which then gives back the watermarked frames with the same picture parameters, such as width, height, pixel format, etc. Finally, the watermarked video frame is encoded by the transcoder into the output media file. The SDK then returns the transcoder the same pixel format as the output in order to convert it to its final form.

Frame-based forensic video watermarking techniques make the process of inserting and extracting the watermark easier. They also have the advantage of being able to use short samples of videos within a few frames for the extraction process. Therefore, frame based watermarking techniques can help add an additional layer of security to DRM-protected content by securely embedding identifiable information that can be extracted in the event of a breach or leakage.