Digital Watermarking Compressed Video Using Macroblock Types

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for digitally watermarking media streams, such as MPEG streams, often require wholesale re-encoding, affect the stream size, are detectable by consumers, or impact content presentation, failing to meet the need for a method that is imperceptible and efficient for content providers.

Innovation Solution

The method identifies sequences of visually indistinguishable frames in a compressed video stream, specifically using macroblock types to encode digital watermark information within B-frames, ensuring the watermarked stream remains valid and identical in size and quality to the original, without affecting content presentation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If existing methods for digitally watermarking media streams are used, then watermarking capability is achieved, but the stream size is affected and processing complexity increases

Engineering Contradiction:

Improvewatermarking capabilityVSAvoidstream size

Core Design Contradiction:

ReliabilityVSQuantity of substance

Solution Approach 1:

The video stream is segmented into individual frames, and the watermarking process is applied selectively to specific frames rather than the entire stream. This allows the watermark information to be embedded in discrete segments without affecting the overall stream size significantly.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies watermarking only to specific frames that meet certain criteria (e.g., frames with sufficient temporal redundancy), rather than uniformly watermarking every frame. This local application maintains watermarking reliability while minimizing the total amount of data modified.

Inventive Principle:

Principle #3Local quality

2Reliability

If existing watermarking methods are applied, then digital marking is achieved, but content presentation is affected and detectability increases

Engineering Contradiction:

Improvewatermarking capabilityVSAvoidcontent presentation degradation

Core Design Contradiction:

ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent embeds watermark information subtly within the natural variations of video frames, using only the minimal necessary modifications to encode the watermark. By applying partial action (watermarking only where needed and with minimal distortion), the method achieves reliable watermarking while keeping degradation imperceptible to consumers.

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The patent exploits the natural temporal redundancy and visual imperceptibility in video sequences as beneficial features. The slight modifications introduced for watermarking are hidden within the natural frame-to-frame variations, converting what would normally be perceived as degradation into an imperceptible part of the video's natural variation.

Inventive Principle:

Principle #22Blessing in disguise (Convert harm into benefit)

3Reliability

If wholesale re-encoding is performed for watermarking, then watermark information is embedded, but processing time and complexity increase significantly

Engineering Contradiction:

Improvewatermark embeddingVSAvoidprocessing efficiency

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent identifies and selects suitable frames for watermarking in advance, based on pre-analyzed properties such as temporal redundancy and visual importance. This preliminary selection avoids the need for complex re-encoding operations during the actual watermarking process, significantly improving processing efficiency.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent extracts only the essential watermark information and embeds it selectively in pre-identified suitable frames, rather than processing the entire video stream through complex re-encoding. This extraction approach minimizes processing overhead while maintaining watermark reliability.

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS7672373B2Digital watermarking of a media stream using coded macroblock types

Publication Date: 2010.03.02 SYNAMEDIA LTD

AI summary

A method and a carrier medium instructing a processor to execute a method. The method includes accepting a compressed video stream compressed by a compression method that uses macroblocks of possibly different macroblock types for each frame predictive encoding frames. The method further includes identifying a sequence of frames that are substantially visually indistinguishable and that include a predictively encoded frame and the at least one reference frame of the predictively encoded frame. The method further includes creating an ordered set of macroblocks, with the macroblock types of different macroblocks in the ordered set used to encode digital watermark information. The method further includes replacing the predictively encoded frame of the identified sequence with a newly-encoded predictively encoded replacement frame that contains the created ordered set of macroblocks, including the encoded watermarking information.