Watermark Detection Using Self-Referencing Video Blocks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing watermark detection methods are inefficient when dealing with luminance variations and video transforms, particularly in camcording scenarios, as they require temporal and spatial registration between the original and pirated videos, making them sensitive to luminance modifications.
Innovation Solution
The method compares watermarked blocks within the same pirated video to potential reference blocks that undergo the same luminance transform, using Euclidean distance or correlation to identify the watermark value, allowing for robust detection without the need for preliminary registration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If watermark detection is performed by comparing watermarked blocks to original master video blocks, then watermark detection accuracy is maintained under ideal conditions, but detection robustness deteriorates when luminance variations or video transforms occur
Solution Approach 1:
The invention creates synthetic reference blocks by copying and transforming existing video blocks within the pirated video itself. Instead of relying on the original master video, the method generates reference blocks that undergo the same luminance transforms and video processing as the watermarked blocks, ensuring both blocks are affected identically by any transformations. This copying approach within the same video copy eliminates the mismatch problem between master and pirated videos.
Solution Approach 2:
The invention introduces an intermediary step of generating synthetic reference blocks that serve as a mediator between the watermarked blocks and the watermark detection process. These reference blocks act as a bridge that accounts for luminance variations and transforms, allowing comparison without directly referencing the original master video. The intermediary reference blocks absorb the effects of transforms, enabling robust watermark detection.
2Measurement precision
If temporal and spatial registration is performed between pirated video and master video, then watermark detection can be performed using original block values, but the process complexity and time consumption increase significantly
Solution Approach 1:
The invention extracts and removes the registration requirement entirely from the watermark detection process. Instead of performing complex temporal and spatial registration between pirated and master videos, the method extracts only the necessary reference blocks from within the pirated video itself and synthesizes them to match the watermarked blocks. This extraction approach eliminates the need for alignment procedures.
Solution Approach 2:
The pirated video serves itself by generating its own reference blocks from within its own content. The method selects blocks from the pirated video, applies the same transforms, and uses them as references for watermark detection. This self-service approach eliminates the need for external master video reference and complex registration processes, making the system autonomous and simpler.
3Ease of manufacture
If average luminance comparison is used for watermark detection, then the method is simple to implement, but it becomes sensitive to luminance transforms and automatic gain control variations
Solution Approach 1:
The invention changes the fundamental parameter being compared from absolute luminance values to relative differences between watermarked and reference blocks. By computing the difference between corresponding blocks (both subjected to the same transforms), the method eliminates sensitivity to automatic gain control and luminance variations. The parameter comparison shifts from absolute to differential, making it invariant to luminance scaling.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a method for decoding a watermark embedded in a target video comprising blocks of samples representative of the target video wherein watermark is embedded in an original video by modifying motion vectors of a compressed representation of the original video. The method comprises the steps of determining in the video a location of a watermarked block; determining a location for each reference block among a plurality of reference blocks wherein a reference block is associated to a watermark value among a plurality of watermark values; comparing the watermarked block with each reference block; and identifying watermark value for the watermarked block from the results of the plurality of comparison. The method is remarkable in that each reference block is a potential prediction block also located in the target video, for instance the pirated video, and in that the reference block that corresponds to the best prediction for the watermark block is selected. Thus the plurality of comparisons between the watermarked block and each reference block is performed among the same target video.