Scale-Independent Disparity Maps for 3D Video
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in efficiently processing and representing disparity maps for stereoscopic video images, particularly in maintaining sub-pixel accuracy and compatibility across different resolutions, which affects the quality of 3D effects and viewer comfort.
Innovation Solution
The use of a non-standard resolution for disparity values, represented as integers, allows for sub-pixel accuracy and compatibility across various display resolutions by converting these values to a common multiple format, enabling seamless transmission and display without transcoding, thus maintaining the integrity of the 3D effect.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If disparity maps are processed using standard display resolutions, then compatibility with existing display devices is improved, but sub-pixel accuracy is lost and information is lost during transcoding
Solution Approach 1:
The patent introduces a new dimension by using a non-standard resolution (e.g., 11520 pixels wide) that is a common multiple of various standard resolutions. This allows disparity values to be represented with higher precision in a unified coordinate system, from which they can be accurately converted to any standard resolution without loss of sub-pixel information.
Solution Approach 2:
The patent changes the resolution parameter from standard display resolutions to a non-standard resolution that is a common multiple of various standards. This parameter change enables the disparity map to maintain sub-pixel accuracy while being adaptable to multiple standard resolutions through mathematical conversion.
2Adaptability or versatility
If disparity values are converted to common multiple format, then compatibility across different resolutions is improved, but data processing complexity increases
Solution Approach 1:
The patent performs preliminary action by pre-calculating disparity values at a non-standard resolution that is a common multiple of various standard resolutions. This preliminary processing at a higher resolution enables simpler downstream conversions to any standard resolution, reducing overall processing complexity despite the initial computational cost.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Various implementations provide a resolution-specific attribute for a resolution-invariant region. Several implementations provide an SCM-based disparity value for scale-invariant cells. Particular implementations provide a signal or structure that includes such an attribute. Yet further implementations access such an attribute, and process the attribute. One particular implementation accesses an attribute for a region of a picture. The region is coextensive with a first integer number of pixels in a particular resolution, and with a second integer number of pixels in a second resolution. The attribute is specific to the region and to a combination resolution determined as a combination of the particular resolution and the second resolution. The attribute is converted to provide an attribute specific to the particular resolution. The converted attribute is associated with the first integer number of pixels. Other implementations use the converted attribute in post-processing operations, such as, for example, subtitle placement using disparity information.