Hierarchical Gamut Mapping for Precision Color Volume
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing gamut mapping algorithms face challenges in accurately representing the target device color volume due to memory limitations, leading to loss of tonal resolution and banding artifacts, especially when dealing with non-linear gamut volumes and high bit depths, where the number of points required to represent the gamut surface exceeds conventional memory capabilities.
Innovation Solution
A hierarchical gamut mapping method that uses coarse sampling to identify the region containing the optimal point, followed by fine sampling to refine the estimate, allowing for the calculation of the optimal solution without interpolation and reducing memory footprint, using a multi-step procedure with progressively higher bit resolutions to determine the closest gamut-mapped value.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional gamut mapping algorithms use high bit depth representation to accurately represent the target device color volume, then mapping precision is improved, but memory requirements exceed conventional capabilities
Solution Approach 1:
The patent segments the color space representation into multiple layers with different bit depths. A first layer uses lower bit depth (e.g., 8-bit) for general color volume representation, while a second layer uses higher bit depth (e.g., 16-bit) only for specific regions requiring precision. This segmentation allows accurate gamut mapping in critical areas while keeping overall memory footprint manageable.
Solution Approach 2:
The patent applies different quality levels (bit depths) to different regions of the color space based on local requirements. Regions with complex gamut boundaries or high importance receive higher bit depth representation, while other regions use lower bit depth. This local quality approach ensures mapping precision where needed without uniformly increasing memory requirements across the entire color space.
2Quantity of substance
If conventional gamut mapping algorithms reduce memory usage by lowering bit depth, then memory footprint is reduced, but tonal resolution is lost and banding artifacts appear
Solution Approach 1:
The patent divides the color space into multiple layers where the first layer provides broad coverage at lower bit depth, and the second layer provides enhanced precision at higher bit depth for specific regions. This segmentation allows the system to maintain acceptable tonal resolution overall while using lower average bit depth, thus reducing memory footprint without uniformly sacrificing quality.
Solution Approach 2:
The patent applies higher bit depth locally to regions where tonal resolution is critical (such as near gamut boundaries or in important color regions), while using lower bit depth in other regions. This ensures that banding artifacts are minimized where they would be most noticeable, while keeping memory usage low in less critical areas.
3Measurement precision
If conventional gamut mapping algorithms represent non-linear gamut volumes with sufficient points, then mapping accuracy is improved, but computational resources and processing time increase
Solution Approach 1:
The patent segments the gamut representation into multiple layers with different resolutions. The first layer provides a coarse approximation of the gamut volume, while the second layer adds detailed refinement only where needed. This segmentation allows the algorithm to achieve high mapping accuracy by focusing computational effort on critical regions rather than uniformly processing the entire color space, thus improving processing speed.
Solution Approach 2:
The patent performs preliminary processing by first establishing the lower bit depth layer that captures the general gamut structure. This preliminary representation allows for quick initial mappings, and only requires additional computational resources for the higher bit depth layer when and where precision is needed, rather than always performing full high-precision calculations.
Data Source
AI summary
Methods and systems for gamut mapping are disclosed. Pixels of an image or points of a look-up-table can be gamut mapped in a multi-step iterative process, by generating a coarse gamut hull and calculating a value of a distance metric for out-of-gamut pixels or LUT points, and subsequently generating a fine gamut hull in the neighborhood of the coarse gamut hull points closest to the out-of-gamut pixel or LUT points under consideration. The out-of-gamut pixel or LUT point is gamut mapped based on the smallest distance metric value calculated.