HDR Image Encoding With Luminance-Dependent Chroma Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image coding technologies struggle to efficiently transition between images with different luminance dynamic ranges while maintaining chromatic accuracy, particularly in the transition from High Dynamic Range (HDR) to Standard Dynamic Range (SDR) for video communication systems.
Innovation Solution
A system that encodes HDR images using a luminance mapping function to convert HDR colors to SDR colors, incorporating metadata for decoding and display tuning to adapt to various display capabilities, ensuring accurate chromatic representation across different dynamic ranges.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If HDR images are converted to SDR using traditional luminance mapping, then the luminance dynamic range is reduced for compatibility, but chromatic accuracy deteriorates due to color distortion
Solution Approach 1:
The patent applies different mapping strategies to different color components based on their local characteristics. The luminance component Y is mapped using a piecewise function that preserves dark region details, while chroma components Cb and Cr are adjusted using luminance-dependent scaling factors. This local differentiation allows the system to maintain chromatic accuracy in different regions of the image simultaneously while adapting to SDR display constraints.
Solution Approach 2:
The patent changes the mapping parameters dynamically based on luminance levels. The chroma scaling factors are not fixed but vary according to the local luminance value, allowing the system to preserve color accuracy in different brightness regions. This parameter adaptation enables the conversion to maintain both display compatibility and chromatic fidelity across the full dynamic range.
2Productivity
If a simple luminance mapping function is used for HDR to SDR conversion, then the conversion process is computationally efficient, but color accuracy is compromised
Solution Approach 1:
The patent segments the luminance mapping function into multiple pieces with different characteristics. The piecewise function divides the luminance range into segments, each handled by a specific mapping formula. This segmentation allows the system to maintain computational efficiency through simple piecewise calculations while achieving high color accuracy by applying different optimization strategies to different luminance regions.
Solution Approach 2:
The patent introduces dynamic chroma scaling factors that adapt to local luminance conditions. Rather than using fixed mapping parameters, the system dynamically adjusts the chroma components based on the luminance value at each pixel location. This dynamic adaptation maintains color accuracy across varying brightness levels while keeping the computational process efficient through straightforward conditional calculations.
3Device complexity
If chroma components are scaled uniformly during luminance mapping, then the implementation is simple, but chromaticity errors increase particularly in dark regions
Solution Approach 1:
The patent applies different chroma scaling strategies to different luminance regions. Dark regions use one scaling approach to preserve color accuracy, while brighter regions use another. This local quality differentiation ensures that chromaticity accuracy is maintained where it matters most (in dark regions) without overly complicating the overall implementation.
Solution Approach 2:
The patent changes the chroma scaling parameters based on luminance levels. The scaling factors are luminance-dependent, allowing the system to compensate for the reduced sensitivity of human vision to color in dark regions. This parameter adaptation maintains chromaticity accuracy across the full luminance range while keeping the implementation relatively simple through straightforward conditional logic.
Data Source
Figure 1A~1D
Figure 2
Figure 3A~3C
AI summary
To obtain better color control in dynamic range mappings, which are e.g. useful in particular for HDR image coding and communication, the inventor proposes a luminance down-mapper (500) and corresponding up-mapper comprising: a luma input (501) for a normalized input luma (Y'_i) which codes an input luminance lying within a first luminance dynamic range (DR_H); a chroma input (502) for two normalized input chroma components (Cb, Cr_i), which together with the input luma define an input color (C _H); a supply circuit (503) arranged to supply a set of multipliers (g1(X'1); g2(X'2)) of respective normalized input values (X' 1; X'2) spanning a range of normalized input values, which is arranged to determine the set of multipliers based on a luma mapping function (F _LBri), by calculating for any one of the respective normalized input values a respective output value resulting from applying the luma mapping function to the respective normalized input value, and calculating the respective multiplier as the result of dividing the respective output value by the respective input value; a first multiplier determination circuit (504) arranged to output a first multiplier (gd(Y')) from the set of multipliers which corresponds to an input value equal to the normalized input luma; a largest primary color component determining circuit (506) arranged to determine the largest component (LC_i) of a red, green and blue color component representation of the input color; a second multiplier determination circuit (507) arranged to output a second multiplier (gd(LC)) from the set of multipliers which corresponds to an input value equal to the largest component (LC_i); a weight determination circuit (508) arranged to receive a control parameter (Ksi) from a source (509), and to output a first weight (A) and a second weight equal to one minus the first weight (1-A), wherein the first weight is calculated, for non-zero values of the input luma, as the minimum of the value one and a first result, wherein the first result equals a multiplication of the control parameter by a second result, wherein the second result is obtained by subtracting the value one from a third result; wherein the third result is obtained by dividing the largest component (LC_i) by the input luma (Y'_i); a first multiplication circuit (510) arranged to obtain a first weighted multiplier (gd_W) as the result of multiplying the first multiplier (gd(Y')) by the second weight (1-A); a second multiplication circuit (511) arranged to obtain a second weighted multiplier (gd_LCW) as the result of multiplying the second multiplier (gd(LC)) by the first weight (A); an adder (512) to yield a final multiplier (gF) by adding the first weighted multiplier (gd_W) to the second weighted multiplier (gd_LCW); a third multiplication circuit (513) to yield a normalized output luma by multiplying the input luma by the final multiplier (gF), and a fourth multiplication circuit (514) to yield two normalized output chroma components by multiplying the two normalized input chroma components (Cb, Cr_i) by the final multiplier (gF).