HDR to SDR Color Space Transformation via ICtCp

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current color space transformation methods fail to accurately convert HDR-WCG images to SDR-SCG images while maintaining perceptual accuracy, as they do not simultaneously consider brightness and chromaticity values, leading to quality losses and incompatibility with modern color representations.

Innovation Solution

The method employs a constant luminance color space transformation using IC T C P and its polar coordinate representation, with steps involving conversion to a transformation color space, calculation of color space boundaries, hue shift vectors, and geometric intersection calculations to ensure accurate representation of HDR-WCG images within the SDR-SCG color gamut.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If older color space transformation methods (e.g., L*a*b*) are used, then the transformation process is simple and well-established, but the method is unsuitable for HDR color space transformations where large differences in brightness and chromaticity must be adjusted

Engineering Contradiction:
Improveease of transformation processVSAvoidadaptability to HDR color space
Core Design Contradiction:
Ease of manufactureVSAdaptability or versatility

Solution Approach 1:

The patent introduces a new transformation color space (ICtCopc) with modified parameters specifically designed for HDR content. The color space uses perceptually uniform chromaticity coordinates (ct, cp) and luminance (Y), replacing the traditional L*a*b* parameters. This parameter change enables accurate representation of large brightness and chromaticity differences in HDR while maintaining computational feasibility.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The transformation process is segmented into distinct stages: conversion to the new ICtCopc color space, separate processing of luminance and chromaticity components, and final conversion to target color space. This segmentation allows independent optimization of brightness and color transformations, addressing the limitations of unified traditional methods.

Inventive Principle:
Principle #1Segmentation

2Adaptability or versatility

If HDR-WCG image content is converted to SDR-SCG image content, then compatibility with older playback devices is improved, but the perceptual accuracy of brightness and color impression is lost

Engineering Contradiction:
Improvecompatibility with playback devicesVSAvoidperceptual accuracy
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent applies perceptual mapping functions that transform HDR-WCG parameters to SDR-SCG parameters while preserving perceptual relationships. The luminance mapping uses a piecewise function that accounts for human brightness perception characteristics, and the chromaticity mapping preserves hue relationships. This ensures that the converted SDR image maintains the original perceptual impression despite the reduced color and brightness gamut.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The new ICtCopc color space serves as an intermediary representation that decouples luminance and chromaticity information. This intermediary format allows independent optimization of the transformation to SDR, enabling accurate perceptual mapping while ensuring compatibility with SDR-SCG playback devices that cannot display the full HDR-WCG gamut.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Productivity

If the color space transformation uses traditional methods, then the processing is computationally efficient, but the method does not simultaneously consider brightness and chromaticity values leading to quality losses

Engineering Contradiction:
Improvecomputational efficiencyVSAvoidtransformation accuracy
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The patent reformulates the transformation using a parameter set (Y, ct, cp) that separates luminance from chromaticity in a perceptually uniform way. This parameter change enables efficient computation through independent processing of luminance and chromaticity components, while simultaneously achieving high transformation accuracy by accounting for their interrelationships in HDR content.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The transformation algorithm is segmented into independent luminance mapping and chromaticity mapping operations. The luminance component is transformed using a perceptual mapping function, while the chromaticity components (ct, cp) are transformed separately to preserve color relationships. This segmentation improves computational efficiency while maintaining accuracy by avoiding the need for complex joint optimization.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP3945719B1Color space conversion method
Publication Date: 2023.12.27 HOCHSCHULE RHEINMAIN UNIV OF APPLIED SCI WIESBADEN RUSSELSHEIM
  • EP3945719B1 patent drawingFigure 1
  • EP3945719B1 patent drawingFigure 2~3
  • EP3945719B1 patent drawingFigure 4~5

AI summary

The invention relates to a method for color space transformation of at least one image or image sequence from a source color space to a target color space, comprising the following steps: First, the transformation color space is defined and all input color points are converted into the transformation color space. Then, the color space boundaries are calculated by determining the relative distance between the boundary of the source color space, the boundary of the target color space, and the input color point, in order to calculate the shift vectors. Subsequently, the shift vectors for the combination of source and target color spaces are calculated using the respective primary and secondary valences. Finally, the appropriate shift vector is applied to the hue angle of the input color point using the calculated relative distance.Then, a color space transformation is performed based on geometric intersection calculations. This involves determining the regression curve by separately calculating the slope and scaling factor, determining the soft-clipping mapping color space boundary, and calculating the intersection point of the regression curve with the soft-clipping mapping color space boundary. Finally, the transformed color coordinates are converted into the target color representation.