Color Gamut Conversion in a Linear Domain to Limit Quantization Errors

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing video conversion methods amplify quantization errors during color gamut conversions, particularly when converting between HDR and SDR formats, leading to significant inaccuracies in video content.

Innovation Solution

A method involving conversion of input YUV data to RGB data in a linear domain, followed by conversions within this domain to achieve the desired color gamut, using a combination of linear and non-linear functions to minimize quantization errors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If color gamut conversion is performed using conventional quantized data processing, then device complexity is reduced, but manufacturing precision deteriorates due to amplified quantization errors

Engineering Contradiction:

Improveconversion accuracyVSAvoidprocessing complexity

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent applies preliminary dithering noise before quantization to reduce quantization errors. By adding controlled random noise to the signal before the quantization step, the method prepares the data in advance to minimize error amplification during subsequent color gamut conversions, thereby improving conversion accuracy without adding complex processing steps during the actual conversion operation.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent changes the precision parameter of intermediate calculations by performing conversions in higher precision (e.g., 16-bit or 32-bit floating point) rather than maintaining the original quantized precision throughout. This parameter change allows accurate color gamut conversion while still using standard quantized input and output formats, resolving the contradiction between accuracy and complexity.

Inventive Principle:

Principle #35Parameter changes

2Adaptability or versatility

If multiple conversion operations are performed during color gamut conversion, then adaptability is improved, but quantization errors are amplified

Engineering Contradiction:

Improvecolor gamut compatibilityVSAvoidvideo content accuracy

Core Design Contradiction:

Adaptability or versatilityVSManufacturing precision

Solution Approach 1:

The patent introduces an intermediary high-precision linear domain representation between the quantized non-linear input and the quantized non-linear output. By converting quantized non-linear data to a high-precision linear domain (serving as an intermediary), performing color gamut transformations, and then converting back to quantized non-linear format, the method enables multiple conversion operations while minimizing quantization error amplification through the use of the high-precision intermediary representation.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS12477090B2Method for limiting effects of quantization in a color gamut modification process applied to a video content

Publication Date: 2025.11.18 INTERDIGITAL CE PATENT HOLDINGS SAS

AI summary

A method comprising applying a conversion function to first RGB picture data in a color gamut and corresponding to a first non-linear domain to obtain second RGB picture data in the same color gamut in a linear domain, said conversion function being a combination (1603) of a first and a second function, the first function being a linear function defined between zero and a second inflexion point the slope of which depending on a steepness value and the second function being a part of a linear to first non-linear domain transfer function defined from the second inflexion point, the second inflexion point being computed (1601) in the first non-linear domain from a first inflexion point obtained (1600) in the linear domain, the steepness value being computed (1602) from the first and second inflexion points.