HDR Image Layer Conversion for Display-Adaptive Restoration

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current HDR transcoding schemes result in exceptions and affect user experience due to direct transcoding of HDR images or videos, leading to incorrect bitstream generation and abnormal visual effects.

Innovation Solution

A dual-layer image processing method involving base-layer and enhancement-layer data conversion based on metadata, including color space and gamut conversion, to enhance restoration and adaptability to display devices.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If HDR image or video is directly transcoded using single-layer scheme, then encoding process is simple, but transcoded HDR image or video contains exceptions and affects user experience

Engineering Contradiction:
Improveencoding process complexityVSAvoidtranscoded HDR image quality
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent divides the HDR image into base-layer data and enhancement-layer data, encoding them separately. The base-layer data contains essential image information while the enhancement-layer data contains additional detail information. This segmentation allows the decoder to reconstruct HDR images with higher quality by combining both layers, resolving the contradiction between encoding simplicity and output quality.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary conversion process where base-layer data and enhancement-layer data are converted to different color spaces or formats before being combined. This intermediary step acts as a mediator that enables proper integration of the two layers, ensuring exception-free transcoding while maintaining encoding efficiency through the structured dual-layer approach.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Manufacturing precision

If base-layer data and enhancement-layer data are converted based on metadata, then restoration degree and adaptability improve, but processing complexity increases

Engineering Contradiction:
Improverestoration degreeVSAvoidprocessing complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent changes key parameters of the base-layer and enhancement-layer data, including color space conversion (e.g., from RGB to YCbCr), bit depth adjustment, and resolution scaling. These parameter changes are guided by metadata that describes the original HDR image properties. By systematically transforming parameters rather than performing complex pixel-level operations, the patent achieves high restoration degree while controlling processing complexity.

Inventive Principle:
Principle #35Parameter changes

3Adaptability or versatility

If dual-layer conversion is performed on base-layer and enhancement-layer data, then adaptability to display devices improves, but encoding time increases

Engineering Contradiction:
Improveadaptability to display devicesVSAvoidencoding time
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent performs preliminary conversion of base-layer and enhancement-layer data to standard color spaces and formats during the encoding phase. Metadata is generated in advance that describes the conversion parameters and display device characteristics. This preliminary action ensures that the data is pre-adapted for various display devices, reducing the need for real-time conversion during playback and thereby reducing the actual encoding time experienced by users.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4697716A1Image processing method and related device
Publication Date: 2026.02.18 HUAWEI TECH CO LTD
  • EP4697716A1 patent drawingFigure 1
  • EP4697716A1 patent drawingFigure 2
  • EP4697716A1 patent drawingFigure 3~4

AI summary

This application provides an image processing method and a related device. The method includes: obtaining first base-layer data, first enhancement-layer data, and first metadata, where the first base-layer data is base-layer data corresponding to a first HDR image, and the first enhancement-layer data is enhancement-layer data corresponding to the first HDR image; performing conversion on target data according to the first metadata, to obtain target conversion data, where the target data is the first base-layer data and/or the first enhancement-layer data; and determining a second HDR image based on the target conversion data. Based on the foregoing technical solution, a more efficient and proper encoding scheme can be selected for the base-layer data and/or the enhancement-layer data, and then the base-layer data and/or the enhancement-layer data may be converted during decoding, so that the finally synthesized HDR image has a higher restoration degree and is more adaptive to a display device.