HDR Layered Encoding with Metadata-Guided Format Conversion

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current HDR transcoding schemes cause exceptions in transcoded HDR images or videos, affecting user experience due to the direct re-encoding of HDR content without proper handling of HDR pixel values and format identifiers.

Innovation Solution

An encoding method that divides HDR image data into base-layer and enhancement-layer data, with metadata containing indication information for decoding schemes, allowing flexible selection of encoding schemes based on region characteristics, improving encoding efficiency and synthesis effects.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If HDR image or video is directly transcoded using a single-layer scheme, then the transcoding process is simple and fast, but exceptions occur in the transcoded content affecting user experience

Engineering Contradiction:
Improvetranscoding speedVSAvoidtranscoded content quality
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent segments HDR content into multiple layers: base-layer data (containing essential information for basic decoding) and enhancement-layer data (containing additional details for improved quality). This segmentation allows the decoder to reconstruct HDR content accurately by combining both layers, resolving the contradiction between simple transcoding and high quality output.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary metadata structure that carries encoding scheme information and layer data between the encoder and decoder. This metadata acts as a mediator that enables the decoder to properly interpret and reconstruct HDR content without requiring complex real-time analysis, thus maintaining both speed and quality.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Manufacturing precision

If multiple encoding schemes are used for different regions of HDR data, then encoding precision and synthesis effect are improved, but the complexity of the encoding process increases

Engineering Contradiction:
Improveencoding precisionVSAvoidencoding process complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent applies local quality by allowing different encoding schemes (e.g., PQ, HLG, gamma) to be used for different regions or layers of HDR data. The metadata specifies which encoding scheme applies to each layer, enabling precise encoding tailored to specific content requirements while keeping the overall process manageable through structured metadata.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent utilizes parameter changes by allowing the encoding scheme parameters to vary across different layers and regions. The metadata carries these parameter variations, enabling the encoder to switch between different encoding approaches (PQ, HLG, gamma) based on content characteristics without requiring a completely different encoding architecture.

Inventive Principle:
Principle #35Parameter changes

3Adaptability or versatility

If HDR pixel values and format information are processed separately before re-encoding, then encoding flexibility is improved, but the processing time and complexity increase

Engineering Contradiction:
Improveencoding flexibilityVSAvoidprocessing time
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent merges HDR pixel values, format information, and encoding scheme metadata into a unified multi-layer data structure. This combining allows all HDR-related information to be processed and transmitted together in a single encoding pass, eliminating the need for separate processing steps and reducing overall processing time while maintaining flexibility.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent creates a universal metadata structure that can carry multiple types of information (encoding scheme identifiers, layer data, region information) in a single format. This multi-functional metadata structure enables the encoder to handle various HDR scenarios (different encoding schemes, different regions, different layer configurations) without requiring separate processing paths.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP4694120A1Encoding method, decoding method, and related device
Publication Date: 2026.02.11 HUAWEI TECH CO LTD
  • EP4694120A1 patent drawingFigure 1
  • EP4694120A1 patent drawingFigure 2
  • EP4694120A1 patent drawingFigure 3~4

AI summary

This application provides an encoding method, a decoding method, and a related device. The method includes: obtaining a first HDR image and N pieces of first base-layer data corresponding to the first HDR image; determining N pieces of first enhancement-layer data based on the first HDR image, the N pieces of first base-layer data, and at least one encoding scheme, where N is a positive integer greater than or equal to 1; and encoding the N pieces of first enhancement-layer data, the N pieces of first base-layer data, and first metadata to obtain a bitstream, where the first metadata includes first indication information, and the first indication information indicates the at least one encoding scheme or a decoding scheme corresponding to the at least one encoding scheme. The foregoing technical solution can implement flexible format conversion, improving encoding efficiency and final synthesis effect.