HDR Layered Encoding with Metadata-Guided Format Conversion
Find Innovative SolutionsGenerate 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
Engineering 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
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.
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.
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
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.
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.
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
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.
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.
Data Source
Figure 1
Figure 2
Figure 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.