Inverse Display Management Coding for SDR to EDR Conversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies face challenges in efficiently encoding and decoding video data to support a wide variety of standard dynamic range (SDR) and high dynamic range (HDR) display devices, particularly in converting SDR images to enhanced dynamic range (EDR) images without significant computational costs or compression.
Innovation Solution
A method involving the generation of composer metadata from SDR-EDR image pairs in a training database, which is used to encode SDR images into a video signal that can be decoded by downstream devices to produce EDR images optimized for EDR displays, leveraging machine learning-based prediction coefficients and backward reshaping mappings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If SDR images are converted to EDR images using traditional methods, then image quality is improved, but computational cost increases significantly
Solution Approach 1:
The patent applies preliminary action by pre-training machine learning models offline to learn the mapping between SDR and EDR images. The trained model parameters are then embedded in the codec, allowing the actual SDR-to-EDR conversion during video coding to use these pre-computed parameters rather than performing heavy computational tasks in real-time, thus improving image quality while reducing computational cost during encoding/decoding
Solution Approach 2:
The patent substitutes traditional mechanical/image-processing methods with machine learning-based approaches. Instead of using conventional image processing algorithms for SDR-to-EDR conversion, the system uses trained neural network models that can perform the conversion with lower computational overhead during actual video coding operations
2Manufacturing precision
If EDR images are compressed explicitly, then image fidelity is maintained, but compression complexity and computational requirements increase
Solution Approach 1:
The patent merges the SDR-to-EDR conversion process with the video compression process into a single unified operation. Instead of first converting SDR to EDR and then compressing the EDR images separately, the system performs both operations simultaneously through the inverse display management coding process, reducing overall complexity while maintaining image fidelity
Solution Approach 2:
The patent creates a universal coding framework that handles both SDR and EDR content through a single inverse display management process. The same codec infrastructure can process both standard dynamic range and enhanced dynamic range images without requiring separate dedicated compression systems, thereby reducing device complexity
3Reliability
If separate coding systems are used for SDR and HDR displays, then display-specific optimization is achieved, but system complexity increases
Solution Approach 1:
The patent implements a universal inverse display management coding system that can adapt to both SDR and EDR displays through a single framework. The system uses a unified SDR base layer that can be processed differently based on the target display type, eliminating the need for completely separate coding systems while still providing display-specific optimization
Solution Approach 2:
The patent segments the video content into an SDR base layer and an EDR enhancement layer. This segmentation allows the system to maintain a common SDR foundation that works on all displays while adding EDR-specific information only when needed for EDR displays, reducing overall system complexity through modular design
Data Source
Figure 1A
Figure 1B
Figure 1C
AI summary
A standard dynamic range (SDR) image is received. Composer metadata is generated for mapping the SDR image to an enhanced dynamic range (EDR) image. The composer metadata specifies a backward reshaping mapping that is generated from SDR-EDR image pairs in a training database. The SDR-EDR image pairs comprise SDR images that do not include the SDR image and EDR images that corresponds to the SDR images. The SDR image and the composer metadata are encoded in an output SDR video signal. An EDR display operating with a receiver of the output SDR video signal is caused to render an EDR display image. The EDR display image is derived from a composed EDR image composed from the SDR image based on the composer metadata.