SEI Messaging of Source Color Volume for HDR Display Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies for communicating and processing source color volume information in images, particularly for HDR and WCG content, face challenges such as providing insufficient metadata, adding significant overhead, and requiring excessive computational resources.
Innovation Solution
The proposed solution involves using SEI messaging to communicate source color volume information, including the largest 2D color gamut, maximum, minimum, and average luminance values, and optionally sliced color gamut for specific luminance values, to enable efficient display management and improve image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If detailed source color volume information is communicated using existing SEI or VUI messaging proposals, then display management accuracy for HDR content is improved, but metadata overhead and computational complexity increase significantly
Solution Approach 1:
The patent extracts only the essential color volume information (chromaticity coordinates of primaries and luminance metadata) needed for display management, separating it from the full content color volume description. This selective extraction reduces metadata overhead while maintaining sufficient accuracy for HDR display mapping.
Solution Approach 2:
Instead of describing the full content color volume and having displays adapt, the patent inverts the approach by having the encoder signal the mastering display color volume and content luminance metadata, allowing decoders to perform efficient tone mapping and color space transformation without complex computations.
2Manufacturing precision
If content color volume is described using multiple primary expressions and spatial regions, then color accuracy is improved, but the complexity of generating and processing the metadata increases
Solution Approach 1:
The patent applies local quality by signaling luminance metadata specifically for regions containing content exceeding the mastering display luminance range, rather than providing uniform detailed information across all content. This reduces overall metadata complexity while maintaining color accuracy where it matters most.
Solution Approach 2:
The patent changes the parameter representation by using luminance metadata (max_content_luminance, min_content_luminance, average_content_luminance) alongside chromaticity coordinates, enabling efficient tone mapping operations that reduce processing complexity compared to full color volume descriptions.
3Reliability
If full content color volume information is transmitted, then rendering accuracy on diverse displays is improved, but bandwidth consumption and processing overhead increase
Solution Approach 1:
The patent performs preliminary action by having the encoder pre-calculate and signal the mastering display color volume and content luminance metadata during encoding. This allows the decoder to perform simple look-up table-based tone mapping and color space transformation, reducing both bandwidth requirements and decoding complexity while maintaining rendering accuracy.
Data Source
AI summary
Methods are described to communicate source color volume information in a coded bitstream using SEI messaging. Such data include at least the minimum, maximum, and average luminance values in the source data plus optional data that may include the color volume x and y chromaticity coordinates for the input color primaries (e.g., red, green, and blue) of the source data, and the color x and y chromaticity coordinates for the color primaries corresponding to the minimum, average, and maximum luminance values in the source data. Messaging data signaling an active region in each picture may also be included.