Backward-Compatible HDR Image Encoding via LDR Base Layer

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for encoding and decoding high dynamic range (HDR) images lack backward compatibility with existing lower-dynamic-range (LDR) systems, leading to loss of information and suboptimal performance in reproducing both HDR and LDR images.

Innovation Solution

The method involves obtaining a lower dynamic range image corresponding to the HDR image, identifying groups of pixels with the same values, generating a prediction function based on HDR pixel values, computing a residual image, and encoding both the LDR and prediction functions to enable backward-compatible decoding of HDR images.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If high dynamic range data formats with larger numbers of bits per pixel are used, then the dynamic range and color fidelity are improved, but backward compatibility with prior lower dynamic range data formats is lost

Engineering Contradiction:
Improvedynamic rangeVSAvoidbackward compatibility
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent segments the HDR image data into multiple layers: an LDR base layer that maintains compatibility with existing systems, and an HDR enhancement layer that provides extended dynamic range. This segmentation allows different parts of the data structure to serve different purposes - the base layer ensures backward compatibility while the enhancement layer provides improved dynamic range for capable devices

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements a nested structure where the LDR representation is embedded within the HDR data structure as a base layer, and the HDR-specific enhancement data is nested within or alongside it. This nested organization allows LDR systems to use only the base layer while HDR systems can access both layers for full functionality

Inventive Principle:
Principle #7Nested doll (Nesting)

2Adaptability or versatility

If 24 bits per pixel are used for image data, then the data format is compatible with existing systems, but the brightness range is insufficient for high dynamic range display

Engineering Contradiction:
Improvesystem compatibilityVSAvoidbrightness range
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent applies partial action by providing more than 24 bits per pixel for HDR capability while maintaining 24-bit compatibility for LDR systems. The enhancement layer contains the additional bits needed for extended brightness range, but this extra data is only utilized by HDR-capable displays, allowing existing systems to function normally with the base layer

Inventive Principle:
Principle #16Partial or excessive action

3Productivity

If lossy HDR video compression methods are used, then the compression ratio is improved, but information loss occurs and backward compatibility is not provided

Engineering Contradiction:
Improvecompression ratioVSAvoidimage information
Core Design Contradiction:
ProductivityVSLoss of information

Solution Approach 1:

The patent employs feedback mechanisms in the compression process where the encoder analyzes the relationship between LDR and HDR image data to optimize the enhancement layer. The compression algorithm uses feedback from the base layer reconstruction to minimize information loss while achieving efficient compression, ensuring that the residual HDR information is preserved with minimal degradation

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3197157B1High dynamic range codecs
Publication Date: 2021.12.15 MAX PLANCK GESELLSCHAFT ZUR FOERDERUNG DER WISSENSCHAFTEN EV
  • EP3197157B1 patent drawingFigure 1
  • EP3197157B1 patent drawingFigure 1A
  • EP3197157B1 patent drawingFigure 2

AI summary

A method for encoding high dynamic range (HDR) images involves providing a lower dynamic range (LDR) image, generating a prediction function for estimating the values for pixels in the HDR image based on the values of corresponding pixels in the LDR image, and obtaining a residual frame based on differences between the pixel values of the HDR image and estimated pixel values. The LDR image, prediction function and residual frame can all be encoded in data from which either the LDR image of HDR image can be recreated.