Adaptive Reshaping for Video Signal Detail Preservation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video codecs fail to preserve visual details when encoding video data at low bit depths, leading to loss of details in high and low luminance ranges, and the metadata required for high-quality decoding is often too large for downstream devices to process effectively.

Innovation Solution

Implementing adaptive reshaping techniques in video codecs that dynamically allocate codewords based on image content, using power functions and lookup tables to approximate reshaping functions, allowing for efficient encoding and decoding of video signals across varying dynamic ranges while minimizing metadata transmission.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If video data is encoded at low bit depth, then device complexity and metadata size are reduced, but visual details in high and low luminance ranges are lost

Engineering Contradiction:
Improvemetadata processing complexityVSAvoidvisual details in luminance ranges
Core Design Contradiction:
Device complexityVSLoss of information

Solution Approach 1:

The patent applies preliminary tone mapping and adaptive reshaping during the encoding stage to pre-adjust the luminance distribution and allocate codewords according to the human visual system's sensitivity. This preliminary action ensures that visual details are preserved before compression, allowing low bit depth encoding to maintain quality without requiring complex metadata for post-processing correction.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent dynamically changes encoding parameters including bit depth, color space transformation matrices, and transfer functions based on the content's luminance characteristics. By adapting these parameters to match the HVS sensitivity at different luminance levels, the system achieves efficient compression while preserving visual details, resolving the contradiction between low complexity and information loss.

Inventive Principle:
Principle #35Parameter changes

2Loss of information

If adaptive reshaping with dynamic codeword allocation is implemented, then visual details are preserved across luminance levels, but device complexity and processing requirements increase

Engineering Contradiction:
Improvevisual details preservationVSAvoidprocessing capabilities
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The patent segments the luminance range into multiple regions (dark, mid-tone, bright) and applies different codeword allocation strategies to each segment based on HVS sensitivity. This segmentation allows the system to preserve visual details where most needed while using simpler encoding in less sensitive regions, reducing overall processing complexity while maintaining quality.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies local quality enhancement by allocating higher precision codewords to luminance regions where the human visual system is most sensitive (particularly mid-tones and transitions) and using coarser quantization in less sensitive regions. This localized approach preserves visual details effectively while minimizing the overall processing burden on downstream devices.

Inventive Principle:
Principle #3Local quality

3Ease of operation

If metadata size is reduced for downstream device processing, then ease of operation improves, but the ability to achieve high quality decoding is compromised

Engineering Contradiction:
Improvedownstream device processingVSAvoiddecoding quality
Core Design Contradiction:
Ease of operationVSManufacturing precision

Solution Approach 1:

The patent creates a simplified copy of the essential transfer function characteristics that can be represented with minimal metadata. Instead of transmitting complete lookup tables or complex transformation matrices, the system encodes the key parameters that define the reshaping behavior, allowing downstream devices to reconstruct the necessary transformation with reduced metadata while maintaining decoding quality.

Inventive Principle:
Principle #26Copying

Data Source

PatentEP3272123B1Signal reshaping approximation
Publication Date: 2019.06.19 DOLBY LABORATORIES LICENSING CORP
  • EP3272123B1 patent drawingFigure 1A
  • EP3272123B1 patent drawingFigure 1B
  • EP3272123B1 patent drawingFigure 2

AI summary

Statistical values are computed based on received source images. An adaptive reshaping function is selected for one or more source images based on the one or more statistical values. A portion of source video content is adaptively reshaped, based on the selected adaptive reshaping function to generate a portion of reshaped video content. The portion of source video content is represented by the one or more source images. An approximation of an inverse of the selected adaptive reshaping function is generated. The reshaped video content and a set of adaptive reshaping parameters defining the approximation of the inverse of the selected adaptive reshaping function are encoded into a reshaped video signal. The reshaped video signal may be processed by a downstream recipient device to generate a version of reconstructed source images, for example, for rendering with a display device.