Content-Adaptive Preprocessing for HDR Static Transfer Functions

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current video coding technologies face challenges in efficiently compressing High Dynamic Range (HDR) and Wide Color Gamut (WCG) video data, as existing transfer functions are static and do not adapt to content properties, leading to suboptimal bit allocation and increased bitrate for bright and dark areas in HDR content.

Innovation Solution

A content-adaptive High Dynamic Range (HDR) video processing system that employs a static transfer function with adaptive signal processing, including pre-processing and post-processing units to adjust signal characteristics, allowing for more efficient utilization of the codeword space and improved signal-to-quantization noise ratio.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If a static transfer function is used to compact dynamic range, then the device complexity is reduced, but the compression efficiency deteriorates due to suboptimal bit allocation

Engineering Contradiction:
Improvetransfer function complexityVSAvoidcompression efficiency
Core Design Contradiction:
Device complexityVSProductivity

Solution Approach 1:

The patent applies preliminary action by performing content-adaptive preprocessing of video data before applying the static transfer function. The system analyzes video content statistics (such as brightness distribution and dynamic range characteristics) and adjusts the input signal accordingly, so that when the static transfer function is applied, it produces optimal compression results for that specific content type.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements parameter changes by modifying the input parameters to the static transfer function based on content analysis. Instead of changing the transfer function itself, the system adjusts parameters such as signal scaling, offset values, and preprocessing transformations to optimize the performance of the fixed transfer function for different video content characteristics.

Inventive Principle:
Principle #35Parameter changes

2Ease of manufacture

If a static transfer function is used, then the ease of manufacture is improved, but the adaptability to different video content deteriorates

Engineering Contradiction:
Improvesystem implementation easeVSAvoidcontent adaptability
Core Design Contradiction:
Ease of manufactureVSAdaptability or versatility

Solution Approach 1:

The patent introduces an intermediary component - a content analysis and preprocessing module - that sits between the video input and the static transfer function. This intermediary analyzes video content characteristics and transforms the input signal adaptively, allowing a simple static transfer function to achieve content-adaptive performance through intelligent signal preparation.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Device complexity

If no preprocessing is applied, then the device complexity is reduced, but the signal-to-quantization noise ratio deteriorates

Engineering Contradiction:
Improveprocessing pipeline complexityVSAvoidsignal-to-quantization noise ratio
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent applies preliminary action by implementing preprocessing operations before the transfer function that optimize the signal characteristics. By analyzing content statistics and adjusting the input signal in advance, the system ensures that the quantization process occurs on optimally prepared data, maximizing the signal-to-quantization noise ratio without requiring complex post-processing.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4090017A1Content-adaptive application of fixed transferfunction to high dynamic range (HDR) and/or widecolor gamut (WCG) video data
Publication Date: 2022.11.16 QUALCOMM INC
  • EP4090017A1 patent drawingFigure 1
  • EP4090017A1 patent drawingFigure 2
  • EP4090017A1 patent drawingFigure 3

AI summary

This disclosure relates to processing video data, including processing video data to conform to a high dynamic range (HDR)/wide color gamut (WCG) color container. The techniques apply, on an encoding side, pre-processing of color values prior to application of a static transfer function and/or apply post-processing on the output from the application of the static transfer function. By applying pre-processing, the examples may generate color values that when compacted into a different dynamic range by application of the static transfer function linearize the output codewords. By applying post-processing, the examples may increase signal to quantization noise ratio. The examples may apply the inverse of the operations on the encoding side on the decoding side to reconstruct the color values.