Texture Synthesis Frequency Adjustment for Video Coding Efficiency

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current video coding technologies, such as HEVC, struggle with efficiently encoding complex textures due to low predictability, leading to inefficient bit rate usage and subjective quality issues.

Innovation Solution

Implement texture synthesis with frequency adjustment by applying a frequency damping function in the spectral domain, combined with motion compensation and luminance reconstruction to enhance encoding efficiency and subjective quality.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If conventional predictive coding is used for encoding complex textures, then bit rate efficiency deteriorates, but implementing texture synthesis with frequency adjustment increases device complexity

Engineering Contradiction:
Improvebit rate efficiencyVSAvoidencoding complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The image is divided into texture regions and non-texture regions. Texture regions are identified and processed separately using texture synthesis with frequency adjustment, while non-texture regions use conventional predictive coding. This segmentation allows the complex processing to be applied only where necessary, improving bit rate efficiency for complex textures without unnecessarily increasing overall device complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies frequency damping parameters to adjust the spectral content of synthesized texture blocks. By modifying frequency domain parameters of the texture synthesis process, the method adapts the synthesized textures to better match the original image characteristics, thereby improving encoding efficiency for complex texture regions while maintaining a manageable level of processing complexity.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If texture synthesis is applied to improve subjective quality, then bit rate overhead increases, but frequency adjustment reduces this overhead

Engineering Contradiction:
Improvesubjective qualityVSAvoidbit rate overhead
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

Frequency adjustment through damping parameters modifies the spectral characteristics of synthesized texture blocks to better match the original image. This parameter adjustment improves the subjective quality of reconstructed textures while reducing the bit rate overhead by enabling more efficient representation of texture information in the frequency domain.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent replaces direct spatial domain texture copying with frequency domain processing. By transforming texture blocks to the frequency domain, applying damping functions, and then transforming back, the method achieves better quality with reduced overhead compared to conventional spatial domain texture synthesis approaches.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Measurement precision

If pixel-wise fidelity is prioritized in encoding, then coding efficiency for complex textures deteriorates, but texture synthesis targeting subjective quality improves efficiency

Engineering Contradiction:
Improvepixel-wise fidelityVSAvoidcoding efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent changes the optimization criterion from pixel-wise fidelity to subjective quality by applying frequency damping parameters during texture synthesis. This parameter change allows the encoding process to prioritize perceptually important features over exact pixel matching, thereby improving coding efficiency for complex textures while maintaining acceptable visual quality.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The method applies different processing strategies to different regions: texture regions use frequency-adjusted texture synthesis optimized for subjective quality, while non-texture regions use conventional predictive coding optimized for pixel-wise fidelity. This local quality approach improves overall coding efficiency by matching the processing method to the local image characteristics.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP3662669B1Frequency adjustment for texture synthesis in video coding
Publication Date: 2026.01.21 HUAWEI TECH CO LTD
  • EP3662669B1 patent drawingFigure 1~2
  • EP3662669B1 patent drawingFigure 3~4
  • EP3662669B1 patent drawingFigure 5

AI summary

The present disclosure relates to encoding a decoding video employing texture coding. In particular, a texture region is identified within a video picture and a texture patch is determined for said region. Moreover, a first set of parameters is derived specifying weighting factors for reconstructing spectral coefficients of the texture region by fitting the texture region in spectral domain to a first function of the texture patch, the first function being defined by the first set of parameters. The texture patch and the first set of parameters are then included into a bitstream which is output of the encoder and provided in this way to the decoder which reconstructs the texture based on the patch and the function applied to the patch.