Cross-Component Chroma Prediction With Adaptive Model Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing video coding standards, such as VVC and ECM, require improvements in cross-component prediction (CCP) for chroma prediction to enhance coding efficiency and reduce computational complexity.
Innovation Solution
Implementing advanced cross-component prediction models, including block-vector guided CCM (BVG-CCCM) and gradient-based GL-CCCM, which adaptively fuse chroma prediction with luma samples using convolutional filters and slope adjustments, and derive models based on adjacent sample gradients.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If cross-component prediction is implemented to improve chroma prediction accuracy, then coding efficiency is improved, but computational complexity increases
Solution Approach 1:
The patent uses lightweight, simplified prediction models that can be quickly computed and discarded for each block, replacing complex persistent models. This allows high prediction accuracy through adaptive modeling while keeping computational cost low by using simple, fast-to-compute operations that don't require extensive processing resources.
Solution Approach 2:
The patent divides the chroma prediction process into multiple independent linear models applied to different sub-blocks or regions. Each sub-block has its own simple linear model parameters, allowing localized accurate prediction without requiring a single complex global model, thus improving accuracy while maintaining computational efficiency through parallelization.
2Manufacturing precision
If multiple linear models are applied to different subdivisions to improve prediction accuracy, then chroma prediction accuracy is improved, but device complexity increases
Solution Approach 1:
The patent segments the chroma block into multiple sub-blocks and applies different linear models to each segment. This allows the system to capture local variations in the luma-chroma relationship more accurately while keeping each individual model simple, resolving the contradiction between accuracy and complexity.
Solution Approach 2:
The patent applies the principle of local quality by using different linear model parameters for different spatial regions or sub-blocks within the chroma block. This allows the prediction to adapt to local characteristics of the image content, improving accuracy in regions with varying luma-chroma relationships while maintaining overall computational efficiency.
Data Source
AI summary
Methods and systems implement cross-component prediction (“CCP”) for chroma prediction, to improve prediction accuracy. A VVC-standard encoder and a VVC-standard decoder can configure one or more processors of a computing system to perform chroma fusion inheritance in CCP merge modes; update a CCP model by a current reconstructed block; perform adaptive fusion for interCCCM mode and inter-CCP merge mode; and perform adaptive model derivation for interCCCM mode.