Chroma Block Prediction Using Neighbor Selection and Nonlinear Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video encoding/decoding methods lack efficient cross-component prediction techniques for chroma blocks, particularly in handling parameters like alpha and beta for linear and non-linear models, and neighbor selection methods for improved prediction accuracy.
Innovation Solution
The proposed methods involve deriving beta using various models, including non-linear models, and selecting neighbors for cross-component prediction, combining these with non-cross-component prediction models to enhance video encoding/decoding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional video encoding methods are used for chroma blocks, then the encoding process is simple, but prediction accuracy and compression efficiency are insufficient
Solution Approach 1:
The patent applies parameter changes by deriving alpha and beta parameters dynamically from neighbor samples rather than using fixed conventional values. The alpha parameter controls the weight of luma samples in cross-component prediction, while beta provides an offset adjustment. These parameters are calculated based on actual video content characteristics, enabling adaptive optimization of prediction accuracy for different video sequences and blocks.
Solution Approach 2:
The patent segments the chroma block into multiple prediction regions and applies different cross-component prediction models to different segments. By dividing the block and applying localized prediction with region-specific alpha and beta parameters, the method achieves higher overall prediction accuracy while managing complexity through selective application of advanced techniques only where beneficial.
2Productivity
If advanced cross-component prediction models are applied, then compression efficiency improves, but computational complexity increases
Solution Approach 1:
The patent implements partial action by applying cross-component prediction with derived alpha and beta parameters selectively rather than universally. The method evaluates whether the computational cost of deriving parameters and applying cross-component prediction is justified by the actual compression benefit for each chroma block, applying the advanced model only when it provides measurable improvement over simpler methods.
Solution Approach 2:
The patent performs preliminary action by deriving alpha and beta parameters from neighbor samples before applying cross-component prediction to the current chroma block. This pre-computation of parameters based on already-decoded neighboring blocks allows the main prediction process to proceed efficiently without iterative optimization, reducing overall computational complexity.
3Measurement precision
If more neighbor samples are used for parameter derivation, then prediction accuracy improves, but processing time increases
Solution Approach 1:
The patent extracts only the essential and most relevant neighbor samples needed for accurate parameter derivation, rather than processing all available neighboring blocks. By selectively choosing a limited set of representative neighbor samples that capture the local video characteristics, the method achieves good parameter accuracy while minimizing the number of samples that need to be processed and stored.
Data Source
AI summary
Example implementations include a method, apparatus and computer-readable medium of video coding, comprising receiving the code block and one or more neighbor samples and determining the value of beta based on at least one of an average chroma value, a midrange chroma value, a median chroma value, an average luma value, a midrange luma value, or a median luma value of two or more neighbor samples.