Prediction-Sample CCRM for Chroma Residual Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in efficiently compressing video data due to the lack of a comprehensive cross-component residual model, particularly in the context of video data compression, especially in handling chroma components, which are not adequately addressed by current technologies, leading to inefficiencies in the context of chroma components, which are not adequately addressed by current technologies, which are not adequately addressed by current technologies, which are not adequately addressed by current technologies, which are not adequately addressed by current technologies, which are not adequately addressed by current technologies.
Innovation Solution
The proposed solution is the implementation of a cross-component residual model (CCRM) that applies a prediction sample domain model (P-CCRM) to derive chroma residual data, which is not adequately addressed by current technologies, which are not adequately addressed by current technologies, which are not adequately addressed by current technologies, which are not adequately addressed by current technologies, which are not adequately addressed by current technologies, which are not adequately addressed by current technologies, which are not adequately addressed by current technologies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a cross-component residual model (CCRM) is applied to the current block, then chroma component compression efficiency is improved, but device complexity increases
Solution Approach 1:
The patent introduces filter coefficients as an intermediary element that maps luma residual data to chroma residual data. These coefficients act as a mediator between the luma and chroma components, enabling the cross-component residual model to function without requiring complex direct modeling. The filter coefficients simplify the relationship between components while maintaining compression efficiency.
Solution Approach 2:
The patent changes the parameter representation by using filter coefficients (alpha and beta) to describe the cross-component relationship. Instead of storing full chroma residual data, the model transforms the representation into parameters that capture the essential relationship between luma and chroma, reducing complexity while preserving information.
2Measurement precision
If filter coefficients are derived based on prediction samples, then chroma residual accuracy is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary action by deriving filter coefficients during the prediction phase using available prediction samples before final reconstruction. This allows the model to prepare the mapping relationship in advance, improving accuracy without adding significant overhead during the critical reconstruction phase. The coefficients are computed once and then applied to multiple residual values.
Solution Approach 2:
The patent applies partial action by selectively deriving filter coefficients only for blocks where cross-component modeling is beneficial. The model uses syntax elements to indicate whether P-CCRM is applied to each block, allowing the processor to skip coefficient derivation for blocks where it would not provide significant accuracy improvement, thus reducing overall processing time.
3Reliability
If cross-component residual modeling is implemented, then video coding performance is enhanced, but bitstream complexity increases
Solution Approach 1:
The patent implements universality by designing a syntax structure that can handle both CCRM and P-CCRM modes through a unified framework. The syntax elements are designed to be versatile, indicating the application of different cross-component models without requiring separate dedicated syntax for each mode. This multi-functional syntax reduces bitstream complexity by avoiding redundancy.
Solution Approach 2:
The patent segments the cross-component modeling into distinct syntactic components: a flag indicating P-CCRM application, another flag indicating CCRM application, and conditional filter coefficient data. This segmentation allows the bitstream to efficiently represent only the necessary information for each block type, reducing overall complexity while maintaining performance.
Data Source
AI summary
An apparatus of video decoding is provided. The apparatus includes processing circuitry. The processing circuitry is configured to receive a bitstream that includes syntax information for a current block. The syntax information indicates whether a P-CCRM is applied to the current block. The current block includes a luma component and a chroma component. When the syntax information indicates that the P-CCRM is applied to the current block, the processing circuitry is configured to derive chroma residual data of the chroma component based on luma residual data of the luma component. The processing circuitry is configured to reconstruct samples of the chroma component based on prediction samples of the chroma component and the derived chroma residual data.