Video Coding Cross-Component Linear Model Prediction Padding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding technologies face challenges in efficiently reducing redundancy in video data, particularly in the context of intra prediction and motion compensation, which can lead to increased bandwidth and storage requirements.
Innovation Solution
The proposed solution involves improvements to the cross-component linear model (CCLM) prediction mode, specifically by padding unavailable neighboring luma samples with available samples, which simplifies the down-sample filtering process and reduces dependencies on the location of coding units within a CTU.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If traditional intra coding without intra prediction is used, then the coding process is simpler, but the compression ratio is lower and bandwidth requirements are higher
Solution Approach 1:
The patent applies preliminary action by performing intra prediction using neighboring sample values before the actual coding process. Predictor blocks are generated in advance by copying and filtering neighboring samples, allowing the main coding process to work with pre-processed data that requires less bandwidth for transmission.
2Measurement precision
If down-sample filtering is performed on unavailable neighboring luma samples using complex boundary detection, then prediction accuracy is improved, but the computational complexity increases
Solution Approach 1:
The patent applies local quality by differentiating the filtering approach based on the location of samples within the CTU structure. Samples at the top and left boundaries use one filtering approach, while other samples use a different approach. This localized differentiation maintains prediction accuracy where it matters most while reducing overall computational complexity.
Solution Approach 2:
The patent performs preliminary action by pre-determining which neighboring luma samples are available and preparing filtered values in advance. By identifying available samples and computing their filtered values before the main prediction process, the patent avoids complex runtime boundary detection and reduces computational overhead during actual coding.
3Measurement precision
If cross-component linear model prediction is implemented, then chroma prediction accuracy is improved, but the dependency on luma sample availability increases complexity
Solution Approach 1:
The patent uses an intermediary approach by introducing a linear model that relates luma and chroma components. Instead of directly requiring every luma sample to be available for chroma prediction, the linear model acts as an intermediary that can estimate chroma values based on available luma samples, reducing the strict dependency on complete luma sample availability.
Solution Approach 2:
The patent applies parameter changes by transforming the prediction problem from direct sample copying to a parameter-based linear model. By changing from a spatial-domain approach (copying samples) to a parameter-domain approach (linear coefficients), the patent can handle missing samples more gracefully while maintaining prediction accuracy.
Data Source
Figure 1A
Figure 1B
Figure 1C
AI summary
Aspects of the disclosure provide methods, apparatuses, and non-transitory computer-readable storage mediums for video encoding/decoding. An apparatus includes processing circuitry that decodes prediction information for a current block in a coded video sequence. The prediction information indicates a cross-component linear model (CCLM) prediction mode for the current block. The processing circuitry determines a sample value of a first unavailable neighboring luma sample of the current block based on at least one luma sample used in the CCLM prediction mode not being available. The sample value of the first unavailable neighboring luma sample is determined based on a sample value of an available neighboring luma sample. The processing circuitry calculates a parameter of the CCLM prediction mode based on the sample value of the first unavailable neighboring luma sample of the current block. Further, the processing circuitry reconstructs the current block based on the calculated parameter of the CCLM prediction mode.