Intra Prediction Plane Fitting for Video Coding Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding standards, such as ITU-T H.264/AVC and ITU-T H.265/HEVC, face challenges in achieving high coding efficiency for intra prediction due to the plane-based mode's regression parameters being determined without considering actual pixel values, leading to low compression efficiency.
Innovation Solution
The proposed solution combines directional intra-prediction mechanisms with the plane-based intra-prediction mode, allowing for clipping of both regression parameters and pixel values, and accounts for intensity changes of reference samples along a selected prediction direction, independent of plane parameters, to improve coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If plane-based intra prediction mode is used with regression parameters, then prediction coverage is improved, but coding efficiency deteriorates because parameters are determined without considering actual pixel values
Solution Approach 1:
The patent changes the parameters used in plane-based prediction from fixed regression coefficients to adaptive parameters derived from actual reference pixel values. Specifically, it uses the average of reference pixels for DC mode and gradient calculations based on actual pixel differences, making the prediction adapt to the actual image content while maintaining coding efficiency.
Solution Approach 2:
The patent introduces feedback mechanisms where the prediction process uses actual reference pixel values from the image to adjust prediction parameters. The gradient calculations incorporate actual pixel differences from reference blocks, creating a feedback loop that adapts prediction to real image characteristics rather than using fixed regression models.
2Measurement precision
If more intra prediction modes are added to improve accuracy, then prediction precision is improved, but device complexity increases
Solution Approach 1:
The patent applies partial action by enhancing only the most important prediction modes (DC and plane modes) with adaptive parameter calculations based on actual pixel values, rather than implementing all possible prediction modes. This provides significant precision improvement with moderate complexity increase.
Solution Approach 2:
The patent improves prediction precision by changing the parameters of existing simple modes (DC and plane modes) to be adaptive rather than fixed, achieving high precision with low computational complexity compared to adding many complex directional modes.
Data Source
AI summary
A video coding block is intra-predicted on the basis of reference samples from a set of neighboring video coding blocks. A fitting plane is determined on the basis of the reference samples. The fitting plane defines a plurality of fitting samples. For a selected directional intra prediction mode, the sample values of the video coding block are predicted on the basis of the reference sample values and the fitting sample values.