Intra Prediction Mode Extension for Video Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video encoding and decoding methods lack efficiency in predicting slow-changing intensity regions, as existing prediction modes inadequately describe such regions, leading to suboptimal compression performance.
Innovation Solution
Introduction of three additional intra prediction modes - Planar_horizontal, Planar_vertical, and Planar_diagonal - to better model slow-changing intensity regions, expanding the set of prediction modes from 67 to 70 and modifying the Most Probable Mode (MPM) list for improved encoding and decoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If existing prediction modes are used, then the encoding complexity is kept low, but the compression efficiency for slow-changing intensity regions is insufficient
Solution Approach 1:
The patent segments the prediction mode set into traditional modes and three new planar modes (horizontal, vertical, diagonal). This segmentation allows the system to selectively apply appropriate modes based on the characteristics of the video block, improving compression efficiency for slow-changing regions while maintaining manageable complexity through structured organization
Solution Approach 2:
The patent changes the parameter of prediction mode diversity by introducing three additional planar modes to the existing set. This parameter change enables more accurate modeling of slow-changing intensity regions, directly addressing the compression efficiency problem without requiring fundamental changes to the prediction framework
2Measurement precision
If the number of prediction modes is increased from 67 to 70, then the accuracy of predicting slow-changing regions is improved, but the encoding and decoding complexity increases
Solution Approach 1:
The patent applies local quality by introducing specific planar modes (horizontal, vertical, diagonal) that are locally optimized for slow-changing intensity regions. Rather than uniformly increasing complexity across all prediction scenarios, the new modes are selectively applied where they provide the most benefit, maintaining prediction accuracy while controlling overall complexity
Solution Approach 2:
The patent implements partial action by adding only the specific planar modes needed for slow-changing regions rather than comprehensively expanding all prediction capabilities. This selective approach achieves the necessary prediction accuracy improvement without the full complexity burden of a complete mode expansion
Data Source
AI summary
A method and apparatus for performing intra prediction for encoding or decoding adds multiple prediction modes to those of prior intra prediction schemes. A horizontal mode, a vertical mode, and a diagonal mode are added. Reference pixels for each of the added modes can come from multiple pixel positions away from a block to be encoded or decoded. Reference pixels to the left and above a block are reconstructed from those already coded. Reference pixels from right of, and below, a block are estimated or extrapolated and can be based on other pixels around the block. An index can be sent in a bitstream indicating which prediction was used.