Occurrence-Based Intra Coding Using Neighbor Histograms
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding systems face inefficiencies in predicting intra prediction modes, leading to suboptimal compression and decoding performance.
Innovation Solution
A device or method that determines intra prediction modes by analyzing neighboring blocks, updating histograms based on sample numbers, and using blending weights to refine intra prediction mode occurrences for improved encoding and decoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional intra prediction mode selection is used, then encoding complexity is reduced, but prediction accuracy and compression efficiency deteriorate
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing histograms of intra prediction mode occurrences from previously decoded neighboring blocks. These pre-computed histograms are then directly used to guide the selection of intra prediction modes for the current block, eliminating the need for complex real-time analysis during encoding while improving prediction accuracy through data-driven mode selection.
Solution Approach 2:
The patent implements self-service by having the decoder perform the histogram analysis and intra prediction mode selection that would traditionally require encoder involvement. The decoder uses the pre-computed histograms to automatically determine the most probable intra prediction modes for current blocks, reducing encoder complexity while maintaining prediction accuracy through autonomous decoder-side decision-making.
2Measurement precision
If histogram updating based on multiple neighboring blocks is performed, then intra prediction mode prediction accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent applies the extraction principle by selectively using only the necessary histogram data from neighboring blocks that are already decoded and available. Instead of analyzing all possible neighboring blocks or performing exhaustive histogram computations, the method extracts and utilizes pre-computed histogram information from relevant neighboring blocks, reducing computational complexity while maintaining prediction accuracy.
Solution Approach 2:
The patent uses preliminary action by pre-computing histograms during the decoding of neighboring blocks before they are needed for the current block's prediction. This advance preparation eliminates the need for complex real-time histogram analysis during the critical path of encoding/decoding the current block, reducing computational complexity while preserving accurate prediction mode selection.
3Manufacturing precision
If blending weights are calculated for multiple intra prediction modes, then prediction quality is enhanced, but encoding overhead increases
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting the number of blending weights calculated based on the specific characteristics of the current block and its neighboring blocks. Instead of always calculating blending weights for all possible intra prediction modes, the method adapts the parameter of how many modes receive blending weight calculations, enhancing prediction quality where needed while minimizing encoding overhead in simpler cases.
Solution Approach 2:
The patent implements partial action by calculating blending weights for only the most probable intra prediction modes identified from the histogram analysis, rather than computing weights for all possible modes. This selective approach enhances prediction quality for the dominant modes while avoiding the excessive computational and encoding overhead that would result from processing all modes equally.
Data Source
Figure 1A
Figure 1B
Figure 1C
AI summary
A device determine that a neighboring block associated with a current block is associated with a plurality of intra prediction modes. The device may obtain a sample number associated with a first intra prediction mode. The device may update a histogram of intra prediction mode occurrence associated with the current block based on the sample number associated with the first intra prediction mode and the first intra prediction mode. The device may encode and/or decode the current block based on the histogram of intra prediction mode occurrence for the current block. The neighboring block may include a plurality of partitions. The device may identify a first partition of the neighboring block associated with the first intra prediction mode and a second partition associated with the second the first intra prediction mode, and update the histogram of intra prediction mode occurrence based on the sample numbers of the two partitions.