VVC Intra Prediction Block Merging to Reduce Coding Load
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The block division pattern in VVC requires a significant amount of processing due to the need to evaluate all division patterns for determining the size of coding units, leading to increased computational load, especially when performing intra-screen prediction and quantization processing.
Innovation Solution
A video coding method that divides images into blocks, acquires intra-image prediction modes, and determines whether to combine adjacent blocks based on these modes to reduce processing by optimizing block division patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If all division patterns are evaluated to determine the size of coding units in VVC, then the block division can match the pattern and feature of the coding target frame more accurately, but the processing amount increases significantly
Solution Approach 1:
The patent segments the evaluation process by dividing it into multiple stages: initial evaluation of all division patterns, intermediate evaluation of promising patterns only, and final evaluation of the best candidate. This segmentation allows the system to maintain high block division accuracy while reducing overall processing time by avoiding complete evaluation of all patterns.
Solution Approach 2:
The patent applies partial action by performing complete evaluation only on a subset of division patterns that show promise in intermediate evaluation, while using simplified evaluation for the remaining patterns. This approach maintains sufficient block division accuracy for the most important cases while reducing total processing load.
2Productivity
If pruning is performed adaptively using adjacent block division results, then the processing amount can be reduced, but the processing amount varies according to video content
Solution Approach 1:
The patent performs preliminary adaptive pruning using adjacent block division results before the main evaluation process. This preliminary action eliminates obviously suboptimal division patterns early, ensuring consistent processing efficiency across different video contents while maintaining the ability to handle content-specific optimizations.
Solution Approach 2:
The patent uses feedback from adjacent block division results to guide the pruning process. By incorporating information from already-encoded adjacent blocks, the system can make informed decisions about which division patterns to evaluate fully and which to prune, achieving both processing efficiency and consistency across different video contents.
3Adaptability or versatility
If QT division, BT division, and TT division are performed for each CU size, then comprehensive block division options are available, but the processing amount is greatly increased by intra-screen prediction and conversion processing
Solution Approach 1:
The patent implements dynamic evaluation where the depth and type of intra-screen prediction and conversion processing depend on the results of intermediate evaluation. Division patterns that show promise undergo complete processing, while others undergo simplified processing or are pruned entirely. This dynamic approach maintains division pattern flexibility while adapting processing complexity to actual needs.
Solution Approach 2:
The patent performs preliminary evaluation of division patterns using simplified metrics before committing to expensive intra-screen prediction and conversion processing. This preliminary action filters out poor candidates early, ensuring that comprehensive processing is applied only to promising patterns, thus maintaining versatility while reducing overall processing complexity.
Data Source
AI summary
A video coding method is a video coding method for coding an image in units of blocks obtained by dividing an image, and in the method a computer performs processing for dividing the image into blocks, acquiring an intra-image prediction mode predicted for each block, and generating, by using a combination pattern based on the intra-image prediction mode of a plurality of blocks adjacent to at least any one block, determining whether or not the plurality of blocks are to be combined, and performing prediction for each combined block when it is determined that the plurality of blocks are to be combined.